feat: support litellm-sdk for reranker endpoint

.claude-plugin/marketplace.json

@@ -1,15 +0,0 @@
-{
-  "$schema": "https://anthropic.com/claude-code/marketplace.schema.json",
-  "name": "hindsight",
-  "description": "Official Hindsight integrations for Claude Code",
-  "owner": {
-    "name": "vectorize-io"
-  },
-  "plugins": [
-    {
-      "name": "hindsight-memory",
-      "description": "Automatic long-term memory for Claude Code via Hindsight",
-      "source": "./hindsight-integrations/claude-code"
-    }
-  ]
-}

.env.example

@@ -2,10 +2,10 @@
 # Copy this file to .env and fill in your values
 
 # LLM Configuration (Required)
-# Supported providers: openai, groq, ollama, gemini, anthropic, lmstudio, vertexai, minimax
+# Supported providers: openai, groq, ollama, gemini, anthropic, lmstudio, vertexai
 HINDSIGHT_API_LLM_PROVIDER=openai
 HINDSIGHT_API_LLM_API_KEY=your-api-key-here
-HINDSIGHT_API_LLM_MODEL=gpt-4o-mini
+HINDSIGHT_API_LLM_MODEL=o3-mini
 HINDSIGHT_API_LLM_BASE_URL=https://api.openai.com/v1
 
 # Example: Anthropic Claude configuration
@@ -20,11 +20,6 @@ HINDSIGHT_API_LLM_BASE_URL=https://api.openai.com/v1
 # HINDSIGHT_API_LLM_VERTEXAI_REGION=us-central1
 # HINDSIGHT_API_LLM_VERTEXAI_SERVICE_ACCOUNT_KEY=/path/to/service-account-key.json  # Optional, uses ADC if not set
 
-# Example: MiniMax configuration (1M context window)
-# HINDSIGHT_API_LLM_PROVIDER=minimax
-# HINDSIGHT_API_LLM_API_KEY=your-minimax-api-key
-# HINDSIGHT_API_LLM_MODEL=MiniMax-M2.7
-
 # Example: LM Studio local configuration (Qwen 2.5 32B recommended)
 # HINDSIGHT_API_LLM_PROVIDER=lmstudio
 # HINDSIGHT_API_LLM_API_KEY=lmstudio
@@ -46,12 +41,6 @@ HINDSIGHT_API_LOG_LEVEL=info
 # HINDSIGHT_API_DATABASE_URL=postgresql://user:pass@host:5432/db
 # HINDSIGHT_API_DATABASE_SCHEMA=public  # PostgreSQL schema name (default: public)
 
-# Vector Extension (Optional - uses pgvector by default)
-# Options: "pgvector" (default), "vchord", "pgvectorscale" (DiskANN)
-# HINDSIGHT_API_VECTOR_EXTENSION=pgvector
-# For Azure PostgreSQL with DiskANN:
-# HINDSIGHT_API_VECTOR_EXTENSION=pgvectorscale  # Auto-detects pg_diskann on Azure
-
 # Embeddings Configuration (Optional - uses local by default)
 # Provider: "local" (default) or "tei" (HuggingFace Text Embeddings Inference)
 # HINDSIGHT_API_EMBEDDINGS_PROVIDER=local

.github/dependabot.yml

@@ -1,6 +0,0 @@
-version: 2
-updates:
-  - package-ecosystem: "github-actions"
-    directory: "/"
-    schedule:
-      interval: "weekly"

.github/workflows/deploy-docs.yml

@@ -21,20 +21,17 @@ jobs:
   build:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v6
-      - uses: actions/setup-node@v6
+      - uses: actions/checkout@v4
+      - uses: actions/setup-node@v4
         with:
           node-version: 20
           cache: npm
           cache-dependency-path: package-lock.json
-      - uses: astral-sh/setup-uv@v7
+      - uses: astral-sh/setup-uv@v4
       - run: npm ci --workspace=hindsight-docs
       - run: uv run generate-llms-full
       - run: npm run build --workspace=hindsight-docs
-        env:
-          UMAMI_URL: https://analytics.hindsight.vectorize.io
-          UMAMI_WEBSITE_ID: ${{ secrets.UMAMI_WEBSITE_ID }}
-      - uses: actions/upload-pages-artifact@v4
+      - uses: actions/upload-pages-artifact@v3
         with:
           path: hindsight-docs/build
   deploy:

.github/workflows/release-integration.yml

@@ -1,111 +0,0 @@
-name: Release Integration
-
-on:
-  push:
-    tags:
-      - 'integrations/**'
-
-jobs:
-  publish:
-    runs-on: ubuntu-latest
-    permissions:
-      id-token: write  # for PyPI trusted publishing
-
-    steps:
-    - uses: actions/checkout@v6
-
-    - name: Extract integration info
-      id: info
-      run: |
-        # refs/tags/integrations/litellm/v0.1.0 → integration=litellm, version=0.1.0
-        TAG="${GITHUB_REF#refs/tags/}"
-        INTEGRATION=$(echo "$TAG" | cut -d'/' -f2)
-        VERSION=$(echo "$TAG" | cut -d'/' -f3 | sed 's/^v//')
-        echo "integration=$INTEGRATION" >> $GITHUB_OUTPUT
-        echo "version=$VERSION" >> $GITHUB_OUTPUT
-        echo "tag=$TAG" >> $GITHUB_OUTPUT
-        echo "Integration: $INTEGRATION, Version: $VERSION"
-
-    - name: Detect integration type
-      id: type
-      run: |
-        if [ -f "hindsight-integrations/${{ steps.info.outputs.integration }}/pyproject.toml" ]; then
-          echo "type=python" >> $GITHUB_OUTPUT
-        elif [ -f "hindsight-integrations/${{ steps.info.outputs.integration }}/package.json" ]; then
-          echo "type=typescript" >> $GITHUB_OUTPUT
-        else
-          echo "type=plugin" >> $GITHUB_OUTPUT
-        fi
-
-    # ── Python integrations (litellm, pydantic-ai, crewai) ──────────────────
-
-    - name: Install uv
-      if: steps.type.outputs.type == 'python'
-      uses: astral-sh/setup-uv@v7
-      with:
-        enable-cache: true
-
-    - name: Set up Python
-      if: steps.type.outputs.type == 'python'
-      uses: actions/setup-python@v6
-      with:
-        python-version-file: ".python-version"
-
-    - name: Build Python package
-      if: steps.type.outputs.type == 'python'
-      working-directory: ./hindsight-integrations/${{ steps.info.outputs.integration }}
-      run: uv build --out-dir dist
-
-    - name: Publish Python package to PyPI
-      if: steps.type.outputs.type == 'python'
-      uses: pypa/gh-action-pypi-publish@release/v1
-      with:
-        packages-dir: ./hindsight-integrations/${{ steps.info.outputs.integration }}/dist
-        skip-existing: true
-
-    # ── TypeScript integrations (ai-sdk, chat, openclaw) ────────────────────
-
-    # ── Plugin integrations (claude-code) — no package to publish ───────────
-
-    - name: Plugin release
-      if: steps.type.outputs.type == 'plugin'
-      run: |
-        echo "Plugin integration ${{ steps.info.outputs.integration }} v${{ steps.info.outputs.version }} — no package to publish."
-        echo "Users install via: claude plugin marketplace add vectorize-io/hindsight --sparse hindsight-integrations"
-
-    # ── TypeScript integrations (ai-sdk, chat, openclaw) ────────────────────
-
-    - name: Set up Node.js
-      if: steps.type.outputs.type == 'typescript'
-      uses: actions/setup-node@v6
-      with:
-        node-version: '22'
-        registry-url: 'https://registry.npmjs.org'
-
-    - name: Install dependencies
-      if: steps.type.outputs.type == 'typescript'
-      working-directory: ./hindsight-integrations/${{ steps.info.outputs.integration }}
-      run: npm ci
-
-    - name: Build TypeScript package
-      if: steps.type.outputs.type == 'typescript'
-      working-directory: ./hindsight-integrations/${{ steps.info.outputs.integration }}
-      run: npm run build
-
-    - name: Publish TypeScript package to npm
-      if: steps.type.outputs.type == 'typescript'
-      working-directory: ./hindsight-integrations/${{ steps.info.outputs.integration }}
-      run: |
-        set +e
-        OUTPUT=$(npm publish --access public 2>&1)
-        EXIT_CODE=$?
-        echo "$OUTPUT"
-        if [ $EXIT_CODE -ne 0 ]; then
-          if echo "$OUTPUT" | grep -q "cannot publish over"; then
-            echo "Package version already published, skipping..."
-            exit 0
-          fi
-          exit $EXIT_CODE
-        fi
-      env:
-        NODE_AUTH_TOKEN: ${{ secrets.NPM_TOKEN }}

.github/workflows/release.yml

@@ -13,15 +13,15 @@ jobs:
       id-token: write
 
     steps:
-    - uses: actions/checkout@v6
+    - uses: actions/checkout@v4
 
     - name: Install uv
-      uses: astral-sh/setup-uv@v7
+      uses: astral-sh/setup-uv@v5
       with:
         enable-cache: true
 
     - name: Set up Python
-      uses: actions/setup-python@v6
+      uses: actions/setup-python@v5
       with:
         python-version-file: ".python-version"
 
@@ -30,39 +30,29 @@ jobs:
       working-directory: ./hindsight-clients/python
       run: uv build --out-dir dist
 
-    - name: Build hindsight-api-slim
-      working-directory: ./hindsight-api-slim
-      run: uv build --out-dir dist
-
     - name: Build hindsight-api
       working-directory: ./hindsight-api
       run: uv build --out-dir dist
 
     - name: Build hindsight-all
-      working-directory: ./hindsight-all
+      working-directory: ./hindsight
       run: uv build --out-dir dist
 
-    - name: Build hindsight-all-slim
-      working-directory: ./hindsight-all-slim
+    - name: Build hindsight-litellm
+      working-directory: ./hindsight-integrations/litellm
       run: uv build --out-dir dist
 
     - name: Build hindsight-embed
       working-directory: ./hindsight-embed
       run: uv build --out-dir dist
 
-    # Publish in order (client and api-slim first, then api/all wrappers which depend on them)
+    # Publish in order (client and api first, then hindsight-all which depends on them)
     - name: Publish hindsight-client to PyPI
       uses: pypa/gh-action-pypi-publish@release/v1
       with:
         packages-dir: ./hindsight-clients/python/dist
         skip-existing: true
 
-    - name: Publish hindsight-api-slim to PyPI
-      uses: pypa/gh-action-pypi-publish@release/v1
-      with:
-        packages-dir: ./hindsight-api-slim/dist
-        skip-existing: true
-
     - name: Publish hindsight-api to PyPI
       uses: pypa/gh-action-pypi-publish@release/v1
       with:
@@ -72,13 +62,13 @@ jobs:
     - name: Publish hindsight-all to PyPI
       uses: pypa/gh-action-pypi-publish@release/v1
       with:
-        packages-dir: ./hindsight-all/dist
+        packages-dir: ./hindsight/dist
         skip-existing: true
 
-    - name: Publish hindsight-all-slim to PyPI
+    - name: Publish hindsight-litellm to PyPI
       uses: pypa/gh-action-pypi-publish@release/v1
       with:
-        packages-dir: ./hindsight-all-slim/dist
+        packages-dir: ./hindsight-integrations/litellm/dist
         skip-existing: true
 
     - name: Publish hindsight-embed to PyPI
@@ -89,15 +79,14 @@ jobs:
 
     # Upload artifacts for GitHub release
     - name: Upload artifacts
-      uses: actions/upload-artifact@v7
+      uses: actions/upload-artifact@v4
       with:
         name: python-packages
         path: |
           hindsight-clients/python/dist/*
-          hindsight-api-slim/dist/*
           hindsight-api/dist/*
-          hindsight-all/dist/*
-          hindsight-all-slim/dist/*
+          hindsight/dist/*
+          hindsight-integrations/litellm/dist/*
           hindsight-embed/dist/*
         retention-days: 1
 
@@ -106,10 +95,10 @@ jobs:
     environment: npm
 
     steps:
-    - uses: actions/checkout@v6
+    - uses: actions/checkout@v4
 
     - name: Set up Node.js
-      uses: actions/setup-node@v6
+      uses: actions/setup-node@v4
       with:
         node-version: '20'
         registry-url: 'https://registry.npmjs.org'
@@ -144,21 +133,119 @@ jobs:
       run: npm pack
 
     - name: Upload artifacts
-      uses: actions/upload-artifact@v7
+      uses: actions/upload-artifact@v4
       with:
         name: typescript-client
         path: hindsight-clients/typescript/*.tgz
         retention-days: 1
 
+  release-openclaw-integration:
+    runs-on: ubuntu-latest
+    environment: npm
+
+    steps:
+    - uses: actions/checkout@v4
+
+    - name: Set up Node.js
+      uses: actions/setup-node@v4
+      with:
+        node-version: '22'
+        registry-url: 'https://registry.npmjs.org'
+
+    - name: Install dependencies
+      working-directory: ./hindsight-integrations/openclaw
+      run: npm ci
+
+    - name: Build
+      working-directory: ./hindsight-integrations/openclaw
+      run: npm run build
+
+    - name: Publish to npm
+      working-directory: ./hindsight-integrations/openclaw
+      run: |
+        set +e
+        OUTPUT=$(npm publish --access public 2>&1)
+        EXIT_CODE=$?
+        echo "$OUTPUT"
+        if [ $EXIT_CODE -ne 0 ]; then
+          if echo "$OUTPUT" | grep -q "cannot publish over"; then
+            echo "Package version already published, skipping..."
+            exit 0
+          fi
+          exit $EXIT_CODE
+        fi
+      env:
+        NODE_AUTH_TOKEN: ${{ secrets.NPM_TOKEN }}
+
+    - name: Pack for GitHub release
+      working-directory: ./hindsight-integrations/openclaw
+      run: npm pack
+
+    - name: Upload artifacts
+      uses: actions/upload-artifact@v4
+      with:
+        name: openclaw-integration
+        path: hindsight-integrations/openclaw/*.tgz
+        retention-days: 1
+
+  release-ai-sdk-integration:
+    runs-on: ubuntu-latest
+    environment: npm
+
+    steps:
+    - uses: actions/checkout@v4
+
+    - name: Set up Node.js
+      uses: actions/setup-node@v4
+      with:
+        node-version: '22'
+        registry-url: 'https://registry.npmjs.org'
+
+    - name: Install dependencies
+      working-directory: ./hindsight-integrations/ai-sdk
+      run: npm ci
+
+    - name: Build
+      working-directory: ./hindsight-integrations/ai-sdk
+      run: npm run build
+
+    - name: Publish to npm
+      working-directory: ./hindsight-integrations/ai-sdk
+      run: |
+        set +e
+        OUTPUT=$(npm publish --access public 2>&1)
+        EXIT_CODE=$?
+        echo "$OUTPUT"
+        if [ $EXIT_CODE -ne 0 ]; then
+          if echo "$OUTPUT" | grep -q "cannot publish over"; then
+            echo "Package version already published, skipping..."
+            exit 0
+          fi
+          exit $EXIT_CODE
+        fi
+      env:
+        NODE_AUTH_TOKEN: ${{ secrets.NPM_TOKEN }}
+
+    - name: Pack for GitHub release
+      working-directory: ./hindsight-integrations/ai-sdk
+      run: npm pack
+
+    - name: Upload artifacts
+      uses: actions/upload-artifact@v4
+      with:
+        name: ai-sdk-integration
+        path: hindsight-integrations/ai-sdk/*.tgz
+        retention-days: 1
+
   release-control-plane:
     runs-on: ubuntu-latest
     environment: npm
 
     steps:
-    - uses: actions/checkout@v6
+    - uses: actions/checkout@v4
 
     - name: Set up Node.js
-      uses: actions/setup-node@v6
+      uses: actions/setup-node@v4
       with:
         node-version: '20'
         registry-url: 'https://registry.npmjs.org'
@@ -181,14 +268,11 @@ jobs:
     - name: Build
       run: npm run build --workspace=hindsight-control-plane
 
-    - name: Verify standalone build
-      run: test -f hindsight-control-plane/standalone/server.js || (echo 'standalone/server.js missing - build failed' && exit 1)
-
     - name: Publish to npm
       working-directory: ./hindsight-control-plane
       run: |
         set +e
-        OUTPUT=$(npm publish --access public --ignore-scripts 2>&1)
+        OUTPUT=$(npm publish --access public 2>&1)
         EXIT_CODE=$?
         echo "$OUTPUT"
         if [ $EXIT_CODE -ne 0 ]; then
@@ -206,7 +290,7 @@ jobs:
       run: npm pack
 
     - name: Upload artifacts
-      uses: actions/upload-artifact@v7
+      uses: actions/upload-artifact@v4
       with:
         name: control-plane
         path: hindsight-control-plane/*.tgz
@@ -229,13 +313,9 @@ jobs:
             target: aarch64-apple-darwin
             artifact_name: hindsight
             asset_name: hindsight-darwin-arm64
-          - os: ubuntu-24.04-arm
-            target: aarch64-unknown-linux-gnu
-            artifact_name: hindsight
-            asset_name: hindsight-linux-arm64
 
     steps:
-    - uses: actions/checkout@v6
+    - uses: actions/checkout@v4
 
     - name: Install Rust
       uses: dtolnay/rust-toolchain@stable
@@ -253,7 +333,7 @@ jobs:
         chmod +x artifacts/${{ matrix.asset_name }}
 
     - name: Upload artifacts
-      uses: actions/upload-artifact@v7
+      uses: actions/upload-artifact@v4
       with:
         name: rust-cli-${{ matrix.asset_name }}
         path: artifacts/${{ matrix.asset_name }}
@@ -294,7 +374,7 @@ jobs:
               PRELOAD_ML_MODELS=false
 
     steps:
-    - uses: actions/checkout@v6
+    - uses: actions/checkout@v4
 
     - name: Free Disk Space
       uses: jlumbroso/free-disk-space@main
@@ -308,13 +388,13 @@ jobs:
         swap-storage: true
 
     - name: Set up QEMU
-      uses: docker/setup-qemu-action@v4
+      uses: docker/setup-qemu-action@v3
 
     - name: Set up Docker Buildx
-      uses: docker/setup-buildx-action@v4
+      uses: docker/setup-buildx-action@v3
 
     - name: Log in to GitHub Container Registry
-      uses: docker/login-action@v4
+      uses: docker/login-action@v3
       with:
         registry: ghcr.io
         username: ${{ github.actor }}
@@ -326,7 +406,7 @@ jobs:
 
     - name: Extract metadata for release tags
       id: meta
-      uses: docker/metadata-action@v6
+      uses: docker/metadata-action@v5
       with:
         images: ghcr.io/${{ github.repository_owner }}/${{ matrix.image_name }}
         flavor: |
@@ -342,7 +422,7 @@ jobs:
     # # Step 1: Build for local testing (single platform, no push)
     # # This creates an identical image to what will be released, just for one platform
     # - name: Build image for testing
-    #   uses: docker/build-push-action@v7
+    #   uses: docker/build-push-action@v6
     #   with:
     #     context: .
     #     file: docker/standalone/Dockerfile
@@ -361,7 +441,7 @@ jobs:
 
     # Build multi-platform and push to release tags
     - name: Build and push release images
-      uses: docker/build-push-action@v7
+      uses: docker/build-push-action@v6
       with:
         context: .
         file: docker/standalone/Dockerfile
@@ -379,7 +459,7 @@ jobs:
       packages: write
 
     steps:
-    - uses: actions/checkout@v6
+    - uses: actions/checkout@v4
 
     - name: Install Helm
       uses: azure/setup-helm@v4
@@ -399,7 +479,7 @@ jobs:
       run: helm push helm-packages/*.tgz oci://ghcr.io/${{ github.repository_owner }}/charts
 
     - name: Upload artifacts
-      uses: actions/upload-artifact@v7
+      uses: actions/upload-artifact@v4
       with:
         name: helm-chart
         path: helm-packages/*.tgz
@@ -407,55 +487,67 @@ jobs:
 
   create-github-release:
     runs-on: ubuntu-latest
-    needs: [release-python-packages, release-typescript-client, release-control-plane, release-rust-cli, release-docker-images, release-helm-chart]
+    needs: [release-python-packages, release-typescript-client, release-openclaw-integration, release-ai-sdk-integration, release-control-plane, release-rust-cli, release-docker-images, release-helm-chart]
     permissions:
       contents: write
 
     steps:
-    - uses: actions/checkout@v6
+    - uses: actions/checkout@v4
 
     - name: Extract version from tag
       id: get_version
       run: echo "VERSION=${GITHUB_REF#refs/tags/v}" >> $GITHUB_OUTPUT
 
     - name: Download Python packages
-      uses: actions/download-artifact@v8
+      uses: actions/download-artifact@v4
       with:
         name: python-packages
         path: ./artifacts/python-packages
 
     - name: Download TypeScript client
-      uses: actions/download-artifact@v8
+      uses: actions/download-artifact@v4
       with:
         name: typescript-client
         path: ./artifacts/typescript-client
 
+    - name: Download OpenClaw Integration
+      uses: actions/download-artifact@v4
+      with:
+        name: openclaw-integration
+        path: ./artifacts/openclaw-integration
+
+    - name: Download AI SDK Integration
+      uses: actions/download-artifact@v4
+      with:
+        name: ai-sdk-integration
+        path: ./artifacts/ai-sdk-integration
+
     - name: Download Control Plane
-      uses: actions/download-artifact@v8
+      uses: actions/download-artifact@v4
       with:
         name: control-plane
         path: ./artifacts/control-plane
 
     - name: Download Rust CLI (Linux)
-      uses: actions/download-artifact@v8
+      uses: actions/download-artifact@v4
       with:
         name: rust-cli-hindsight-linux-amd64
         path: ./artifacts/rust-cli-linux
 
     - name: Download Rust CLI (macOS Intel)
-      uses: actions/download-artifact@v8
+      uses: actions/download-artifact@v4
       with:
         name: rust-cli-hindsight-darwin-amd64
         path: ./artifacts/rust-cli-darwin-amd64
 
     - name: Download Rust CLI (macOS ARM)
-      uses: actions/download-artifact@v8
+      uses: actions/download-artifact@v4
       with:
         name: rust-cli-hindsight-darwin-arm64
         path: ./artifacts/rust-cli-darwin-arm64
 
     - name: Download Helm chart
-      uses: actions/download-artifact@v8
+      uses: actions/download-artifact@v4
       with:
         name: helm-chart
         path: ./artifacts/helm-chart
@@ -465,13 +557,16 @@ jobs:
         mkdir -p release-assets
         # Python packages
         cp artifacts/python-packages/hindsight-clients/python/dist/* release-assets/ || true
-        cp artifacts/python-packages/hindsight-api-slim/dist/* release-assets/ || true
         cp artifacts/python-packages/hindsight-api/dist/* release-assets/ || true
-        cp artifacts/python-packages/hindsight-all/dist/* release-assets/ || true
-        cp artifacts/python-packages/hindsight-all-slim/dist/* release-assets/ || true
+        cp artifacts/python-packages/hindsight/dist/* release-assets/ || true
+        cp artifacts/python-packages/hindsight-integrations/litellm/dist/* release-assets/ || true
         cp artifacts/python-packages/hindsight-embed/dist/* release-assets/ || true
         # TypeScript client
         cp artifacts/typescript-client/*.tgz release-assets/ || true
+        # OpenClaw Integration
+        cp artifacts/openclaw-integration/*.tgz release-assets/ || true
+        # AI SDK Integration
+        cp artifacts/ai-sdk-integration/*.tgz release-assets/ || true
         # Control Plane
         cp artifacts/control-plane/*.tgz release-assets/ || true
         # Rust CLI binaries

.gitignore

@@ -46,7 +46,6 @@ hindsight-docs/static/llms-full.txt
 hindsight-dev/benchmarks/locomo/results/
 hindsight-dev/benchmarks/longmemeval/results/
 hindsight-dev/benchmarks/consolidation/results/
-hindsight-dev/benchmarks/perf/results/
 benchmarks/results/
 hindsight-cli/target
 hindsight-clients/rust/target

CLAUDE.md

@@ -17,20 +17,20 @@ Hindsight is an agent memory system that provides long-term memory for AI agents
 ./scripts/dev/start-api.sh
 
 # Run all tests (parallelized with pytest-xdist)
-cd hindsight-api-slim && uv run pytest tests/
+cd hindsight-api && uv run pytest tests/
 
 # Run specific test file
-cd hindsight-api-slim && uv run pytest tests/test_http_api_integration.py -v
+cd hindsight-api && uv run pytest tests/test_http_api_integration.py -v
 
 # Run single test function
-cd hindsight-api-slim && uv run pytest tests/test_retain.py::test_retain_simple -v
+cd hindsight-api && uv run pytest tests/test_retain.py::test_retain_simple -v
 
 # Lint and format
-cd hindsight-api-slim && uv run ruff check .
-cd hindsight-api-slim && uv run ruff format .
+cd hindsight-api && uv run ruff check .
+cd hindsight-api && uv run ruff format .
 
 # Type checking (uses ty - extremely fast type checker from Astral)
-cd hindsight-api-slim && uv run ty check hindsight_api/
+cd hindsight-api && uv run ty check hindsight_api/
 ```
 
 ### Control Plane (Next.js)
@@ -57,22 +57,15 @@ cd hindsight-control-plane && npm run dev
 
 ### Benchmarks
 ```bash
-# Accuracy benchmarks
 ./scripts/benchmarks/run-longmemeval.sh
 ./scripts/benchmarks/run-locomo.sh
-
-# Performance benchmarks
-./scripts/benchmarks/run-consolidation.sh
-./scripts/benchmarks/run-retain-perf.sh --document <path>  # Requires API server running
-
-# Results viewer
 ./scripts/benchmarks/start-visualizer.sh  # View results at localhost:8001
 ```
 
 ## Architecture
 
 ### Monorepo Structure
-- **hindsight-api-slim/**: Core FastAPI server with memory engine (Python, uv)
+- **hindsight-api/**: Core FastAPI server with memory engine (Python, uv)
 - **hindsight/**: Embedded Python bundle (hindsight-all package)
 - **hindsight-control-plane/**: Admin UI (Next.js, npm)
 - **hindsight-cli/**: CLI tool (Rust, cargo, uses progenitor for API client)
@@ -81,9 +74,9 @@ cd hindsight-control-plane && npm run dev
 - **hindsight-integrations/**: Framework integrations (LiteLLM, OpenAI)
 - **hindsight-dev/**: Development tools and benchmarks
 
-### Core Engine (hindsight-api-slim/hindsight_api/engine/)
+### Core Engine (hindsight-api/hindsight_api/engine/)
 - `memory_engine.py`: Main orchestrator (~170KB) for retain/recall/reflect operations
-- `llm_wrapper.py`: LLM abstraction supporting OpenAI, Anthropic, Gemini, Groq, MiniMax, Ollama, LM Studio
+- `llm_wrapper.py`: LLM abstraction supporting OpenAI, Anthropic, Gemini, Groq, Ollama, LM Studio
 - `embeddings.py`: Embedding generation (local sentence-transformers or TEI)
 - `cross_encoder.py`: Reranking (local or TEI)
 - `entity_resolver.py`: Entity extraction and normalization
@@ -101,7 +94,7 @@ cd hindsight-control-plane && npm run dev
 - `fusion.py`: Reciprocal rank fusion for combining results
 - `reranking.py`: Cross-encoder reranking
 
-### API Layer (hindsight-api-slim/hindsight_api/api/)
+### API Layer (hindsight-api/hindsight_api/api/)
 - `http.py`: FastAPI HTTP routers (~80KB) for all REST endpoints
 - `mcp.py`: Model Context Protocol server implementation
 
@@ -111,13 +104,13 @@ Main operations:
 - **Reflect**: Disposition-aware reasoning using memories and mental models.
 
 ### Database
-PostgreSQL with pgvector. Schema managed via Alembic migrations in `hindsight-api-slim/hindsight_api/alembic/`. Migrations run automatically on API startup.
+PostgreSQL with pgvector. Schema managed via Alembic migrations in `hindsight-api/hindsight_api/alembic/`. Migrations run automatically on API startup.
 
 Key tables: `banks`, `memory_units`, `documents`, `entities`, `entity_links`
 
 ### Adding Database Migrations
 
-1. **Create a new migration file** in `hindsight-api-slim/hindsight_api/alembic/versions/`:
+1. **Create a new migration file** in `hindsight-api/hindsight_api/alembic/versions/`:
    - File name format: `<revision_id>_<description>.py` (e.g., `f1a2b3c4d5e6_add_new_index.py`)
    - Use a unique hex revision ID (12 chars)
    - Set `down_revision` to the previous migration's revision ID
@@ -154,7 +147,7 @@ Key tables: `banks`, `memory_units`, `documents`, `entities`, `entity_links`
 
 3. **Run migrations locally**:
    ```bash
-   # Set database URL and run migrations for the base schema plus all tenants
+   # Set database URL and run migrations
    uv run hindsight-admin run-db-migration
 
    # Run on a specific tenant schema
@@ -251,7 +244,7 @@ Fields must be categorized as either **hierarchical** (can be overridden per-ten
 
 #### Adding a New Configuration Field
 
-1. **config.py** (`hindsight-api-slim/hindsight_api/config.py`):
+1. **config.py** (`hindsight-api/hindsight_api/config.py`):
    - Add `ENV_*` constant for the environment variable name (e.g., `ENV_MY_SETTING = "HINDSIGHT_API_MY_SETTING"`)
    - Add `DEFAULT_*` constant for the default value
    - Add field to `HindsightConfig` dataclass with type annotation
@@ -268,7 +261,7 @@ Fields must be categorized as either **hierarchical** (can be overridden per-ten
    # Static field - just don't add to _HIERARCHICAL_FIELDS
    ```
 
-2. **main.py** (`hindsight-api-slim/hindsight_api/main.py`):
+2. **main.py** (`hindsight-api/hindsight_api/main.py`):
    - Add field to the manual `HindsightConfig()` constructor call (search for "CLI override")
 
 3. **Use hierarchical config in MemoryEngine**:
@@ -308,19 +301,19 @@ cp .env.example .env
 # Edit .env with LLM API key
 
 # Python deps
-uv sync --directory hindsight-api-slim/
+uv sync --directory hindsight-api/
 
 # Node deps (uses npm workspaces)
 npm install
 ```
 
 Required env vars:
-- `HINDSIGHT_API_LLM_PROVIDER`: openai, anthropic, gemini, groq, minimax, ollama, lmstudio
+- `HINDSIGHT_API_LLM_PROVIDER`: openai, anthropic, gemini, groq, ollama, lmstudio
 - `HINDSIGHT_API_LLM_API_KEY`: Your API key
-- `HINDSIGHT_API_LLM_MODEL`: Model name (e.g., gpt-4o-mini, claude-sonnet-4-20250514)
+- `HINDSIGHT_API_LLM_MODEL`: Model name (e.g., o3-mini, claude-sonnet-4-20250514)
 
 Optional (uses local models by default):
 - `HINDSIGHT_API_EMBEDDINGS_PROVIDER`: local (default) or tei
 - `HINDSIGHT_API_RERANKER_PROVIDER`: local (default) or tei
 - `HINDSIGHT_API_DATABASE_URL`: External PostgreSQL (uses embedded pg0 by default)
-- `HINDSIGHT_API_ENABLE_BANK_CONFIG_API`: Enable per-bank config API (default: true)
+- `HINDSIGHT_API_ENABLE_BANK_CONFIG_API`: Enable per-bank config API (default: false, disabled for security)

README.md

@@ -7,12 +7,10 @@
 [![CI](https://github.com/vectorize-io/hindsight/actions/workflows/release.yml/badge.svg)](https://github.com/vectorize-io/hindsight/actions/workflows/release.yml)
 [![Slack Community](https://img.shields.io/badge/Slack-Join%20Community-4A154B?logo=slack)](https://join.slack.com/t/hindsight-space/shared_invite/zt-3nhbm4w29-LeSJ5Ixi6j8PdiYOCPlOgg)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
-[![gitcgr](https://gitcgr.com/badge/vectorize-io/hindsight.svg)](https://gitcgr.com/vectorize-io/hindsight)
 ![PyPI - Downloads](https://img.shields.io/pypi/dm/hindsight-api?label=PyPI)
 ![NPM Downloads](https://img.shields.io/npm/dm/%40vectorize-io%2Fhindsight-client?logoColor=orange&label=NPM&color=blue&link=https%3A%2F%2Fwww.npmjs.com%2Fpackage%2F%40vectorize-io%2Fhindsight-client)
-<br/>
 
-<a href="https://trendshift.io/repositories/15603" target="_blank"><img src="https://trendshift.io/api/badge/repositories/15603" alt="vectorize-io%2Fhindsight | Trendshift" style="width: 250px; height: 55px;" width="250" height="55"/></a>
+
 </div>
 
 ---
@@ -38,7 +36,7 @@ Hindsight is being used in production at Fortune 500 enterprises and by a growin
 
 ## Adding Hindsight to Your AI Agents
 
-The easiest way to use Hindsight with an existing agent is with the LLM Wrapper. You can add memory to your agent with 2 lines of code. That will swap your current LLM client out with the Hindsight wrapper. After that, memories will be stored and retrieved automatically as you make LLM calls.
+The easiest way use Hindsight with an existing agent is with the LLM Wrapper. You can add memory to your agent with 2 lines of code. That will swap your current LLM client out with the Hindsight wrapper. After that, memories will be stored and retrieved automatically as you make LLM calls.
 
 If you need more control over how and when your agent stores and recalls memories, there's also a simple API you can integrate with using the SDKs or directly via HTTP.
 
@@ -71,7 +69,7 @@ docker run --rm -it --pull always -p 8888:8888 -p 9999:9999 \
 >API: http://localhost:8888
 >UI: http://localhost:9999
 
-You can modify the LLM provider by setting `HINDSIGHT_API_LLM_PROVIDER`. Valid options are `openai`, `anthropic`, `gemini`, `groq`, `ollama`, `lmstudio`, and `minimax`. The documentation provides more details on [supported models](https://hindsight.vectorize.io/developer/models).
+You can modify the LLM provider by setting `HINDSIGHT_API_LLM_PROVIDER`. Valid options are `openai`, `anthropic`, `gemini`, `groq`, `ollama`, and `lmstudio`. The documentation provides more details on [supported models](https://hindsight.vectorize.io/developer/models).
 
 
 
@@ -183,7 +181,7 @@ Satisfying these requirements in Hindsight is straightforward. When new user inp
 
 ![Overview](./hindsight-docs/static/img/hindsight-overview.webp)
 
-Most agent memory implementations rely on basic vector search or sometimes use a knowledge graph. Hindsight uses biomimetic data structures to organize agent memories in a way that is more like how human memory works:
+Most agent memory implementation rely on basic vector search or sometimes use a knowledge graph. Hindsight uses biomimetic data structures to organize agent memories in a way that is more like how human memory works:
 
 - **World:** Facts about the world ("The stove gets hot")
 - **Experiences:** Agent's own experiences ("I touched the stove and it really hurt")
@@ -309,5 +307,3 @@ MIT — see [LICENSE](./LICENSE)
 ---
 
 Built by [Vectorize.io](https://vectorize.io)
-
-<img src="https://umami-pixel.chris-latimer.workers.dev/?id=a8b043e6-6964-454d-80df-69b69d3f0d50&host=github.com&url=/vectorize-io/hindsight" width="1" height="1" alt="" />

deno.lock

@@ -1,139 +0,0 @@
-{
-  "version": "5",
-  "specifiers": {
-    "jsr:@std/assert@^1.0.17": "1.0.19",
-    "jsr:@std/assert@^1.0.19": "1.0.19",
-    "jsr:@std/expect@*": "1.0.18",
-    "jsr:@std/internal@^1.0.12": "1.0.12",
-    "jsr:@std/path@^1.1.4": "1.1.4",
-    "jsr:@std/testing@*": "1.0.17"
-  },
-  "jsr": {
-    "@std/[email protected]": {
-      "integrity": "eaada96ee120cb980bc47e040f82814d786fe8162ecc53c91d8df60b8755991e",
-      "dependencies": [
-        "jsr:@std/internal"
-      ]
-    },
-    "@std/[email protected]": {
-      "integrity": "8566eab35200466f8609eb7e7aed062ed0db314e9a258d5d201b1b8997ce801a",
-      "dependencies": [
-        "jsr:@std/assert@^1.0.19",
-        "jsr:@std/internal",
-        "jsr:@std/path"
-      ]
-    },
-    "@std/[email protected]": {
-      "integrity": "972a634fd5bc34b242024402972cd5143eac68d8dffaca5eaa4dba30ce17b027"
-    },
-    "@std/[email protected]": {
-      "integrity": "1d2d43f39efb1b42f0b1882a25486647cb851481862dc7313390b2bb044314b5",
-      "dependencies": [
-        "jsr:@std/internal"
-      ]
-    },
-    "@std/[email protected]": {
-      "integrity": "87bdc2700fa98249d48a17cd72413352d3d3680dcfbdb64947fd0982d6bbf681",
-      "dependencies": [
-        "jsr:@std/assert@^1.0.17",
-        "jsr:@std/internal"
-      ]
-    }
-  },
-  "workspace": {
-    "members": {
-      "hindsight-clients/typescript": {
-        "packageJson": {
-          "dependencies": [
-            "npm:@hey-api/[email protected]",
-            "npm:@types/jest@29",
-            "npm:@types/node@20",
-            "npm:jest@29",
-            "npm:ts-jest@29",
-            "npm:tsup@^8.5.1",
-            "npm:typescript@5"
-          ]
-        }
-      },
-      "hindsight-control-plane": {
-        "packageJson": {
-          "dependencies": [
-            "npm:@eslint/eslintrc@^3.3.3",
-            "npm:@eslint/js@^9.39.2",
-            "npm:@radix-ui/react-alert-dialog@^1.1.15",
-            "npm:@radix-ui/react-checkbox@^1.3.3",
-            "npm:@radix-ui/react-dialog@^1.1.15",
-            "npm:@radix-ui/react-dropdown-menu@^2.1.16",
-            "npm:@radix-ui/react-label@^2.1.8",
-            "npm:@radix-ui/react-popover@^1.1.15",
-            "npm:@radix-ui/react-radio-group@^1.3.8",
-            "npm:@radix-ui/react-select@^2.2.6",
-            "npm:@radix-ui/react-slider@^1.3.6",
-            "npm:@radix-ui/react-slot@^1.2.4",
-            "npm:@radix-ui/react-switch@^1.2.6",
-            "npm:@radix-ui/react-tabs@^1.1.13",
-            "npm:@radix-ui/react-tooltip@^1.2.8",
-            "npm:@tailwindcss/postcss@^4.1.17",
-            "npm:@tailwindcss/typography@~0.5.19",
-            "npm:@types/cytoscape@^3.21.9",
-            "npm:@types/node@^24.10.0",
-            "npm:@types/react-dom@^19.2.2",
-            "npm:@types/react@^19.2.2",
-            "npm:autoprefixer@^10.4.21",
-            "npm:class-variance-authority@~0.7.1",
-            "npm:clsx@^2.1.1",
-            "npm:cmdk@^1.1.1",
-            "npm:cytoscape-fcose@^2.2.0",
-            "npm:cytoscape@^3.33.1",
-            "npm:eslint-config-next@^16.0.1",
-            "npm:eslint-plugin-react-hooks@^7.0.1",
-            "npm:eslint-plugin-react@^7.37.5",
-            "npm:eslint@^9.39.1",
-            "npm:[email protected]",
-            "npm:next-themes@~0.4.6",
-            "npm:next@^16.1.6",
-            "npm:postcss@^8.5.6",
-            "npm:prettier@^3.7.4",
-            "npm:react-chrono@^2.9.1",
-            "npm:react-dom@^19.2.0",
-            "npm:react-markdown@^10.1.0",
-            "npm:react18-json-view@~0.2.9",
-            "npm:react@^19.2.0",
-            "npm:recharts@^3.5.1",
-            "npm:remark-gfm@^4.0.1",
-            "npm:sonner@^2.0.7",
-            "npm:tailwind-merge@^3.4.0",
-            "npm:tailwindcss-animate@^1.0.7",
-            "npm:tailwindcss@^4.1.17",
-            "npm:[email protected]",
-            "npm:typescript-eslint@^8.50.0",
-            "npm:typescript@^5.9.3"
-          ]
-        }
-      },
-      "hindsight-docs": {
-        "packageJson": {
-          "dependencies": [
-            "npm:@docusaurus/[email protected]",
-            "npm:@docusaurus/[email protected]",
-            "npm:@docusaurus/[email protected]",
-            "npm:@docusaurus/theme-common@^3.9.2",
-            "npm:@docusaurus/theme-mermaid@^3.9.2",
-            "npm:@docusaurus/[email protected]",
-            "npm:@docusaurus/[email protected]",
-            "npm:@easyops-cn/docusaurus-search-local@~0.52.2",
-            "npm:@mdx-js/react@3",
-            "npm:clsx@2",
-            "npm:prism-react-renderer@^2.3.0",
-            "npm:raw-loader@^4.0.2",
-            "npm:react-dom@19",
-            "npm:react-icons@^5.6.0",
-            "npm:react@19",
-            "npm:redocusaurus@^2.5.0",
-            "npm:typescript@~5.6.2"
-          ]
-        }
-      }
-    }
-  }
-}

docker/docker-compose/pg_textsearch/Dockerfile

@@ -1,32 +0,0 @@
-# PostgreSQL with pgvector and pg_textsearch extensions
-# Note: pg_textsearch requires PostgreSQL 17+
-FROM postgres:17
-
-# Install build dependencies
-RUN apt-get update && apt-get install -y \
-    build-essential \
-    git \
-    postgresql-server-dev-17 \
-    libpq-dev \
-    && rm -rf /var/lib/apt/lists/*
-
-# Install pgvector
-RUN cd /tmp && \
-    git clone --branch v0.8.0 https://github.com/pgvector/pgvector.git && \
-    cd pgvector && \
-    make && \
-    make install
-
-# Install pg_textsearch
-RUN cd /tmp && \
-    git clone https://github.com/timescale/pg_textsearch.git && \
-    cd pg_textsearch && \
-    make && \
-    make install
-
-# Clean up source files and build dependencies
-RUN rm -rf /tmp/pgvector /tmp/pg_textsearch && \
-    apt-get purge -y --auto-remove build-essential git postgresql-server-dev-17
-
-# Ensure extensions are preloaded
-RUN echo "shared_preload_libraries = 'pg_textsearch'" >> /usr/share/postgresql/postgresql.conf.sample

docker/docker-compose/pg_textsearch/docker-compose.yaml

@@ -1,91 +0,0 @@
-name: hindsight
-# Docker Compose file for Hindsight with PostgreSQL and Timescale pg_textsearch
-# docker compose -f docker/docker-compose/pg_textsearch/docker-compose.yaml down && sleep 2 && docker compose -f docker/docker-compose/pg_textsearch/docker-compose.yaml up -d
-# Make sure to set the required environment variables before running:
-# - HINDSIGHT_DB_PASSWORD: Password for the PostgreSQL user
-# - Configure LLM provider variables as needed (see below in the hindsight service)
-#
-# Usage:
-# docker compose up -d
-#
-# Optional environment variables with defaults:
-# - HINDSIGHT_VERSION: Hindsight application version (default: latest)
-# - HINDSIGHT_DB_USER: PostgreSQL user (default: hindsight_user)
-# - HINDSIGHT_DB_NAME: PostgreSQL database name (default: hindsight_db)
-
-services:
-  db:
-    # Use custom PostgreSQL image with pgvector and pg_textsearch extensions
-    build:
-      context: .
-      dockerfile: Dockerfile
-    container_name: hindsight-db
-    restart: always
-    # Expose PostgreSQL port
-    ports:
-      - "5437:5432"
-    environment:
-      POSTGRES_USER: ${HINDSIGHT_DB_USER:-hindsight_user}
-      POSTGRES_PASSWORD: ${HINDSIGHT_DB_PASSWORD:-hindsight_password}
-      POSTGRES_DB: ${HINDSIGHT_DB_NAME:-hindsight_db}
-    volumes:
-      - pg_data:/var/lib/postgresql/data
-    networks:
-      - hindsight-net
-
-  pg-textsearch-init:
-    build:
-      context: .
-      dockerfile: Dockerfile
-    depends_on:
-      - db
-    environment:
-      - PGPASSWORD=${HINDSIGHT_DB_PASSWORD:-hindsight_password}
-    command: >
-      bash -c "
-        echo 'Waiting for PostgreSQL to be ready...';
-        until pg_isready -h hindsight-db -p 5432 -U hindsight_user; do
-          echo 'PostgreSQL is unavailable - sleeping';
-          sleep 2;
-        done;
-        echo 'PostgreSQL is ready - creating hindsight_db database';
-        psql -h hindsight-db -p 5432 -U hindsight_user -c 'CREATE DATABASE hindsight_db;' 2>/dev/null || echo 'Database already exists';
-        echo 'Creating extensions in hindsight_db database';
-        psql -h hindsight-db -p 5432 -U hindsight_user -d hindsight_db -c 'CREATE EXTENSION IF NOT EXISTS vector CASCADE;';
-        psql -h hindsight-db -p 5432 -U hindsight_user -d hindsight_db -c 'CREATE EXTENSION IF NOT EXISTS pg_textsearch CASCADE;';
-        echo 'Database and extensions created successfully';
-      "
-    restart: "no"
-    networks:
-      - hindsight-net
-
-  hindsight:
-    image: ghcr.io/vectorize-io/hindsight:${HINDSIGHT_VERSION:-latest}
-    container_name: hindsight-app
-    ports:
-      - "8888:8888"
-      - "9999:9999"
-    environment:
-      # LLM Configuration
-      HINDSIGHT_API_LLM_PROVIDER: ${HINDSIGHT_API_LLM_PROVIDER:-openai}
-      HINDSIGHT_API_LLM_API_KEY: ${OPENAI_API_KEY:-your-api-key}
-
-      # Database Configuration
-      HINDSIGHT_API_DATABASE_URL: postgresql://${HINDSIGHT_DB_USER:-hindsight_user}:${HINDSIGHT_DB_PASSWORD:-hindsight_password}@db:5432/${HINDSIGHT_DB_NAME:-hindsight_db}
-
-      # Vector and Text Search Extensions
-      HINDSIGHT_API_VECTOR_EXTENSION: pgvector
-      HINDSIGHT_API_TEXT_SEARCH_EXTENSION: pg_textsearch
-
-    depends_on:
-      - db
-    networks:
-      - hindsight-net
-
-
-networks:
-  hindsight-net:
-    driver: bridge
-
-volumes:
-  pg_data:

docker/docker-compose/s3-file-storage/docker-compose.yaml

@@ -1,83 +0,0 @@
-# Docker Compose file for Hindsight with S3 file storage (SeaweedFS)
-#
-# SeaweedFS (Apache 2.0) provides an S3-compatible object storage backend
-# for storing uploaded files instead of PostgreSQL BYTEA storage.
-#
-# Make sure to set the required environment variables before running:
-# - HINDSIGHT_DB_PASSWORD: Password for the PostgreSQL user
-# - Configure LLM provider variables as needed (see below in the hindsight service)
-#
-# Usage:
-# docker compose up -d
-#
-# Optional environment variables with defaults:
-# - HINDSIGHT_VERSION: Hindsight application version (default: latest)
-# - HINDSIGHT_DB_USER: PostgreSQL user (default: hindsight_user)
-# - HINDSIGHT_DB_NAME: PostgreSQL database name (default: hindsight_db)
-# - HINDSIGHT_DB_VERSION: PostgreSQL version (default: 18)
-# - SEAWEEDFS_S3_ACCESS_KEY: S3 access key (default: hindsight_s3_key)
-# - SEAWEEDFS_S3_SECRET_KEY: S3 secret key (default: hindsight_s3_secret)
-
-services:
-  db:
-    image: pgvector/pgvector:pg${HINDSIGHT_DB_VERSION:-18}
-    container_name: hindsight-db
-    restart: always
-    environment:
-      POSTGRES_USER: ${HINDSIGHT_DB_USER:-hindsight_user}
-      POSTGRES_PASSWORD: ${HINDSIGHT_DB_PASSWORD:?Please set the HINDSIGHT_DB_PASSWORD env variable}
-      POSTGRES_DB: ${HINDSIGHT_DB_NAME:-hindsight_db}
-    volumes:
-      - pg_data:/var/lib/postgresql/${HINDSIGHT_DB_VERSION:-18}/docker
-    networks:
-      - hindsight-net
-
-  seaweedfs:
-    image: chrislusf/seaweedfs:latest
-    container_name: hindsight-seaweedfs
-    restart: always
-    # Single-node mode: master + volume + filer + S3 gateway all in one process
-    command: >
-      server
-      -s3
-      -s3.port=8333
-      -s3.config=/etc/seaweedfs/s3.json
-      -ip.bind=0.0.0.0
-    volumes:
-      - seaweedfs_data:/data
-      - ./s3.json:/etc/seaweedfs/s3.json:ro
-    # Expose S3 API port (uncomment to access from host)
-    # ports:
-    #   - "8333:8333"
-    networks:
-      - hindsight-net
-
-  hindsight:
-    image: ghcr.io/vectorize-io/hindsight:${HINDSIGHT_VERSION:-latest}
-    container_name: hindsight-app
-    ports:
-      - "8888:8888"
-      - "9999:9999"
-    environment:
-      - HINDSIGHT_API_LLM_API_KEY=${OPENAI_API_KEY?Please set the OPENAI_API_KEY env variable}
-      - HINDSIGHT_API_DATABASE_URL=postgresql://${HINDSIGHT_DB_USER:-hindsight_user}:${HINDSIGHT_DB_PASSWORD:?Please set the HINDSIGHT_DB_PASSWORD env variable}@db:5432/${HINDSIGHT_DB_NAME:-hindsight_db}
-      # S3 file storage configuration (SeaweedFS)
-      - HINDSIGHT_API_FILE_STORAGE_TYPE=s3
-      - HINDSIGHT_API_FILE_STORAGE_S3_BUCKET=hindsight
-      - HINDSIGHT_API_FILE_STORAGE_S3_ENDPOINT=http://seaweedfs:8333
-      - HINDSIGHT_API_FILE_STORAGE_S3_REGION=us-east-1
-      - HINDSIGHT_API_FILE_STORAGE_S3_ACCESS_KEY_ID=${SEAWEEDFS_S3_ACCESS_KEY:-hindsight_s3_key}
-      - HINDSIGHT_API_FILE_STORAGE_S3_SECRET_ACCESS_KEY=${SEAWEEDFS_S3_SECRET_KEY:-hindsight_s3_secret}
-    depends_on:
-      - db
-      - seaweedfs
-    networks:
-      - hindsight-net
-
-networks:
-  hindsight-net:
-    driver: bridge
-
-volumes:
-  pg_data:
-  seaweedfs_data:

docker/docker-compose/s3-file-storage/s3.json

@@ -1,19 +0,0 @@
-{
-  "identities": [
-    {
-      "name": "hindsight",
-      "credentials": [
-        {
-          "accessKey": "hindsight_s3_key",
-          "secretKey": "hindsight_s3_secret"
-        }
-      ],
-      "actions": [
-        "Admin",
-        "Read",
-        "Write",
-        "List"
-      ]
-    }
-  ]
-}

docker/docker-compose/timescale/.dockerignore

@@ -1,16 +0,0 @@
-# Git
-.git
-.gitignore
-.gitattributes
-
-# Docker
-docker-compose.yaml
-.dockerignore
-
-# Documentation
-README.md
-*.md
-
-# Environment
-.env
-.env.example

docker/docker-compose/timescale/.env.example

@@ -1,25 +0,0 @@
-# PostgreSQL Configuration
-HINDSIGHT_DB_USER=hindsight_user
-HINDSIGHT_DB_PASSWORD=change-me-to-secure-password
-HINDSIGHT_DB_NAME=hindsight_db
-
-# Hindsight Version
-HINDSIGHT_VERSION=latest
-
-# LLM Configuration
-HINDSIGHT_API_LLM_PROVIDER=openai
-OPENAI_API_KEY=your-openai-api-key-here
-
-# Alternative LLM providers (uncomment and configure as needed):
-# HINDSIGHT_API_LLM_PROVIDER=anthropic
-# ANTHROPIC_API_KEY=your-anthropic-api-key
-
-# HINDSIGHT_API_LLM_PROVIDER=gemini
-# GEMINI_API_KEY=your-gemini-api-key
-
-# HINDSIGHT_API_LLM_PROVIDER=groq
-# GROQ_API_KEY=your-groq-api-key
-
-# Vector and Text Search (already configured in docker-compose.yaml)
-# HINDSIGHT_API_VECTOR_EXTENSION=pgvectorscale
-# HINDSIGHT_API_TEXT_SEARCH_EXTENSION=pg_textsearch

docker/docker-compose/timescale/Dockerfile

@@ -1,55 +0,0 @@
-# PostgreSQL with pgvector, pgvectorscale, and pg_textsearch extensions
-# All three extensions from Timescale/pgvector for high-performance vector and text search
-# Note: Requires PostgreSQL 16+
-FROM postgres:17
-
-# Install build dependencies and Rust toolchain
-RUN apt-get update && apt-get install -y \
-    build-essential \
-    git \
-    postgresql-server-dev-17 \
-    libpq-dev \
-    cmake \
-    curl \
-    pkg-config \
-    libssl-dev \
-    && rm -rf /var/lib/apt/lists/*
-
-# Install Rust toolchain (required for pgvectorscale)
-RUN curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y
-ENV PATH="/root/.cargo/bin:${PATH}"
-
-# Install pgvector (required by pgvectorscale)
-RUN cd /tmp && \
-    git clone --branch v0.8.0 https://github.com/pgvector/pgvector.git && \
-    cd pgvector && \
-    make && \
-    make install && \
-    rm -rf /tmp/pgvector
-
-# Install cargo-pgrx (PostgreSQL extension framework for Rust)
-RUN cargo install cargo-pgrx --version 0.12.5 --locked && \
-    cargo pgrx init --pg17 /usr/bin/pg_config
-
-# Install pgvectorscale (DiskANN index support)
-RUN cd /tmp && \
-    git clone --branch 0.5.1 https://github.com/timescale/pgvectorscale.git && \
-    cd pgvectorscale/pgvectorscale && \
-    cargo pgrx install --release && \
-    rm -rf /tmp/pgvectorscale
-
-# Install pg_textsearch (BM25 text search)
-RUN cd /tmp && \
-    git clone https://github.com/timescale/pg_textsearch.git && \
-    cd pg_textsearch && \
-    make && \
-    make install && \
-    rm -rf /tmp/pg_textsearch
-
-# Clean up build dependencies (keep runtime dependencies)
-RUN apt-get purge -y --auto-remove git cmake curl && \
-    rm -rf /root/.cargo/registry /root/.cargo/git
-
-# Ensure extensions are preloaded (pg_textsearch requires preloading)
-RUN echo "shared_preload_libraries = 'pg_textsearch'" >> /usr/share/postgresql/postgresql.conf.sample
-

docker/docker-compose/timescale/README.md

@@ -1,101 +0,0 @@
-# Hindsight with Timescale Extensions
-
-This Docker Compose setup provides a complete Hindsight deployment with **Timescale extensions**:
-- **pgvectorscale** - DiskANN algorithm for disk-based scalable vector search  
-- **pg_textsearch** - High-performance BM25 text search
-
-Both extensions are from [Timescale](https://github.com/timescale) and provide production-grade performance.
-
-## Prerequisites
-
-- Docker and Docker Compose installed
-- OpenAI API key (or another LLM provider)
-
-## Quick Start
-
-```bash
-# Set environment variables
-export HINDSIGHT_DB_PASSWORD="your-secure-password"
-export OPENAI_API_KEY="your-openai-api-key"
-
-# Build and start
-docker compose -f docker/docker-compose/timescale/docker-compose.yaml up -d --build
-
-# Check logs
-
-docker compose -f docker/docker-compose/timescale/docker-compose.yaml logs -f
-```
-
-**Access:**
-- API: http://localhost:8888
-- Control Plane: http://localhost:9999
-
-## Stop and Clean Up
-
-```bash
-# Stop services
-docker compose -f docker/docker-compose/timescale/docker-compose.yaml down
-
-# Remove volumes (deletes all data)
-docker compose -f docker/docker-compose/timescale/docker-compose.yaml down -v
-```
-
-## Configuration
-
-### Environment Variables
-
-| Variable | Description | Default |
-|----------|-------------|---------|
-| `HINDSIGHT_DB_PASSWORD` | PostgreSQL password | `hindsight_password` |
-| `HINDSIGHT_DB_USER` | PostgreSQL username | `hindsight_user` |
-| `HINDSIGHT_DB_NAME` | Database name | `hindsight_db` |
-| `HINDSIGHT_VERSION` | Hindsight Docker image version | `latest` |
-| `OPENAI_API_KEY` | OpenAI API key | (required) |
-| `HINDSIGHT_API_LLM_PROVIDER` | LLM provider | `openai` |
-
-### Why Timescale Extensions?
-
-**pgvectorscale (DiskANN):**
-- 28x lower p95 latency vs dedicated vector databases
-- 16x higher query throughput at 99% recall
-- 60-75% cost reduction (disk is cheaper than RAM)
-- Best for large datasets (10M+ vectors)
-
-**pg_textsearch (BM25):**
-- High-performance keyword retrieval
-- Native BM25 ranking algorithm
-- Optimized for full-text search
-
-## Troubleshooting
-
-### Extensions not installed
-
-Check if extensions are available:
-
-```bash
-docker exec -it hindsight-db-timescale psql -U hindsight_user -d hindsight_db -c "\dx"
-```
-
-You should see:
-- `vector` (pgvector)
-- `vectorscale` (pgvectorscale/DiskANN)
-- `pg_textsearch` (BM25 search)
-
-### Build fails
-
-If the Docker build fails during pgvectorscale compilation:
-
-1. Ensure you have sufficient memory (recommended: 4GB+)
-2. Check Docker build logs for Rust compilation errors
-3. Try building with more resources: `docker compose build --no-cache --memory 4g`
-
-### Port conflicts
-
-If port 5438 is already in use, modify the `ports` section in docker-compose.yaml.
-
-## Learn More
-
-- [pgvectorscale GitHub](https://github.com/timescale/pgvectorscale)
-- [pg_textsearch GitHub](https://github.com/timescale/pg_textsearch)
-- [HNSW vs DiskANN](https://www.tigerdata.com/learn/hnsw-vs-diskann)
-- [Hindsight Documentation](https://hindsight.dev)

docker/docker-compose/timescale/docker-compose.yaml

@@ -1,108 +0,0 @@
-name: hindsight
-# Docker Compose file for Hindsight with Timescale extensions
-# - pgvectorscale: DiskANN vector search (disk-based, scalable)
-# - pg_textsearch: BM25 text search (high-performance keyword retrieval)
-#
-# Quick start:
-# docker compose -f docker/docker-compose/timescale/docker-compose.yaml up -d --build
-#
-# Required environment variables:
-# - HINDSIGHT_DB_PASSWORD: Password for the PostgreSQL user
-# - OPENAI_API_KEY (or configure another LLM provider)
-#
-# Optional environment variables with defaults:
-# - HINDSIGHT_VERSION: Hindsight application version (default: latest)
-# - HINDSIGHT_DB_USER: PostgreSQL user (default: hindsight_user)
-# - HINDSIGHT_DB_NAME: PostgreSQL database name (default: hindsight_db)
-
-services:
-  db:
-    # Custom PostgreSQL image with Timescale extensions (pgvectorscale + pg_textsearch)
-    build:
-      context: .
-      dockerfile: Dockerfile
-    container_name: hindsight-db-timescale
-    restart: always
-    # Expose PostgreSQL port (using 5438 to avoid conflicts with other setups)
-    ports:
-      - "5438:5432"
-    environment:
-      POSTGRES_USER: ${HINDSIGHT_DB_USER:-hindsight_user}
-      POSTGRES_PASSWORD: ${HINDSIGHT_DB_PASSWORD:-hindsight_password}
-      POSTGRES_DB: ${HINDSIGHT_DB_NAME:-hindsight_db}
-    volumes:
-      - pg_data:/var/lib/postgresql/data
-    networks:
-      - hindsight-net
-    # Health check to ensure database is ready
-    healthcheck:
-      test: ["CMD-SHELL", "pg_isready -U hindsight_user"]
-      interval: 5s
-      timeout: 5s
-      retries: 5
-
-  timescale-init:
-    build:
-      context: .
-      dockerfile: Dockerfile
-    depends_on:
-      db:
-        condition: service_healthy
-    environment:
-      - PGPASSWORD=${HINDSIGHT_DB_PASSWORD:-hindsight_password}
-    command: >
-      bash -c "
-        echo 'PostgreSQL is ready - creating hindsight_db database';
-        psql -h hindsight-db-timescale -p 5432 -U hindsight_user -c 'CREATE DATABASE hindsight_db;' 2>/dev/null || echo 'Database already exists';
-        echo 'Installing Timescale extensions...';
-        echo '1/3: Installing pgvector (required by pgvectorscale)...';
-        psql -h hindsight-db-timescale -p 5432 -U hindsight_user -d hindsight_db -c 'CREATE EXTENSION IF NOT EXISTS vector CASCADE;';
-        echo '2/3: Installing pgvectorscale (DiskANN vector search)...';
-        psql -h hindsight-db-timescale -p 5432 -U hindsight_user -d hindsight_db -c 'CREATE EXTENSION IF NOT EXISTS vectorscale CASCADE;';
-        echo '3/3: Installing pg_textsearch (BM25 text search)...';
-        psql -h hindsight-db-timescale -p 5432 -U hindsight_user -d hindsight_db -c 'CREATE EXTENSION IF NOT EXISTS pg_textsearch CASCADE;';
-        echo '';
-        echo '✅ Timescale extensions installed successfully';
-        echo '';
-        echo 'Installed extensions:';
-        psql -h hindsight-db-timescale -p 5432 -U hindsight_user -d hindsight_db -c \"\\dx\" | grep -E '(vector|vectorscale|pg_textsearch)';
-      "
-    restart: "no"
-    networks:
-      - hindsight-net
-
-  hindsight:
-    image: ghcr.io/vectorize-io/hindsight:${HINDSIGHT_VERSION:-latest}
-    container_name: hindsight-app-timescale
-    ports:
-      - "8888:8888"
-      - "9999:9999"
-    environment:
-      # LLM Configuration
-      HINDSIGHT_API_LLM_PROVIDER: ${HINDSIGHT_API_LLM_PROVIDER:-openai}
-      HINDSIGHT_API_LLM_API_KEY: ${OPENAI_API_KEY:-your-api-key}
-
-      # Database Configuration
-      HINDSIGHT_API_DATABASE_URL: postgresql://${HINDSIGHT_DB_USER:-hindsight_user}:${HINDSIGHT_DB_PASSWORD:-hindsight_password}@db:5432/${HINDSIGHT_DB_NAME:-hindsight_db}
-
-      # Timescale Extensions
-      # pgvectorscale: DiskANN algorithm for disk-based scalable vector search
-      HINDSIGHT_API_VECTOR_EXTENSION: pgvectorscale
-      # pg_textsearch: High-performance BM25 text search
-      HINDSIGHT_API_TEXT_SEARCH_EXTENSION: pg_textsearch
-
-    depends_on:
-      db:
-        condition: service_healthy
-      timescale-init:
-        condition: service_completed_successfully
-    networks:
-      - hindsight-net
-
-
-networks:
-  hindsight-net:
-    driver: bridge
-
-volumes:
-  pg_data:

docker/standalone/Dockerfile

@@ -42,22 +42,25 @@ RUN apt-get update && apt-get install -y \
     && pip install --no-cache-dir uv
 
 # Copy dependency files and README (required by pyproject.toml)
-COPY hindsight-api-slim/pyproject.toml ./api/
-COPY hindsight-api-slim/README.md ./api/
+COPY hindsight-api/pyproject.toml ./api/
+COPY hindsight-api/README.md ./api/
 
 WORKDIR /app/api
 
-# Sync dependencies using appropriate extras based on INCLUDE_LOCAL_MODELS
-# local-ml: torch, sentence-transformers, transformers, einops, flashrank, mlx (optional)
-# embedded-db: pg0-embedded (always included for embedded PostgreSQL support)
-RUN if [ "$INCLUDE_LOCAL_MODELS" = "true" ]; then \
-        uv sync --extra local-ml --extra embedded-db; \
-    else \
-        uv sync --extra embedded-db; \
+# Remove local ML model dependencies if INCLUDE_LOCAL_MODELS=false
+# This creates a smaller image when using external providers (TEI, OpenAI, Cohere)
+RUN if [ "$INCLUDE_LOCAL_MODELS" != "true" ]; then \
+        echo "Removing local-models dependencies (sentence-transformers, torch, transformers)..." && \
+        sed -i '/"sentence-transformers/d' pyproject.toml && \
+        sed -i '/"transformers/d' pyproject.toml && \
+        sed -i '/"torch/d' pyproject.toml; \
     fi
 
+# Sync dependencies (will create lock file if needed)
+RUN uv sync
+
 # Copy source code (alembic migrations are inside hindsight_api/)
-COPY hindsight-api-slim/hindsight_api ./hindsight_api
+COPY hindsight-api/hindsight_api ./hindsight_api
 
 # Install the local package (uv sync only installed dependencies, not the package itself)
 RUN uv pip install -e .
@@ -167,11 +170,6 @@ RUN chown -R hindsight:hindsight /app
 
 USER hindsight
 
-# Create pg0 data directory as hindsight user so that Docker seeds new named
-# volumes with correct ownership (UID 1000) on first use, avoiding the
-# "Permission denied" error when mounting a fresh root-owned volume.
-RUN mkdir -p /home/hindsight/.pg0
-
 ENV PATH="/app/api/.venv/bin:${PATH}"
 
 # Pre-download tiktoken encoding (ALWAYS - required for token counting even in air-gapped envs)
@@ -323,11 +321,6 @@ RUN chown -R hindsight:hindsight /app
 
 USER hindsight
 
-# Create pg0 data directory as hindsight user so that Docker seeds new named
-# volumes with correct ownership (UID 1000) on first use, avoiding the
-# "Permission denied" error when mounting a fresh root-owned volume.
-RUN mkdir -p /home/hindsight/.pg0
-
 ENV PATH="/app/api/.venv/bin:${PATH}"
 
 # Pre-download tiktoken encoding (ALWAYS - required for token counting even in air-gapped envs)

docker/standalone/start-all.sh

@@ -77,32 +77,18 @@ PIDS=()
 # Start API if enabled
 if [ "$ENABLE_API" = "true" ]; then
     cd /app/api
-    API_HEALTH_URL="${HINDSIGHT_API_HEALTH_URL:-http://localhost:8888/health}"
-    API_STARTUP_WAIT_SECONDS="${HINDSIGHT_API_STARTUP_WAIT_SECONDS:-300}"
-
     # Run API directly - Python's PYTHONUNBUFFERED=1 handles output buffering
     hindsight-api &
     API_PID=$!
     PIDS+=($API_PID)
 
     # Wait for API to be ready
-    api_ready=false
-    for ((i=1; i<=API_STARTUP_WAIT_SECONDS; i++)); do
-        if ! kill -0 "$API_PID" 2>/dev/null; then
-            wait "$API_PID"
-            exit $?
-        fi
-        if curl -sf "$API_HEALTH_URL" &>/dev/null; then
-            api_ready=true
+    for i in {1..60}; do
+        if curl -sf http://localhost:8888/health &>/dev/null; then
             break
         fi
         sleep 1
     done
-
-    if [ "$api_ready" != "true" ]; then
-        echo "❌ API did not become healthy within ${API_STARTUP_WAIT_SECONDS}s"
-        exit 1
-    fi
 else
     echo "API disabled (HINDSIGHT_ENABLE_API=false)"
 fi
@@ -111,8 +97,7 @@ fi
 if [ "$ENABLE_CP" = "true" ]; then
     echo "🎛️  Starting Control Plane..."
     cd /app/control-plane
-    export HOSTNAME="${HINDSIGHT_CP_HOSTNAME:-0.0.0.0}"
-    PORT="${HINDSIGHT_CP_PORT:-9999}" node server.js &
+    PORT=9999 node server.js &
     CP_PID=$!
     PIDS+=($CP_PID)
 else
@@ -125,7 +110,7 @@ echo "✅ Hindsight is running!"
 echo ""
 echo "📍 Access:"
 if [ "$ENABLE_CP" = "true" ]; then
-    echo "   Control Plane: http://localhost:${HINDSIGHT_CP_PORT:-9999}"
+    echo "   Control Plane: http://localhost:9999"
 fi
 if [ "$ENABLE_API" = "true" ]; then
     echo "   API:           http://localhost:8888"

docker/test-image.sh

@@ -13,9 +13,9 @@
 #   target  - Optional: 'cp-only' for control plane, otherwise assumes API image (default: api)
 #
 # Environment variables:
-#   HINDSIGHT_API_LLM_API_KEY                   - Required for API/standalone images (LLM verification)
-#   HINDSIGHT_API_LLM_PROVIDER                  - LLM provider (default: openai)
-#   HINDSIGHT_API_LLM_MODEL                     - LLM model (default: gpt-4o-mini)
+#   GROQ_API_KEY                                - Required for API/standalone images (LLM verification)
+#   HINDSIGHT_API_LLM_PROVIDER                  - LLM provider (default: groq)
+#   HINDSIGHT_API_LLM_MODEL                     - LLM model (default: llama-3.3-70b-versatile)
 #   HINDSIGHT_API_EMBEDDINGS_PROVIDER           - Embeddings provider (optional, for slim images: openai, cohere, tei)
 #   HINDSIGHT_API_EMBEDDINGS_OPENAI_API_KEY     - OpenAI API key for embeddings (optional)
 #   HINDSIGHT_API_RERANKER_PROVIDER             - Reranker provider (optional, for slim images: cohere, tei)
@@ -34,7 +34,7 @@
 #   ./docker/test-image.sh hindsight-control-plane:test cp-only
 #
 #   # Test slim image with external providers
-#   export HINDSIGHT_API_LLM_API_KEY=sk_xxx
+#   export GROQ_API_KEY=gsk_xxx
 #   export HINDSIGHT_API_EMBEDDINGS_PROVIDER=openai
 #   export HINDSIGHT_API_EMBEDDINGS_OPENAI_API_KEY=sk-xxx
 #   export HINDSIGHT_API_RERANKER_PROVIDER=cohere
@@ -49,9 +49,6 @@
 
 set -euo pipefail
 
-SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
-REPO_ROOT="$(dirname "$SCRIPT_DIR")"
-
 # Colors for output
 RED='\033[0;31m'
 GREEN='\033[0;32m'
@@ -63,8 +60,8 @@ IMAGE="${1:-}"
 TARGET="${2:-api}"
 TIMEOUT="${SMOKE_TEST_TIMEOUT:-120}"
 CONTAINER_NAME="${SMOKE_TEST_CONTAINER_NAME:-hindsight-smoke-test}"
-LLM_PROVIDER="${HINDSIGHT_API_LLM_PROVIDER:-openai}"
-LLM_MODEL="${HINDSIGHT_API_LLM_MODEL:-gpt-4o-mini}"
+LLM_PROVIDER="${HINDSIGHT_API_LLM_PROVIDER:-groq}"
+LLM_MODEL="${HINDSIGHT_API_LLM_MODEL:-llama-3.3-70b-versatile}"
 
 # Validate arguments
 if [ -z "$IMAGE" ]; then
@@ -91,9 +88,9 @@ else
 fi
 
 # Check for required environment variables
-if [ "$NEEDS_LLM" = true ] && [ "$LLM_PROVIDER" != "vertexai" ] && [ -z "${HINDSIGHT_API_LLM_API_KEY:-}" ]; then
-    echo -e "${RED}Error: HINDSIGHT_API_LLM_API_KEY environment variable is required for API/standalone images${NC}"
-    echo "Set it with: export HINDSIGHT_API_LLM_API_KEY=your-api-key"
+if [ "$NEEDS_LLM" = true ] && [ -z "${GROQ_API_KEY:-}" ]; then
+    echo -e "${RED}Error: GROQ_API_KEY environment variable is required for API/standalone images${NC}"
+    echo "Set it with: export GROQ_API_KEY=your-api-key"
     exit 2
 fi
 
@@ -126,25 +123,9 @@ else
     # Build docker run command with required and optional env vars
     DOCKER_CMD="docker run -d --name $CONTAINER_NAME"
     DOCKER_CMD="$DOCKER_CMD -e HINDSIGHT_API_LLM_PROVIDER=$LLM_PROVIDER"
-    if [ -n "${HINDSIGHT_API_LLM_API_KEY:-}" ]; then
-        DOCKER_CMD="$DOCKER_CMD -e HINDSIGHT_API_LLM_API_KEY=${HINDSIGHT_API_LLM_API_KEY}"
-    fi
+    DOCKER_CMD="$DOCKER_CMD -e HINDSIGHT_API_LLM_API_KEY=${GROQ_API_KEY}"
     DOCKER_CMD="$DOCKER_CMD -e HINDSIGHT_API_LLM_MODEL=$LLM_MODEL"
 
-    # Add Vertex AI config if provider is vertexai
-    if [ "$LLM_PROVIDER" = "vertexai" ]; then
-        if [ -n "${HINDSIGHT_API_LLM_VERTEXAI_SERVICE_ACCOUNT_KEY:-}" ]; then
-            DOCKER_CMD="$DOCKER_CMD -v ${HINDSIGHT_API_LLM_VERTEXAI_SERVICE_ACCOUNT_KEY}:/tmp/gcp-credentials.json:ro"
-            DOCKER_CMD="$DOCKER_CMD -e HINDSIGHT_API_LLM_VERTEXAI_SERVICE_ACCOUNT_KEY=/tmp/gcp-credentials.json"
-        fi
-        if [ -n "${HINDSIGHT_API_LLM_VERTEXAI_PROJECT_ID:-}" ]; then
-            DOCKER_CMD="$DOCKER_CMD -e HINDSIGHT_API_LLM_VERTEXAI_PROJECT_ID=${HINDSIGHT_API_LLM_VERTEXAI_PROJECT_ID}"
-        fi
-        if [ -n "${HINDSIGHT_API_LLM_VERTEXAI_REGION:-}" ]; then
-            DOCKER_CMD="$DOCKER_CMD -e HINDSIGHT_API_LLM_VERTEXAI_REGION=${HINDSIGHT_API_LLM_VERTEXAI_REGION}"
-        fi
-    fi
-
     # Add optional embeddings provider config
     if [ -n "${HINDSIGHT_API_EMBEDDINGS_PROVIDER:-}" ]; then
         DOCKER_CMD="$DOCKER_CMD -e HINDSIGHT_API_EMBEDDINGS_PROVIDER=${HINDSIGHT_API_EMBEDDINGS_PROVIDER}"
@@ -181,21 +162,6 @@ for i in $(seq 1 "$TIMEOUT"); do
         echo "=== Health Response ==="
         curl -s "http://localhost:${HEALTH_PORT}${HEALTH_PATH}" | python3 -m json.tool 2>/dev/null || curl -s "http://localhost:${HEALTH_PORT}${HEALTH_PATH}"
         echo ""
-
-        # Run retain/recall smoke test for API targets
-        if [ "$TARGET" != "cp-only" ]; then
-            echo ""
-            echo "=== Retain/Recall Smoke Test ==="
-            if ! "$REPO_ROOT/scripts/smoke-test-slim.sh" "http://localhost:${HEALTH_PORT}"; then
-                echo ""
-                echo "=== Container Logs (last 50 lines) ==="
-                docker logs "$CONTAINER_NAME" 2>&1 | tail -50
-                echo ""
-                echo -e "${RED}Smoke test FAILED${NC}"
-                exit 1
-            fi
-        fi
-
         echo ""
         echo "=== Container Logs (last 50 lines) ==="
         docker logs "$CONTAINER_NAME" 2>&1 | tail -50

docker/test-slim-local.sh

@@ -6,17 +6,24 @@
 # It expects API keys to be set in environment variables.
 #
 # Usage:
+#   export GROQ_API_KEY=gsk_xxx
 #   export OPENAI_API_KEY=sk-xxx
 #   export COHERE_API_KEY=xxx
 #   ./docker/test-slim-local.sh
 #
 # Or inline:
-#   OPENAI_API_KEY=sk_xxx COHERE_API_KEY=xxx ./docker/test-slim-local.sh
+#   GROQ_API_KEY=gsk_xxx OPENAI_API_KEY=sk_xxx COHERE_API_KEY=xxx ./docker/test-slim-local.sh
 #
 
 set -euo pipefail
 
 # Check for required API keys
+if [ -z "${GROQ_API_KEY:-}" ]; then
+    echo "❌ Error: GROQ_API_KEY environment variable is required"
+    echo "Set it with: export GROQ_API_KEY=gsk_xxx"
+    exit 1
+fi
+
 if [ -z "${OPENAI_API_KEY:-}" ]; then
     echo "❌ Error: OPENAI_API_KEY environment variable is required"
     echo "Set it with: export OPENAI_API_KEY=sk-xxx"
@@ -34,10 +41,7 @@ IMAGE="${1:-hindsight-slim:test}"
 echo "Testing image: $IMAGE"
 echo ""
 
-# Set up LLM and external providers
-export HINDSIGHT_API_LLM_PROVIDER=openai
-export HINDSIGHT_API_LLM_API_KEY=$OPENAI_API_KEY
-export HINDSIGHT_API_LLM_MODEL=gpt-4o-mini
+# Set up external providers
 export HINDSIGHT_API_EMBEDDINGS_PROVIDER=openai
 export HINDSIGHT_API_EMBEDDINGS_OPENAI_API_KEY=$OPENAI_API_KEY
 export HINDSIGHT_API_RERANKER_PROVIDER=cohere

helm/hindsight/Chart.yaml

@@ -2,8 +2,8 @@ apiVersion: v2
 name: hindsight
 description: Hindsight helm chart
 type: application
-version: 0.4.20
-appVersion: "0.4.20"
+version: 0.4.10
+appVersion: "0.4.10"
 keywords:
   - ai
   - memory

hindsight-all-slim/pyproject.toml

@@ -1,33 +0,0 @@
-[build-system]
-requires = ["setuptools>=61"]
-build-backend = "setuptools.build_meta"
-
-[project]
-name = "hindsight-all-slim"
-version = "0.4.20"
-description = "Hindsight: Agent Memory That Works Like Human Memory - Slim All-in-One Bundle"
-readme = "README.md"
-requires-python = ">=3.11"
-dependencies = [
-    "hindsight-api-slim>=0.4.17",
-    "hindsight-client>=0.0.7",
-    "hindsight-embed>=0.1.0",
-]
-
-[tool.uv.sources]
-hindsight-api-slim = { workspace = true }
-hindsight-client = { workspace = true }
-hindsight-embed = { workspace = true }
-
-[project.optional-dependencies]
-test = [
-    "pytest>=7.0.0",
-    "pytest-asyncio>=0.21.0",
-]
-
-[tool.setuptools]
-packages = []
-
-[tool.pytest.ini_options]
-asyncio_mode = "auto"
-asyncio_default_fixture_loop_scope = "function"

hindsight-all/README.md

@@ -1,48 +0,0 @@
-# hindsight-all
-
-All-in-one package for Hindsight - Agent Memory That Works Like Human Memory
-
-## Quick Start
-
-```python
-from hindsight import start_server, HindsightClient
-
-# Start server with embedded PostgreSQL
-server = start_server(
-    llm_provider="groq",
-    llm_api_key="your-api-key",
-    llm_model="openai/gpt-oss-120b"
-)
-
-# Create client
-client = HindsightClient(base_url=server.url)
-
-# Store memories
-client.put(agent_id="assistant", content="User prefers Python for data analysis")
-
-# Search memories
-results = client.search(agent_id="assistant", query="programming preferences")
-
-# Generate contextual response
-response = client.think(agent_id="assistant", query="What languages should I recommend?")
-
-# Stop server when done
-server.stop()
-```
-
-## Using Context Manager
-
-```python
-from hindsight import HindsightServer, HindsightClient
-
-with HindsightServer(llm_provider="groq", llm_api_key="...") as server:
-    client = HindsightClient(base_url=server.url)
-    # ... use client ...
-# Server automatically stops
-```
-
-## Installation
-
-```bash
-pip install hindsight-all
-```

hindsight-all/hindsight/client_wrapper.py

@@ -1,423 +0,0 @@
-"""
-Wrapper for Hindsight client that adds API namespaces.
-
-Provides organized access to different parts of the Hindsight API through
-namespaces like .banks, .mental_models, etc.
-"""
-
-from __future__ import annotations
-
-from typing import Any
-
-from hindsight_client import Hindsight
-
-
-class BanksAPI:
-    """Namespace for bank-related operations.
-    
-    Provides methods to create, delete, and manage memory banks.
-    """
-
-    def __init__(self, client: Hindsight):
-        self._client = client
-
-    def create(
-        self,
-        bank_id: str,
-        name: str | None = None,
-        mission: str | None = None,
-        disposition: dict[str, Any] | None = None,
-    ) -> Any:
-        """Create a new bank.
-        
-        Args:
-            bank_id: Unique identifier for the bank.
-            name: Optional display name for the bank.
-            mission: Optional mission statement for the bank.
-            disposition: Optional disposition configuration dict.
-            
-        Returns:
-            Bank creation response from the API.
-        """
-        return self._client.create_bank(
-            bank_id=bank_id,
-            name=name,
-            mission=mission,
-            disposition=disposition,
-        )
-
-    def delete(self, bank_id: str) -> Any:
-        """Delete a bank.
-        
-        Args:
-            bank_id: The ID of the bank to delete.
-            
-        Returns:
-            Deletion response from the API.
-        """
-        return self._client.delete_bank(bank_id=bank_id)
-
-    def set_mission(self, bank_id: str, mission: str) -> Any:
-        """Set or update the mission for a bank.
-        
-        Args:
-            bank_id: The ID of the bank.
-            mission: The mission statement to set.
-            
-        Returns:
-            API response confirming the update.
-        """
-        return self._client.set_mission(bank_id=bank_id, mission=mission)
-
-    def set_disposition(self, bank_id: str, disposition: dict[str, Any]) -> Any:
-        """Set or update the disposition for a bank.
-        
-        Args:
-            bank_id: The ID of the bank.
-            disposition: The disposition configuration dict.
-            
-        Returns:
-            API response confirming the update.
-        """
-        return self._client.set_disposition(bank_id=bank_id, disposition=disposition)
-
-    def list(self) -> Any:
-        """List all banks.
-        
-        Returns:
-            List of banks from the API.
-        """
-        from hindsight_client.hindsight_client import _run_async
-
-        return _run_async(self._client._banks_api.list_banks())
-
-
-class MentalModelsAPI:
-    """Namespace for mental model operations.
-    
-    Mental models are reusable knowledge structures that guide agent behavior.
-    """
-
-    def __init__(self, client: Hindsight):
-        self._client = client
-
-    def create(
-        self,
-        bank_id: str,
-        name: str,
-        content: str,
-        tags: list[str] | None = None,
-    ) -> Any:
-        """Create a new mental model.
-        
-        Args:
-            bank_id: The ID of the bank to add the model to.
-            name: Name for the mental model.
-            content: The content/instructions for the mental model.
-            tags: Optional list of tags for categorization.
-            
-        Returns:
-            Creation response from the API.
-        """
-        return self._client.create_mental_model(
-            bank_id=bank_id,
-            name=name,
-            content=content,
-            tags=tags,
-        )
-
-    def list(self, bank_id: str, tags: list[str] | None = None) -> Any:
-        """List all mental models for a bank.
-        
-        Args:
-            bank_id: The ID of the bank.
-            tags: Optional filter by tags.
-            
-        Returns:
-            List of mental models.
-        """
-        return self._client.list_mental_models(bank_id=bank_id, tags=tags)
-
-    def get(self, bank_id: str, mental_model_id: str) -> Any:
-        """Get a specific mental model.
-        
-        Args:
-            bank_id: The ID of the bank.
-            mental_model_id: The ID of the mental model.
-            
-        Returns:
-            The mental model details.
-        """
-        return self._client.get_mental_model(bank_id=bank_id, mental_model_id=mental_model_id)
-
-    def refresh(self, bank_id: str, mental_model_id: str) -> Any:
-        """Refresh a mental model.
-        
-        Args:
-            bank_id: The ID of the bank.
-            mental_model_id: The ID of the mental model to refresh.
-            
-        Returns:
-            Refresh response from the API.
-        """
-        return self._client.refresh_mental_model(bank_id=bank_id, mental_model_id=mental_model_id)
-
-    def update(
-        self,
-        bank_id: str,
-        mental_model_id: str,
-        name: str | None = None,
-        content: str | None = None,
-        tags: list[str] | None = None,
-    ) -> Any:
-        """Update a mental model.
-        
-        Args:
-            bank_id: The ID of the bank.
-            mental_model_id: The ID of the mental model to update.
-            name: Optional new name.
-            content: Optional new content.
-            tags: Optional new tags list.
-            
-        Returns:
-            Update response from the API.
-        """
-        return self._client.update_mental_model(
-            bank_id=bank_id,
-            mental_model_id=mental_model_id,
-            name=name,
-            content=content,
-            tags=tags,
-        )
-
-    def delete(self, bank_id: str, mental_model_id: str) -> Any:
-        """Delete a mental model.
-        
-        Args:
-            bank_id: The ID of the bank.
-            mental_model_id: The ID of the mental model to delete.
-            
-        Returns:
-            Deletion response from the API.
-        """
-        return self._client.delete_mental_model(bank_id=bank_id, mental_model_id=mental_model_id)
-
-
-class DirectivesAPI:
-    """Namespace for directive operations.
-    
-    Directives are explicit instructions that guide agent behavior.
-    """
-
-    def __init__(self, client: Hindsight):
-        self._client = client
-
-    def create(
-        self,
-        bank_id: str,
-        name: str,
-        content: str,
-        tags: list[str] | None = None,
-    ) -> Any:
-        """Create a new directive.
-        
-        Args:
-            bank_id: The ID of the bank to add the directive to.
-            name: Name for the directive.
-            content: The directive content/instructions.
-            tags: Optional list of tags for categorization.
-            
-        Returns:
-            Creation response from the API.
-        """
-        return self._client.create_directive(
-            bank_id=bank_id,
-            name=name,
-            content=content,
-            tags=tags,
-        )
-
-    def list(self, bank_id: str, tags: list[str] | None = None) -> Any:
-        """List all directives for a bank.
-        
-        Args:
-            bank_id: The ID of the bank.
-            tags: Optional filter by tags.
-            
-        Returns:
-            List of directives.
-        """
-        return self._client.list_directives(bank_id=bank_id, tags=tags)
-
-    def get(self, bank_id: str, directive_id: str) -> Any:
-        """Get a specific directive.
-        
-        Args:
-            bank_id: The ID of the bank.
-            directive_id: The ID of the directive.
-            
-        Returns:
-            The directive details.
-        """
-        return self._client.get_directive(bank_id=bank_id, directive_id=directive_id)
-
-    def update(
-        self,
-        bank_id: str,
-        directive_id: str,
-        name: str | None = None,
-        content: str | None = None,
-        tags: list[str] | None = None,
-    ) -> Any:
-        """Update a directive.
-        
-        Args:
-            bank_id: The ID of the bank.
-            directive_id: The ID of the directive to update.
-            name: Optional new name.
-            content: Optional new content.
-            tags: Optional new tags list.
-            
-        Returns:
-            Update response from the API.
-        """
-        return self._client.update_directive(
-            bank_id=bank_id,
-            directive_id=directive_id,
-            name=name,
-            content=content,
-            tags=tags,
-        )
-
-    def delete(self, bank_id: str, directive_id: str) -> Any:
-        """Delete a directive.
-        
-        Args:
-            bank_id: The ID of the bank.
-            directive_id: The ID of the directive to delete.
-            
-        Returns:
-            Deletion response from the API.
-        """
-        return self._client.delete_directive(bank_id=bank_id, directive_id=directive_id)
-
-
-class MemoriesAPI:
-    """Namespace for memory operations.
-    
-    Provides methods to query and retrieve stored memories.
-    """
-
-    def __init__(self, client: Hindsight):
-        self._client = client
-
-    def list(
-        self,
-        bank_id: str,
-        type: str | None = None,
-        search_query: str | None = None,
-        limit: int = 100,
-        offset: int = 0,
-    ) -> Any:
-        """List memories in a bank.
-        
-        Args:
-            bank_id: The ID of the bank to query.
-            type: Optional filter by memory type.
-            search_query: Optional search query for filtering.
-            limit: Maximum number of results to return (default: 100).
-            offset: Number of results to skip for pagination (default: 0).
-            
-        Returns:
-            List of memories matching the criteria.
-        """
-        return self._client.list_memories(
-            bank_id=bank_id,
-            type=type,
-            search_query=search_query,
-            limit=limit,
-            offset=offset,
-        )
-
-
-class HindsightClient(Hindsight):
-    """
-    Enhanced Hindsight client with organized API namespaces.
-
-    This wrapper extends the auto-generated Hindsight client with organized
-    access to different parts of the API through namespaces.
-
-    Example:
-        ```python
-        from hindsight import HindsightClient
-
-        client = HindsightClient(base_url="http://localhost:8888")
-
-        # Core operations (inherited from Hindsight)
-        client.retain(bank_id="test", content="Hello")
-        results = client.recall(bank_id="test", query="Hello")
-
-        # Organized API access through namespaces
-        client.banks.create(bank_id="test", name="Test Bank")
-        models = client.mental_models.list(bank_id="test")
-        directives = client.directives.list(bank_id="test")
-        memories = client.memories.list(bank_id="test")
-        ```
-        
-    Attributes:
-        banks: Namespace for bank management operations.
-        mental_models: Namespace for mental model operations.
-        directives: Namespace for directive operations.
-        memories: Namespace for memory listing operations.
-    """
-
-    def __init__(self, *args: Any, **kwargs: Any) -> None:
-        super().__init__(*args, **kwargs)
-        self._banks_namespace: BanksAPI | None = None
-        self._mental_models_namespace: MentalModelsAPI | None = None
-        self._directives_namespace: DirectivesAPI | None = None
-        self._memories_namespace: MemoriesAPI | None = None
-
-    @property
-    def banks(self) -> BanksAPI:
-        """Access bank management operations.
-        
-        Returns:
-            BanksAPI instance for bank operations.
-        """
-        if self._banks_namespace is None:
-            self._banks_namespace = BanksAPI(self)
-        return self._banks_namespace
-
-    @property
-    def mental_models(self) -> MentalModelsAPI:
-        """Access mental model operations.
-        
-        Returns:
-            MentalModelsAPI instance for mental model operations.
-        """
-        if self._mental_models_namespace is None:
-            self._mental_models_namespace = MentalModelsAPI(self)
-        return self._mental_models_namespace
-
-    @property
-    def directives(self) -> DirectivesAPI:
-        """Access directive operations.
-        
-        Returns:
-            DirectivesAPI instance for directive operations.
-        """
-        if self._directives_namespace is None:
-            self._directives_namespace = DirectivesAPI(self)
-        return self._directives_namespace
-
-    @property
-    def memories(self) -> MemoriesAPI:
-        """Access memory listing operations.
-        
-        Returns:
-            MemoriesAPI instance for memory operations.
-        """
-        if self._memories_namespace is None:
-            self._memories_namespace = MemoriesAPI(self)
-        return self._memories_namespace

hindsight-api-slim/README.md

@@ -1,137 +0,0 @@
-# Hindsight API
-
-**Memory System for AI Agents** — Temporal + Semantic + Entity Memory Architecture using PostgreSQL with pgvector.
-
-Hindsight gives AI agents persistent memory that works like human memory: it stores facts, tracks entities and relationships, handles temporal reasoning ("what happened last spring?"), and forms opinions based on configurable disposition traits.
-
-## Installation
-
-```bash
-pip install hindsight-api
-```
-
-## Quick Start
-
-### Run the Server
-
-```bash
-# Set your LLM provider
-export HINDSIGHT_API_LLM_PROVIDER=openai
-export HINDSIGHT_API_LLM_API_KEY=sk-xxxxxxxxxxxx
-
-# Start the server (uses embedded PostgreSQL by default)
-hindsight-api
-```
-
-The server starts at http://localhost:8888 with:
-- REST API for memory operations
-- MCP server at `/mcp` for tool-use integration
-
-### Use the Python API
-
-```python
-from hindsight_api import MemoryEngine
-
-# Create and initialize the memory engine
-memory = MemoryEngine()
-await memory.initialize()
-
-# Create a memory bank for your agent
-bank = await memory.create_memory_bank(
-    name="my-assistant",
-    background="A helpful coding assistant"
-)
-
-# Store a memory
-await memory.retain(
-    memory_bank_id=bank.id,
-    content="The user prefers Python for data science projects"
-)
-
-# Recall memories
-results = await memory.recall(
-    memory_bank_id=bank.id,
-    query="What programming language does the user prefer?"
-)
-
-# Reflect with reasoning
-response = await memory.reflect(
-    memory_bank_id=bank.id,
-    query="Should I recommend Python or R for this ML project?"
-)
-```
-
-## CLI Options
-
-```bash
-hindsight-api --help
-
-# Common options
-hindsight-api --port 9000          # Custom port (default: 8888)
-hindsight-api --host 127.0.0.1     # Bind to localhost only
-hindsight-api --workers 4          # Multiple worker processes
-hindsight-api --log-level debug    # Verbose logging
-```
-
-## Configuration
-
-Configure via environment variables:
-
-| Variable | Description | Default |
-|----------|-------------|---------|
-| `HINDSIGHT_API_DATABASE_URL` | PostgreSQL connection string | `pg0` (embedded) |
-| `HINDSIGHT_API_LLM_PROVIDER` | `openai`, `anthropic`, `gemini`, `groq`, `ollama`, `lmstudio` | `openai` |
-| `HINDSIGHT_API_LLM_API_KEY` | API key for LLM provider | - |
-| `HINDSIGHT_API_LLM_MODEL` | Model name | `gpt-4o-mini` |
-| `HINDSIGHT_API_HOST` | Server bind address | `0.0.0.0` |
-| `HINDSIGHT_API_PORT` | Server port | `8888` |
-
-### Example with External PostgreSQL
-
-```bash
-export HINDSIGHT_API_DATABASE_URL=postgresql://user:pass@localhost:5432/hindsight
-export HINDSIGHT_API_LLM_PROVIDER=groq
-export HINDSIGHT_API_LLM_API_KEY=gsk_xxxxxxxxxxxx
-
-hindsight-api
-```
-
-## Docker
-
-```bash
-docker run --rm -it -p 8888:8888 \
-  -e HINDSIGHT_API_LLM_API_KEY=$OPENAI_API_KEY \
-  -v $HOME/.hindsight-docker:/home/hindsight/.pg0 \
-  ghcr.io/vectorize-io/hindsight:latest
-```
-
-## MCP Server
-
-For local MCP integration without running the full API server:
-
-```bash
-hindsight-local-mcp
-```
-
-This runs a stdio-based MCP server that can be used directly with MCP-compatible clients.
-
-## Key Features
-
-- **Multi-Strategy Retrieval (TEMPR)** — Semantic, keyword, graph, and temporal search combined with RRF fusion
-- **Entity Graph** — Automatic entity extraction and relationship tracking
-- **Temporal Reasoning** — Native support for time-based queries
-- **Disposition Traits** — Configurable skepticism, literalism, and empathy influence opinion formation
-- **Three Memory Types** — World facts, bank actions, and formed opinions with confidence scores
-
-## Documentation
-
-Full documentation: [https://hindsight.vectorize.io](https://hindsight.vectorize.io)
-
-- [Installation Guide](https://hindsight.vectorize.io/developer/installation)
-- [Configuration Reference](https://hindsight.vectorize.io/developer/configuration)
-- [API Reference](https://hindsight.vectorize.io/api-reference)
-- [Python SDK](https://hindsight.vectorize.io/sdks/python)
-
-## License
-
-Apache 2.0

hindsight-api-slim/hindsight_api/alembic/versions/a1b2c3d4e5f6_add_file_storage_table.py

@@ -1,70 +0,0 @@
-"""Add file_storage table for BYTEA-based file storage
-
-Revision ID: a1b2c3d4e5f6
-Revises: y0t1u2v3w4x5
-Create Date: 2026-02-16
-
-Creates a dedicated table for storing uploaded files using BYTEA.
-This provides zero-config file storage that "just works" for development
-and small deployments. For production/scale, use S3-compatible storage.
-
-Files are stored in a separate table to avoid bloating the documents table.
-"""
-
-from collections.abc import Sequence
-
-from alembic import context, op
-
-revision: str = "a1b2c3d4e5f6"
-down_revision: str | Sequence[str] | None = "y0t1u2v3w4x5"
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-
-def _get_schema_prefix() -> str:
-    """Get schema prefix for table names (required for multi-tenant support)."""
-    schema = context.config.get_main_option("target_schema")
-    return f'"{schema}".' if schema else ""
-
-
-def upgrade() -> None:
-    """Create file_storage table for BYTEA storage."""
-    schema = _get_schema_prefix()
-
-    # Create file_storage table (minimal: just key + data)
-    op.execute(
-        f"""
-        CREATE TABLE IF NOT EXISTS {schema}file_storage (
-            storage_key TEXT PRIMARY KEY,
-            data BYTEA NOT NULL
-        )
-    """
-    )
-
-    # Add file tracking columns to documents table
-    op.execute(
-        f"""
-        ALTER TABLE {schema}documents
-        ADD COLUMN IF NOT EXISTS file_storage_key TEXT,
-        ADD COLUMN IF NOT EXISTS file_original_name TEXT,
-        ADD COLUMN IF NOT EXISTS file_content_type TEXT
-    """
-    )
-
-
-def downgrade() -> None:
-    """Remove file_storage table and related columns."""
-    schema = _get_schema_prefix()
-
-    # Drop columns from documents table
-    op.execute(
-        f"""
-        ALTER TABLE {schema}documents
-        DROP COLUMN IF EXISTS file_storage_key,
-        DROP COLUMN IF EXISTS file_original_name,
-        DROP COLUMN IF EXISTS file_content_type
-    """
-    )
-
-    # Drop file_storage table
-    op.execute(f"DROP TABLE IF EXISTS {schema}file_storage")

hindsight-api-slim/hindsight_api/alembic/versions/a2b3c4d5e6f7_add_text_signals_column.py

@@ -1,88 +0,0 @@
-"""Add text_signals column to memory_units for enriched BM25 indexing.
-
-text_signals stores a denormalized space-separated string of entity names
-(and future signals) to improve full-text search recall without polluting
-the stored fact text.
-
-- vchord: text_signals included in tokenize() at insert time
-- native: search_vector GENERATED column regenerated to include text_signals
-- pg_textsearch: no change (index only supports a single base column)
-
-Revision ID: a2b3c4d5e6f7
-Revises: z1u2v3w4x5y6
-Create Date: 2026-02-28
-"""
-
-import os
-from collections.abc import Sequence
-
-from alembic import context, op
-
-revision: str = "a2b3c4d5e6f7"
-down_revision: str | Sequence[str] | None = "aa2b3c4d5e6f"
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-
-def _get_schema_prefix() -> str:
-    schema = context.config.get_main_option("target_schema")
-    return f'"{schema}".' if schema else ""
-
-
-def _detect_text_search_extension() -> str:
-    return os.getenv("HINDSIGHT_API_TEXT_SEARCH_EXTENSION", "native").lower()
-
-
-def upgrade() -> None:
-    schema = _get_schema_prefix()
-    table = f"{schema}memory_units"
-    text_search_ext = _detect_text_search_extension()
-
-    # Add text_signals column (nullable TEXT, populated at retain time)
-    op.execute(f"ALTER TABLE {table} ADD COLUMN IF NOT EXISTS text_signals TEXT")
-
-    if text_search_ext == "native":
-        # Native PostgreSQL: drop and recreate the GENERATED tsvector column to include text_signals
-        op.execute(f"ALTER TABLE {table} DROP COLUMN IF EXISTS search_vector")
-        op.execute(f"""
-            ALTER TABLE {table}
-            ADD COLUMN search_vector tsvector
-            GENERATED ALWAYS AS (
-                to_tsvector('english',
-                    COALESCE(text, '') || ' ' ||
-                    COALESCE(context, '') || ' ' ||
-                    COALESCE(text_signals, '')
-                )
-            ) STORED
-        """)
-        # Recreate GIN index (was dropped with the column)
-        op.execute(f"""
-            CREATE INDEX IF NOT EXISTS idx_memory_units_text_search
-            ON {table} USING gin(search_vector)
-        """)
-
-    # vchord: tokenize() call in fact_storage.py is updated to include text_signals at insert time
-    # pg_textsearch: no change — index operates on the base `text` column only
-
-
-def downgrade() -> None:
-    schema = _get_schema_prefix()
-    table = f"{schema}memory_units"
-    text_search_ext = _detect_text_search_extension()
-
-    if text_search_ext == "native":
-        op.execute(f"DROP INDEX IF EXISTS {schema}idx_memory_units_text_search")
-        op.execute(f"ALTER TABLE {table} DROP COLUMN IF EXISTS search_vector")
-        op.execute(f"""
-            ALTER TABLE {table}
-            ADD COLUMN search_vector tsvector
-            GENERATED ALWAYS AS (
-                to_tsvector('english', COALESCE(text, '') || ' ' || COALESCE(context, ''))
-            ) STORED
-        """)
-        op.execute(f"""
-            CREATE INDEX idx_memory_units_text_search
-            ON {table} USING gin(search_vector)
-        """)
-
-    op.execute(f"ALTER TABLE {table} DROP COLUMN IF EXISTS text_signals")

hindsight-api-slim/hindsight_api/alembic/versions/a2b3c4d5e6f8_add_gin_index_source_memory_ids.py

@@ -1,54 +0,0 @@
-"""Add GIN index on source_memory_ids for observation lookup performance
-
-Without this index, queries using the array overlap operator (&&) or array
-containment (@>) on source_memory_ids require a full sequential scan over all
-observation memory_units. At ~77k observations this was measured at 45ms per
-query, becoming a bottleneck during consolidation recall (57-64s timeouts) and
-user recall (18-27s average).
-
-The GIN index reduces these queries to index scans: 45ms → 0.049ms (927x
-speedup). Recall dropped from 18-27s to ~6s, and consolidation recall
-stabilised from timeout to ~15s.
-
-Created with CONCURRENTLY so the migration does not block reads or writes.
-CONCURRENTLY requires running outside a transaction block, so the migration
-emits an explicit COMMIT before the statement and uses IF NOT EXISTS for
-idempotency.
-
-Revision ID: a2b3c4d5e6f8
-Revises: f7g8h9i0j1k2
-Create Date: 2026-03-04
-"""
-
-from collections.abc import Sequence
-
-from alembic import context, op
-
-revision: str = "a2b3c4d5e6f8"
-down_revision: str | Sequence[str] | None = "f7g8h9i0j1k2"
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-
-def _get_schema_prefix() -> str:
-    schema = context.config.get_main_option("target_schema")
-    return f'"{schema}".' if schema else ""
-
-
-def upgrade() -> None:
-    schema = _get_schema_prefix()
-
-    # CREATE INDEX CONCURRENTLY cannot run inside a transaction block.
-    # Commit the current Alembic transaction first.
-    op.execute("COMMIT")
-    op.execute(
-        f"CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_memory_units_source_memory_ids "
-        f"ON {schema}memory_units USING GIN (source_memory_ids) "
-        f"WHERE source_memory_ids IS NOT NULL"
-    )
-
-
-def downgrade() -> None:
-    schema = _get_schema_prefix()
-    op.execute("COMMIT")
-    op.execute(f"DROP INDEX CONCURRENTLY IF EXISTS {schema}idx_memory_units_source_memory_ids")

hindsight-api-slim/hindsight_api/alembic/versions/a3b4c5d6e7f8_add_consolidation_failed_at_to_memory_units.py

@@ -1,52 +0,0 @@
-"""Add consolidation_failed_at column to memory_units for tracking persistent LLM failures.
-
-When all LLM retries are exhausted on a single-memory batch, the memory is marked
-with consolidation_failed_at instead of consolidated_at, so it is not silently lost
-and can be retried later via the API.
-
-Revision ID: a3b4c5d6e7f8
-Revises: g7h8i9j0k1l2
-Create Date: 2026-03-17
-"""
-
-from collections.abc import Sequence
-
-from alembic import context, op
-
-revision: str = "a3b4c5d6e7f8"
-down_revision: str | Sequence[str] | None = "g7h8i9j0k1l2"
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-
-def _get_schema_prefix() -> str:
-    """Get schema prefix for table names (required for multi-tenant support)."""
-    schema = context.config.get_main_option("target_schema")
-    return f'"{schema}".' if schema else ""
-
-
-def upgrade() -> None:
-    schema = _get_schema_prefix()
-
-    op.execute(
-        f"""
-        ALTER TABLE {schema}memory_units
-        ADD COLUMN IF NOT EXISTS consolidation_failed_at TIMESTAMPTZ DEFAULT NULL
-        """
-    )
-
-    # Index to efficiently query memories that failed consolidation for a given bank
-    op.execute(
-        f"""
-        CREATE INDEX IF NOT EXISTS idx_memory_units_consolidation_failed
-        ON {schema}memory_units (bank_id, consolidation_failed_at)
-        WHERE consolidation_failed_at IS NOT NULL AND fact_type IN ('experience', 'world')
-        """
-    )
-
-
-def downgrade() -> None:
-    schema = _get_schema_prefix()
-
-    op.execute(f"DROP INDEX IF EXISTS {schema}idx_memory_units_consolidation_failed")
-    op.execute(f"ALTER TABLE {schema}memory_units DROP COLUMN IF EXISTS consolidation_failed_at")

hindsight-api-slim/hindsight_api/alembic/versions/aa2b3c4d5e6f_nullable_event_date.py

@@ -1,36 +0,0 @@
-"""Make event_date nullable in memory_units to support timestamp-free content
-
-Revision ID: aa2b3c4d5e6f
-Revises: z1u2v3w4x5y6
-Create Date: 2026-03-02
-
-When callers retain content without a timestamp (e.g. fictional documents, static text),
-the event_date column should be allowed to be NULL rather than defaulting to utcnow().
-"""
-
-from collections.abc import Sequence
-
-from alembic import context, op
-
-revision: str = "aa2b3c4d5e6f"
-down_revision: str | Sequence[str] | None = "z1u2v3w4x5y6"
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-
-def _get_schema_prefix() -> str:
-    """Get schema prefix for table names (required for multi-tenant support)."""
-    schema = context.config.get_main_option("target_schema")
-    return f'"{schema}".' if schema else ""
-
-
-def upgrade() -> None:
-    schema = _get_schema_prefix()
-    op.execute(f"ALTER TABLE {schema}memory_units ALTER COLUMN event_date DROP NOT NULL")
-
-
-def downgrade() -> None:
-    schema = _get_schema_prefix()
-    # Backfill NULLs with now() before restoring the NOT NULL constraint
-    op.execute(f"UPDATE {schema}memory_units SET event_date = now() WHERE event_date IS NULL")
-    op.execute(f"ALTER TABLE {schema}memory_units ALTER COLUMN event_date SET NOT NULL")

hindsight-api-slim/hindsight_api/alembic/versions/b3c4d5e6f7g8_add_temporal_date_indexes.py

@@ -1,68 +0,0 @@
-"""Add partial indexes on memory_units temporal date fields for fast temporal retrieval
-
-Revision ID: b3c4d5e6f7g8
-Revises: c1a2b3d4e5f6
-Create Date: 2026-03-02
-
-The temporal retrieval entry-point query filters memory_units by occurred_start,
-occurred_end, and mentioned_at using OR conditions. Without dedicated indexes the
-planner falls back to a sequential scan of all bank rows after applying the
-(bank_id, fact_type) index, then re-checks each date field.
-
-These three partial indexes give the planner bitmap-index scan options for the
-three most common date predicates, dramatically reducing the row set before any
-embedding computation is required.
-
-All indexes are created CONCURRENTLY so the migration does not block writes on
-memory_units during production deployments. CONCURRENTLY requires running outside
-a transaction block; see migrations.py for how this is handled safely.
-"""
-
-from collections.abc import Sequence
-
-from alembic import context, op
-
-revision: str = "b3c4d5e6f7g8"
-down_revision: str | Sequence[str] | None = "c1a2b3d4e5f6"
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-
-def _get_schema_prefix() -> str:
-    schema = context.config.get_main_option("target_schema")
-    return f'"{schema}".' if schema else ""
-
-
-def upgrade() -> None:
-    schema = _get_schema_prefix()
-    # Partial index on occurred_start (covers "occurred_start BETWEEN $4 AND $5")
-    op.execute("COMMIT")
-    op.execute(
-        f"CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_memory_units_bank_occurred_start "
-        f"ON {schema}memory_units(bank_id, fact_type, occurred_start) "
-        f"WHERE occurred_start IS NOT NULL"
-    )
-    # Partial index on occurred_end (covers "occurred_end BETWEEN $4 AND $5")
-    op.execute("COMMIT")
-    op.execute(
-        f"CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_memory_units_bank_occurred_end "
-        f"ON {schema}memory_units(bank_id, fact_type, occurred_end) "
-        f"WHERE occurred_end IS NOT NULL"
-    )
-    # Partial index on mentioned_at (covers "mentioned_at BETWEEN $4 AND $5")
-    op.execute("COMMIT")
-    op.execute(
-        f"CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_memory_units_bank_mentioned_at "
-        f"ON {schema}memory_units(bank_id, fact_type, mentioned_at) "
-        f"WHERE mentioned_at IS NOT NULL"
-    )
-
-
-def downgrade() -> None:
-    schema = _get_schema_prefix()
-    op.execute("COMMIT")
-    op.execute(f"DROP INDEX CONCURRENTLY IF EXISTS {schema}idx_memory_units_bank_mentioned_at")
-    op.execute("COMMIT")
-    op.execute(f"DROP INDEX CONCURRENTLY IF EXISTS {schema}idx_memory_units_bank_occurred_end")
-    op.execute("COMMIT")
-    op.execute(f"DROP INDEX CONCURRENTLY IF EXISTS {schema}idx_memory_units_bank_occurred_start")

hindsight-api-slim/hindsight_api/alembic/versions/b4c5d6e7f8a9_backfill_observation_scopes.py

@@ -1,34 +0,0 @@
-"""Backfill observation_scopes column if missing.
-
-This migration ensures observation_scopes exists even on databases that had
-revision z1u2v3w4x5y6 applied when it referred to the old text_signals migration
-(before it was renamed to a2b3c4d5e6f7). The ADD COLUMN IF NOT EXISTS makes this
-a no-op on databases that already have the column.
-
-Revision ID: b4c5d6e7f8a9
-Revises: a2b3c4d5e6f7
-Create Date: 2026-03-02
-"""
-
-from collections.abc import Sequence
-
-from alembic import context, op
-
-revision: str = "b4c5d6e7f8a9"
-down_revision: str | Sequence[str] | None = "a2b3c4d5e6f7"
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-
-def _get_schema_prefix() -> str:
-    schema = context.config.get_main_option("target_schema")
-    return f'"{schema}".' if schema else ""
-
-
-def upgrade() -> None:
-    schema = _get_schema_prefix()
-    op.execute(f"ALTER TABLE {schema}memory_units ADD COLUMN IF NOT EXISTS observation_scopes JSONB")
-
-
-def downgrade() -> None:
-    pass  # intentionally no-op — safe to leave the column in place

hindsight-api-slim/hindsight_api/alembic/versions/c1a2b3d4e5f6_enable_pg_trgm_and_entities_trgm_index.py

@@ -1,59 +0,0 @@
-"""Enable pg_trgm extension and add GIN trigram index on entities.canonical_name
-
-Revision ID: c1a2b3d4e5f6
-Revises: b4c5d6e7f8a9
-Create Date: 2026-03-02
-
-Index is created CONCURRENTLY so the migration does not block writes on entities
-during production deployments. CONCURRENTLY requires running outside a transaction
-block; see migrations.py for how this is handled safely.
-"""
-
-from collections.abc import Sequence
-
-import sqlalchemy as sa
-from alembic import context, op
-
-revision: str = "c1a2b3d4e5f6"
-down_revision: str | Sequence[str] | None = "b4c5d6e7f8a9"
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-
-def _get_schema_prefix() -> str:
-    schema = context.config.get_main_option("target_schema")
-    return f'"{schema}".' if schema else ""
-
-
-def upgrade() -> None:
-    # pg_trgm ships with most PostgreSQL installations as a contrib module.
-    # It enables fast similarity lookups via GIN indexes, used for entity name matching.
-    # On managed services (e.g. Azure Flexible Server), the extension may not be
-    # available or may require manual enablement.  We gracefully skip the index
-    # creation if the extension cannot be loaded — the entity resolver will
-    # auto-detect and fall back to the "full" lookup strategy at runtime.  See #626.
-    conn = op.get_bind()
-    try:
-        conn.execute(sa.text("CREATE EXTENSION IF NOT EXISTS pg_trgm"))
-    except Exception:
-        # Extension not available (managed Postgres, insufficient privileges, etc.)
-        # Roll back the failed statement and skip index creation.
-        conn.execute(sa.text("ROLLBACK"))
-        conn.execute(sa.text("BEGIN"))
-        return
-
-    schema = _get_schema_prefix()
-    # GIN index on canonical_name enables sub-millisecond trigram similarity queries
-    # (% operator, similarity()) instead of full-table scans across all bank entities.
-    op.execute("COMMIT")
-    op.execute(
-        f"CREATE INDEX CONCURRENTLY IF NOT EXISTS entities_canonical_name_trgm_idx "
-        f"ON {schema}entities USING GIN (canonical_name gin_trgm_ops)"
-    )
-
-
-def downgrade() -> None:
-    schema = _get_schema_prefix()
-    op.execute("COMMIT")
-    op.execute(f"DROP INDEX CONCURRENTLY IF EXISTS {schema}entities_canonical_name_trgm_idx")
-    # Note: not dropping pg_trgm extension as other indexes may depend on it

hindsight-api-slim/hindsight_api/alembic/versions/c3d4e5f6g7h8_add_history_to_mental_models.py

@@ -1,30 +0,0 @@
-"""Add history column to mental_models
-
-Revision ID: c3d4e5f6g7h8
-Revises: a2b3c4d5e6f7, a2b3c4d5e6f8
-Create Date: 2026-03-06
-"""
-
-from collections.abc import Sequence
-
-from alembic import context, op
-
-revision: str = "c3d4e5f6g7h8"
-down_revision: str | Sequence[str] | None = ("a2b3c4d5e6f7", "a2b3c4d5e6f8")
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-
-def _get_schema_prefix() -> str:
-    schema = context.config.get_main_option("target_schema")
-    return f'"{schema}".' if schema else ""
-
-
-def upgrade() -> None:
-    schema = _get_schema_prefix()
-    op.execute(f"ALTER TABLE {schema}mental_models ADD COLUMN IF NOT EXISTS history JSONB DEFAULT '[]'::jsonb")
-
-
-def downgrade() -> None:
-    schema = _get_schema_prefix()
-    op.execute(f"ALTER TABLE {schema}mental_models DROP COLUMN IF EXISTS history")

hindsight-api-slim/hindsight_api/alembic/versions/d2e3f4a5b6c7_add_memory_links_expansion_indexes.py

@@ -1,83 +0,0 @@
-"""Add covering and composite indexes to speed up link expansion graph retrieval.
-
-Two indexes target the two bottlenecks identified by EXPLAIN ANALYZE on a 17M-row
-memory_links table:
-
-1. idx_memory_links_to_type_weight  (to_unit_id, link_type, weight DESC)
-   The semantic incoming direction — finding facts that consider seeds as their
-   nearest neighbour — currently hits an expensive BitmapAnd of two separate
-   bitmap scans (to_unit_id bitmap ∩ link_type bitmap).  A composite index
-   on (to_unit_id, link_type) turns this into a single index scan and reduces
-   latency from ~36 ms to < 5 ms per query.
-
-2. idx_memory_links_entity_covering  (from_unit_id) INCLUDE (to_unit_id, entity_id)
-   WHERE link_type = 'entity'
-   The entity co-occurrence expansion uses COUNT(DISTINCT ml.entity_id) and
-   joins on ml.to_unit_id.  Without a covering index the planner must read
-   ~2 500 heap pages to fetch entity_id and to_unit_id after the bitmap index
-   scan, adding ~230 ms of random I/O.  INCLUDE adds those two columns to the
-   index leaf pages so the entire query can be served from the index (index-only
-   scan), eliminating the heap reads entirely.
-   Partial index (WHERE link_type = 'entity') keeps index size ~40 % smaller.
-
-Both indexes are created with CONCURRENTLY so the migration does not block
-concurrent reads or writes on memory_links.  CONCURRENTLY requires running
-outside a transaction block, so the migration emits an explicit COMMIT before
-each statement and uses IF NOT EXISTS for idempotency.
-
-Revision ID: d2e3f4a5b6c7
-Revises: b3c4d5e6f7g8
-Create Date: 2026-03-02
-"""
-
-from collections.abc import Sequence
-
-from alembic import context, op
-
-revision: str = "d2e3f4a5b6c7"
-down_revision: str | Sequence[str] | None = "b3c4d5e6f7g8"
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-
-def _get_schema_prefix() -> str:
-    schema = context.config.get_main_option("target_schema")
-    return f'"{schema}".' if schema else ""
-
-
-def upgrade() -> None:
-    schema = _get_schema_prefix()
-
-    # CREATE INDEX CONCURRENTLY cannot run inside a transaction block.
-    # Commit the current Alembic transaction, then issue each CONCURRENTLY
-    # statement in its own implicit autocommit transaction.
-    # IF NOT EXISTS makes each statement idempotent if the migration is retried.
-
-    # Index for the semantic *incoming* direction in link_expansion_retrieval.py.
-    # Replaces the BitmapAnd of idx_memory_links_to_unit ∩ idx_memory_links_link_type
-    # with a single composite index scan.
-    op.execute("COMMIT")
-    op.execute(
-        f"CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_memory_links_to_type_weight "
-        f"ON {schema}memory_links(to_unit_id, link_type, weight DESC)"
-    )
-
-    # Covering index for entity co-occurrence expansion.
-    # Enables an index-only scan: entity_id and to_unit_id are read from the
-    # index leaf pages instead of the heap, eliminating ~2 500 random heap-page
-    # reads per expansion query.
-    op.execute("COMMIT")
-    op.execute(
-        f"CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_memory_links_entity_covering "
-        f"ON {schema}memory_links(from_unit_id) "
-        f"INCLUDE (to_unit_id, entity_id) "
-        f"WHERE link_type = 'entity'"
-    )
-
-
-def downgrade() -> None:
-    schema = _get_schema_prefix()
-    op.execute("COMMIT")
-    op.execute(f"DROP INDEX CONCURRENTLY IF EXISTS {schema}idx_memory_links_entity_covering")
-    op.execute("COMMIT")
-    op.execute(f"DROP INDEX CONCURRENTLY IF EXISTS {schema}idx_memory_links_to_type_weight")

hindsight-api-slim/hindsight_api/alembic/versions/d4e5f6g7h8i9_gin_source_memory_ids_fastupdate_off.py

@@ -1,53 +0,0 @@
-"""Recreate idx_memory_units_source_memory_ids GIN index with fastupdate=off
-
-GIN indexes use a "fastupdate" pending list by default: small writes are
-buffered there and flushed to the main GIN tree in bulk. Flushing requires
-AccessExclusiveLock on the index. Under high insert concurrency (e.g. 8
-parallel pytest-xdist workers all calling retain_async) two transactions can
-each trigger a flush simultaneously and deadlock.
-
-Disabling fastupdate makes every insert write directly to the GIN tree
-(slightly slower per insert, but no pending-list lock cycles).
-
-Revision ID: d4e5f6g7h8i9
-Revises: d5e6f7a8b9c0
-Create Date: 2026-03-11
-"""
-
-from collections.abc import Sequence
-
-from alembic import context, op
-
-revision: str = "d4e5f6g7h8i9"
-down_revision: str | Sequence[str] | None = "d5e6f7a8b9c0"
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-
-def _get_schema_prefix() -> str:
-    schema = context.config.get_main_option("target_schema")
-    return f'"{schema}".' if schema else ""
-
-
-def upgrade() -> None:
-    schema = _get_schema_prefix()
-    # DROP + CREATE CONCURRENTLY must run outside a transaction block.
-    op.execute("COMMIT")
-    op.execute(f"DROP INDEX CONCURRENTLY IF EXISTS {schema}idx_memory_units_source_memory_ids")
-    op.execute(
-        f"CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_memory_units_source_memory_ids "
-        f"ON {schema}memory_units USING GIN (source_memory_ids) "
-        f"WITH (fastupdate=off) "
-        f"WHERE source_memory_ids IS NOT NULL"
-    )
-
-
-def downgrade() -> None:
-    schema = _get_schema_prefix()
-    op.execute("COMMIT")
-    op.execute(f"DROP INDEX CONCURRENTLY IF EXISTS {schema}idx_memory_units_source_memory_ids")
-    op.execute(
-        f"CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_memory_units_source_memory_ids "
-        f"ON {schema}memory_units USING GIN (source_memory_ids) "
-        f"WHERE source_memory_ids IS NOT NULL"
-    )

hindsight-api-slim/hindsight_api/alembic/versions/d5e6f7a8b9c0_add_bank_internal_id_and_per_bank_hnsw.py

@@ -1,131 +0,0 @@
-"""Add internal_id to banks and per-(bank, fact_type) partial HNSW indexes
-
-Revision ID: d5e6f7a8b9c0
-Revises: a3b4c5d6e7f8
-Create Date: 2026-03-11
-
-This migration:
-1. Adds internal_id UUID column to banks (stable identifier for index naming)
-2. Drops the global HNSW index (competes with per-bank partial indexes)
-3. Creates per-(bank_id, fact_type) partial HNSW indexes for all existing banks
-   (new banks get indexes created at bank-creation time via bank_utils.create_bank_hnsw_indexes)
-
-Why per-(bank, fact_type) indexes:
-- fact_type-only partial indexes are never chosen by the planner when bank_id is in the WHERE
-  clause, because the idx_memory_units_bank_id B-tree index always wins at planning time.
-- Per-(bank, fact_type) partial indexes have both predicates matching → planner selects them.
-- The global HNSW index competes for larger partitions (world, observation) and must be dropped.
-
-For large deployments, create indexes CONCURRENTLY before running this migration:
-    SELECT internal_id, bank_id FROM banks;
-    -- for each bank and each fact_type in (world, experience, observation):
-    CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_mu_emb_{ft}_{uid16}
-        ON memory_units USING hnsw (embedding vector_cosine_ops)
-        WHERE fact_type = '{ft}' AND bank_id = '{bank_id}';
-    DROP INDEX CONCURRENTLY IF EXISTS idx_memory_units_embedding;
-"""
-
-from collections.abc import Sequence
-
-from alembic import context, op
-from sqlalchemy import text
-
-revision: str = "d5e6f7a8b9c0"
-down_revision: str | Sequence[str] | None = "c3d4e5f6g7h8"
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-_HNSW_FACT_TYPES: dict[str, str] = {
-    "world": "worl",
-    "experience": "expr",
-    "observation": "obsv",
-}
-
-
-def _get_schema_prefix() -> str:
-    schema = context.config.get_main_option("target_schema")
-    return f'"{schema}".' if schema else ""
-
-
-def upgrade() -> None:
-    schema = _get_schema_prefix()
-
-    # 1. Add internal_id column to banks
-    op.execute(
-        f"ALTER TABLE {schema}banks ADD COLUMN IF NOT EXISTS internal_id UUID DEFAULT gen_random_uuid() NOT NULL"
-    )
-    op.execute(f"ALTER TABLE {schema}banks ADD CONSTRAINT banks_internal_id_unique UNIQUE (internal_id)")
-
-    # 2. Drop any fact_type-only partial HNSW indexes that may exist from prior migrations
-    #    (bank_id B-tree always wins over them when bank_id is in the WHERE clause)
-    op.execute(f"DROP INDEX IF EXISTS {schema}idx_mu_emb_world")
-    op.execute(f"DROP INDEX IF EXISTS {schema}idx_mu_emb_observation")
-    op.execute(f"DROP INDEX IF EXISTS {schema}idx_mu_emb_experience")
-
-    # 4. Drop global HNSW index (competes with per-bank partial indexes)
-    op.execute(f"DROP INDEX IF EXISTS {schema}idx_memory_units_embedding")
-
-    # 5. Create per-(bank, fact_type) partial HNSW indexes for all existing banks
-    bind = op.get_bind()
-    schema_name = context.config.get_main_option("target_schema")
-    table_ref = f'"{schema_name}".memory_units' if schema_name else "memory_units"
-    banks_ref = f'"{schema_name}".banks' if schema_name else "banks"
-
-    rows = bind.execute(text(f"SELECT bank_id, internal_id FROM {banks_ref}")).fetchall()  # noqa: S608
-    for row in rows:
-        bank_id = row[0]
-        internal_id = str(row[1]).replace("-", "")[:16]
-        escaped_bank_id = bank_id.replace("'", "''")
-        for ft, ft_short in _HNSW_FACT_TYPES.items():
-            idx_name = f"idx_mu_emb_{ft_short}_{internal_id}"
-            # Index name is schema-unqualified (indexes live in the schema of their table)
-            bind.execute(
-                text(
-                    f"CREATE INDEX IF NOT EXISTS {idx_name} "
-                    f"ON {table_ref} USING hnsw (embedding vector_cosine_ops) "
-                    f"WHERE fact_type = '{ft}' AND bank_id = '{escaped_bank_id}'"
-                )
-            )
-
-
-def downgrade() -> None:
-    schema = _get_schema_prefix()
-
-    # Drop per-bank HNSW indexes (iterate existing banks)
-    bind = op.get_bind()
-    schema_name = context.config.get_main_option("target_schema")
-    banks_ref = f'"{schema_name}".banks' if schema_name else "banks"
-
-    rows = bind.execute(text(f"SELECT internal_id FROM {banks_ref}")).fetchall()  # noqa: S608
-    for row in rows:
-        internal_id = str(row[0]).replace("-", "")[:16]
-        for ft_short in _HNSW_FACT_TYPES.values():
-            idx_name = f"idx_mu_emb_{ft_short}_{internal_id}"
-            bind.execute(text(f"DROP INDEX IF EXISTS {schema}{idx_name}"))
-
-    # Restore the global HNSW index
-    table_ref = f'"{schema_name}".memory_units' if schema_name else "memory_units"
-    op.execute(
-        f"CREATE INDEX IF NOT EXISTS idx_memory_units_embedding ON {table_ref} USING hnsw (embedding vector_cosine_ops)"
-    )
-
-    # Restore old fact_type-only partial indexes
-    op.execute(
-        f"CREATE INDEX IF NOT EXISTS idx_mu_emb_world "
-        f"ON {table_ref} USING hnsw (embedding vector_cosine_ops) "
-        f"WHERE fact_type = 'world'"
-    )
-    op.execute(
-        f"CREATE INDEX IF NOT EXISTS idx_mu_emb_observation "
-        f"ON {table_ref} USING hnsw (embedding vector_cosine_ops) "
-        f"WHERE fact_type = 'observation'"
-    )
-    op.execute(
-        f"CREATE INDEX IF NOT EXISTS idx_mu_emb_experience "
-        f"ON {table_ref} USING hnsw (embedding vector_cosine_ops) "
-        f"WHERE fact_type = 'experience'"
-    )
-
-    # Drop internal_id column
-    op.execute(f"ALTER TABLE {schema}banks DROP CONSTRAINT IF EXISTS banks_internal_id_unique")
-    op.execute(f"ALTER TABLE {schema}banks DROP COLUMN IF EXISTS internal_id")

hindsight-api-slim/hindsight_api/alembic/versions/e4f5a6b7c8d9_add_webhooks_tables.py

@@ -1,62 +0,0 @@
-"""Add webhooks table and next_retry_at to async_operations.
-
-Webhook deliveries are handled as async_operations tasks (operation_type='webhook_delivery')
-rather than a dedicated webhook_deliveries table.
-
-Revision ID: e4f5a6b7c8d9
-Revises: d2e3f4a5b6c7
-Create Date: 2026-03-04
-"""
-
-from collections.abc import Sequence
-
-from alembic import context, op
-
-revision: str = "e4f5a6b7c8d9"
-down_revision: str | Sequence[str] | None = "d2e3f4a5b6c7"
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-
-def _get_schema_prefix() -> str:
-    schema = context.config.get_main_option("target_schema")
-    return f'"{schema}".' if schema else ""
-
-
-def upgrade() -> None:
-    schema = _get_schema_prefix()
-
-    op.execute(
-        f"""
-        CREATE TABLE IF NOT EXISTS {schema}webhooks (
-            id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
-            bank_id TEXT,
-            url TEXT NOT NULL,
-            secret TEXT,
-            event_types TEXT[] NOT NULL DEFAULT '{{}}',
-            enabled BOOLEAN NOT NULL DEFAULT TRUE,
-            created_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
-            updated_at TIMESTAMPTZ NOT NULL DEFAULT NOW()
-        )
-        """
-    )
-
-    # Index for bank-scoped webhook lookup
-    op.execute(f"CREATE INDEX IF NOT EXISTS idx_webhooks_bank_id ON {schema}webhooks(bank_id)")
-
-    # Add next_retry_at to async_operations for task-owned retry scheduling
-    op.execute(f"ALTER TABLE {schema}async_operations ADD COLUMN IF NOT EXISTS next_retry_at TIMESTAMPTZ NULL")
-
-    # Index for polling: status + next_retry_at
-    op.execute(
-        f"CREATE INDEX IF NOT EXISTS idx_async_operations_status_retry "
-        f"ON {schema}async_operations(status, next_retry_at)"
-    )
-
-
-def downgrade() -> None:
-    schema = _get_schema_prefix()
-    op.execute(f"DROP INDEX IF EXISTS {schema}idx_async_operations_status_retry")
-    op.execute(f"ALTER TABLE {schema}async_operations DROP COLUMN IF EXISTS next_retry_at")
-    op.execute(f"DROP INDEX IF EXISTS {schema}idx_webhooks_bank_id")
-    op.execute(f"DROP TABLE IF EXISTS {schema}webhooks")

hindsight-api-slim/hindsight_api/alembic/versions/e5f6g7h8i9j0_cascade_delete_ops_on_bank_delete.py

@@ -1,73 +0,0 @@
-"""Add CASCADE DELETE FK from async_operations and webhooks to banks.
-
-When a bank is deleted, all its async_operations and webhooks rows are
-automatically deleted by the database. This ensures that any in-flight
-worker tasks detect the deletion via _check_op_alive() and abort early.
-
-Revision ID: e5f6g7h8i9j0
-Revises: d4e5f6g7h8i9
-Create Date: 2026-03-11
-"""
-
-from collections.abc import Sequence
-
-from alembic import context, op
-
-revision: str = "e5f6g7h8i9j0"
-down_revision: str | Sequence[str] | None = "d4e5f6g7h8i9"
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-
-def _get_schema_prefix() -> str:
-    schema = context.config.get_main_option("target_schema")
-    return f'"{schema}".' if schema else ""
-
-
-def upgrade() -> None:
-    schema = _get_schema_prefix()
-
-    # Remove orphaned async_operations rows whose bank no longer exists
-    # (can happen because there was no FK before this migration).
-    op.execute(
-        f"""
-        DELETE FROM {schema}async_operations
-        WHERE bank_id IS NOT NULL
-          AND bank_id NOT IN (SELECT bank_id FROM {schema}banks)
-        """
-    )
-
-    # Remove orphaned webhooks rows whose bank no longer exists.
-    op.execute(
-        f"""
-        DELETE FROM {schema}webhooks
-        WHERE bank_id IS NOT NULL
-          AND bank_id NOT IN (SELECT bank_id FROM {schema}banks)
-        """
-    )
-
-    # Add FK with ON DELETE CASCADE so that deleting a bank automatically
-    # cleans up all its pending/processing operations and webhook configs.
-    op.execute(
-        f"""
-        ALTER TABLE {schema}async_operations
-            ADD CONSTRAINT fk_async_operations_bank_id
-            FOREIGN KEY (bank_id) REFERENCES {schema}banks(bank_id)
-            ON DELETE CASCADE
-        """
-    )
-
-    op.execute(
-        f"""
-        ALTER TABLE {schema}webhooks
-            ADD CONSTRAINT fk_webhooks_bank_id
-            FOREIGN KEY (bank_id) REFERENCES {schema}banks(bank_id)
-            ON DELETE CASCADE
-        """
-    )
-
-
-def downgrade() -> None:
-    schema = _get_schema_prefix()
-    op.execute(f"ALTER TABLE {schema}async_operations DROP CONSTRAINT IF EXISTS fk_async_operations_bank_id")
-    op.execute(f"ALTER TABLE {schema}webhooks DROP CONSTRAINT IF EXISTS fk_webhooks_bank_id")

hindsight-api-slim/hindsight_api/alembic/versions/f6g7h8i9j0k1_chunk_fk_cascade_delete.py

@@ -1,38 +0,0 @@
-"""chunk_fk_cascade_delete
-
-Revision ID: f6g7h8i9j0k1
-Revises: e5f6g7h8i9j0
-Create Date: 2026-03-16 00:00:00.000000
-
-"""
-
-from collections.abc import Sequence
-
-from alembic import op
-
-# revision identifiers, used by Alembic.
-revision: str = "f6g7h8i9j0k1"
-down_revision: str | Sequence[str] | None = "e5f6g7h8i9j0"
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-
-def upgrade() -> None:
-    """Change memory_units.chunk_id FK from SET NULL to CASCADE.
-
-    When a document is deleted the CASCADE reaches chunks first; with SET NULL
-    the memory_units rows survived with chunk_id = NULL, leaving ghost records.
-    Switching to CASCADE ensures they are removed together with their chunk.
-    """
-    op.drop_constraint("memory_units_chunk_fkey", "memory_units", type_="foreignkey")
-    op.create_foreign_key(
-        "memory_units_chunk_fkey", "memory_units", "chunks", ["chunk_id"], ["chunk_id"], ondelete="CASCADE"
-    )
-
-
-def downgrade() -> None:
-    """Revert to SET NULL behaviour."""
-    op.drop_constraint("memory_units_chunk_fkey", "memory_units", type_="foreignkey")
-    op.create_foreign_key(
-        "memory_units_chunk_fkey", "memory_units", "chunks", ["chunk_id"], ["chunk_id"], ondelete="SET NULL"
-    )

hindsight-api-slim/hindsight_api/alembic/versions/f7g8h9i0j1k2_add_webhook_http_config.py

@@ -1,33 +0,0 @@
-"""Add http_config JSONB column to webhooks table.
-
-Stores HTTP delivery configuration (method, timeout, headers, params) as a
-single JSONB column rather than separate columns.
-
-Revision ID: f7g8h9i0j1k2
-Revises: e4f5a6b7c8d9
-Create Date: 2026-03-04
-"""
-
-from collections.abc import Sequence
-
-from alembic import context, op
-
-revision: str = "f7g8h9i0j1k2"
-down_revision: str | Sequence[str] | None = "e4f5a6b7c8d9"
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-
-def _get_schema_prefix() -> str:
-    schema = context.config.get_main_option("target_schema")
-    return f'"{schema}".' if schema else ""
-
-
-def upgrade() -> None:
-    schema = _get_schema_prefix()
-    op.execute(f"ALTER TABLE {schema}webhooks ADD COLUMN IF NOT EXISTS http_config JSONB NOT NULL DEFAULT '{{}}'")
-
-
-def downgrade() -> None:
-    schema = _get_schema_prefix()
-    op.execute(f"ALTER TABLE {schema}webhooks DROP COLUMN IF EXISTS http_config")

hindsight-api-slim/hindsight_api/alembic/versions/g7h8i9j0k1l2_backsweep_orphan_observations.py

@@ -1,71 +0,0 @@
-"""backsweep_orphan_memory_units
-
-Two-pass cleanup of memory_units rows that were never removed by earlier bugs:
-
-Pass 1 — any fact_type, bank gone:
-  memory_units whose bank_id no longer exists in banks.  These accumulate when
-  a bank is deleted without a proper cascade (no FK from memory_units to banks
-  exists in the schema).
-
-Pass 2 — observations only, all sources gone:
-  observation rows whose bank still exists but every source_memory_id points
-  to a deleted memory unit.  These were left behind before PR #580 fixed the
-  chunk FK cascade and before delete_document() called
-  _delete_stale_observations_for_memories.
-
-Revision ID: g7h8i9j0k1l2
-Revises: f6g7h8i9j0k1
-Create Date: 2026-03-16
-"""
-
-from collections.abc import Sequence
-
-from alembic import context, op
-
-revision: str = "g7h8i9j0k1l2"
-down_revision: str | Sequence[str] | None = "f6g7h8i9j0k1"
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-
-def _get_schema_prefix() -> str:
-    schema = context.config.get_main_option("target_schema")
-    return f'"{schema}".' if schema else ""
-
-
-def upgrade() -> None:
-    schema = _get_schema_prefix()
-    mu = f"{schema}memory_units"
-    banks = f"{schema}banks"
-
-    # Pass 1: delete all memory_units (any fact_type) whose bank no longer exists.
-    # There is no FK from memory_units to banks, so these never cascade away.
-    op.execute(
-        f"""
-        DELETE FROM {mu}
-        WHERE NOT EXISTS (
-            SELECT 1 FROM {banks} b WHERE b.bank_id = {mu}.bank_id
-        )
-        """
-    )
-
-    # Pass 2: delete orphaned observations whose bank still exists but every
-    # source_memory_id refers to a now-deleted memory unit (or the array is
-    # empty).  Observations with at least one surviving source are left alone.
-    op.execute(
-        f"""
-        DELETE FROM {mu} orphan
-        WHERE orphan.fact_type = 'observation'
-          AND NOT EXISTS (
-              SELECT 1
-              FROM {mu} src
-              WHERE src.id = ANY(orphan.source_memory_ids)
-                AND src.bank_id = orphan.bank_id
-          )
-        """
-    )
-
-
-def downgrade() -> None:
-    # Deleted rows cannot be restored.
-    pass

hindsight-api-slim/hindsight_api/alembic/versions/y0t1u2v3w4x5_add_result_metadata_gin_index.py

@@ -1,49 +0,0 @@
-"""Add GIN index on async_operations.result_metadata for parent_operation_id queries
-
-Revision ID: y0t1u2v3w4x5
-Revises: x9s0t1u2v3w4
-Create Date: 2026-02-13
-
-This migration adds a GIN index on the result_metadata JSONB column in the
-async_operations table to support efficient queries for child operations by
-parent_operation_id.
-
-The index enables fast lookups when querying for child operations:
-  SELECT * FROM async_operations
-  WHERE result_metadata::jsonb @> '{"parent_operation_id": "uuid"}'::jsonb
-"""
-
-from collections.abc import Sequence
-
-from alembic import context, op
-
-revision: str = "y0t1u2v3w4x5"
-down_revision: str | Sequence[str] | None = "x9s0t1u2v3w4"
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-
-def _get_schema_prefix() -> str:
-    """Get schema prefix for table names (required for multi-tenant support)."""
-    schema = context.config.get_main_option("target_schema")
-    return f'"{schema}".' if schema else ""
-
-
-def upgrade() -> None:
-    """Add GIN index on result_metadata for efficient parent_operation_id queries."""
-    schema = _get_schema_prefix()
-
-    # Add GIN index for JSONB containment queries (@> operator)
-    op.execute(f"""
-        CREATE INDEX IF NOT EXISTS idx_async_operations_result_metadata
-        ON {schema}async_operations
-        USING gin(result_metadata)
-    """)
-
-
-def downgrade() -> None:
-    """Remove GIN index on result_metadata."""
-    schema = _get_schema_prefix()
-
-    # Drop index
-    op.execute(f"DROP INDEX IF EXISTS {schema}idx_async_operations_result_metadata")

hindsight-api-slim/hindsight_api/alembic/versions/z1u2v3w4x5y6_add_observation_tags_to_memory_units.py

@@ -1,35 +0,0 @@
-"""Add observation_scopes column to memory_units table
-
-Revision ID: z1u2v3w4x5y6
-Revises: a1b2c3d4e5f6
-Create Date: 2026-02-25
-
-Adds observation_scopes JSONB column to memory_units to control how observations
-are scoped during consolidation. Accepts "per_tag", "combined", or an explicit
-list of tag-set lists for custom multi-pass consolidation.
-"""
-
-from collections.abc import Sequence
-
-from alembic import context, op
-
-revision: str = "z1u2v3w4x5y6"
-down_revision: str | Sequence[str] | None = "a1b2c3d4e5f6"
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-
-def _get_schema_prefix() -> str:
-    """Get schema prefix for table names (required for multi-tenant support)."""
-    schema = context.config.get_main_option("target_schema")
-    return f'"{schema}".' if schema else ""
-
-
-def upgrade() -> None:
-    schema = _get_schema_prefix()
-    op.execute(f"ALTER TABLE {schema}memory_units ADD COLUMN IF NOT EXISTS observation_scopes JSONB")
-
-
-def downgrade() -> None:
-    schema = _get_schema_prefix()
-    op.execute(f"ALTER TABLE {schema}memory_units DROP COLUMN IF EXISTS observation_scopes")

hindsight-api-slim/hindsight_api/engine/consolidation/prompts.py

@@ -1,83 +0,0 @@
-"""Prompts for the consolidation engine."""
-
-# Default mission when no bank-specific mission is set
-_DEFAULT_MISSION = "Track every detail: names, numbers, dates, places, and relationships. Prefer specifics over abstractions, never generalise."
-
-# Processing rules — always present regardless of mission
-_PROCESSING_RULES = """Processing rules (always apply):
-- REDUNDANT: same info worded differently → UPDATE the existing observation.
-- CONTRADICTION/UPDATE: capture both states with temporal markers ("used to X, now Y").
-- RESOLVE REFERENCES: when a new fact provides a concrete value resolving a vague placeholder in an existing observation (e.g. "home country", "hometown", "birthplace", "native language", "her ex", "that city"), UPDATE the observation to embed the resolved value explicitly. Example: new fact says "grandma in Sweden" + existing observation says "moved from her home country" → update to "home country is Sweden".
-- NEVER merge observations about different people or unrelated topics."""
-
-# Data section — format placeholders {facts_text} and {observations_text} are substituted at call time
-_BATCH_DATA_SECTION = """
-NEW FACTS:
-{facts_text}
-
-EXISTING OBSERVATIONS (JSON array, pooled from recalls across all facts above):
-{observations_text}
-
-Each observation includes:
-- id: unique identifier for updating
-- text: the observation content
-- proof_count: number of supporting memories
-- occurred_start/occurred_end: temporal range of source facts
-- source_memories: array of supporting facts with their text and dates
-
-Compare the facts against existing observations:
-- Same topic as an existing observation → UPDATE it (observation_id + source_fact_ids)
-- New topic with durable knowledge → CREATE a new observation (source_fact_ids)
-- Cross-reference facts within the batch: a later fact may resolve a vague reference in an earlier one
-- Purely ephemeral facts → omit them unless the MISSION above explicitly targets such data (e.g. timestamped events, session state, screen content)"""
-
-# Output format — JSON braces escaped as {{ }} so .format() leaves them literal
-_BATCH_OUTPUT_FORMAT = """
-Output a JSON object with three arrays.
-
-## EXAMPLE
-
-Input facts:
-[a1b2c3d4-e5f6-7890-abcd-ef1234567890] Alice mentioned she works long hours, often past midnight | Involving: Alice (occurred_start=2024-01-15, mentioned_at=2024-01-15)
-[b2c3d4e5-f6a7-8901-bcde-f12345678901] Alice said she's exhausted from the project deadlines | Involving: Alice (occurred_start=2024-01-20, mentioned_at=2024-01-20)
-
-Good observation text — clean prose, no metadata, each fact tracked distinctly:
-  "Alice works long hours, often past midnight."
-  "Alice feels exhausted from project deadlines."
-
-Bad observation text — NEVER do this (verbatim copy of fact text with metadata):
-  "Alice mentioned she works long hours, often past midnight | Involving: Alice (occurred_start=2024-01-15, mentioned_at=2024-01-15)"
-
-Observation text rules:
-- Write clean prose — NEVER copy raw fact lines or their metadata (temporal fields, "Involving:", "When:" labels, UUIDs).
-- Parenthesized metadata like (occurred_start=...) and pipe-separated labels like "| Involving: ..." are fact formatting — strip them entirely from observation text.
-- How many observations to create and how much to aggregate is driven by the MISSION above.
-
-{{"creates": [{{"text": "Alice works long hours, often past midnight.", "source_fact_ids": ["a1b2c3d4-e5f6-7890-abcd-ef1234567890"]}}, {{"text": "Alice feels exhausted from project deadlines.", "source_fact_ids": ["b2c3d4e5-f6a7-8901-bcde-f12345678901"]}}],
-  "updates": [{{"text": "Alice works at Acme Corp as a senior engineer", "observation_id": "c3d4e5f6-a7b8-9012-cdef-123456789012", "source_fact_ids": ["d4e5f6a7-b8c9-0123-defa-234567890123"]}}],
-  "deletes": [{{"observation_id": "e5f6a7b8-c9d0-1234-efab-345678901234"}}]}}
-
-Rules:
-- "source_fact_ids": copy the EXACT UUID strings shown in brackets [uuid] from NEW FACTS — never use integers or positions.
-- "observation_id": copy the EXACT "id" UUID string from EXISTING OBSERVATIONS.
-- One create/update may reference multiple facts when they jointly support the observation.
-- "deletes": only when an observation is directly superseded or contradicted by new facts.
-- Do NOT include "tags" — handled automatically.
-- Return {{"creates": [], "updates": [], "deletes": []}} if nothing durable is found."""
-
-
-def build_batch_consolidation_prompt(observations_mission: str | None = None) -> str:
-    """
-    Build the consolidation prompt for batch mode (multiple facts per LLM call).
-
-    The mission defines *what* to track (customisable per bank).
-    Processing rules and output format are always present regardless of mission.
-    """
-    mission = observations_mission or _DEFAULT_MISSION
-
-    return (
-        "You are a memory consolidation system. Synthesize facts into observations "
-        "and merge with existing observations when appropriate.\n\n"
-        f"## MISSION\n{mission}\n\n"
-        f"{_PROCESSING_RULES}" + _BATCH_DATA_SECTION + _BATCH_OUTPUT_FORMAT
-    )

hindsight-api-slim/hindsight_api/engine/jina_mlx_reranker.py

@@ -1,144 +0,0 @@
-"""
-MLX implementation of jina-reranker-v3 for Apple Silicon.
-
-This file is adapted from the official model repository:
-  https://huggingface.co/jinaai/jina-reranker-v3-mlx/blob/main/rerank.py
-
-License: CC BY-NC 4.0 (contact Jina AI for commercial usage)
-
-Changes from upstream:
-- Removed the __main__ example block
-- Type annotations added to public methods
-- top_n parameter added to rerank() (upstream only exposed it implicitly)
-"""
-
-import numpy as np
-
-
-class _MLPProjector:
-    def __init__(self):
-        import mlx.nn as nn
-
-        self.linear1 = nn.Linear(1024, 512, bias=False)
-        self.linear2 = nn.Linear(512, 512, bias=False)
-
-    def __call__(self, x):
-        import mlx.nn as nn
-
-        x = self.linear1(x)
-        x = nn.relu(x)
-        x = self.linear2(x)
-        return x
-
-
-def _load_projector(projector_path: str) -> _MLPProjector:
-    import mlx.core as mx
-    from safetensors import safe_open
-
-    projector = _MLPProjector()
-    with safe_open(projector_path, framework="numpy") as f:
-        projector.linear1.weight = mx.array(f.get_tensor("linear1.weight"))
-        projector.linear2.weight = mx.array(f.get_tensor("linear2.weight"))
-    return projector
-
-
-def _sanitize(text: str, special_tokens: dict[str, str]) -> str:
-    for token in special_tokens.values():
-        text = text.replace(token, "")
-    return text
-
-
-def _format_prompt(query: str, docs: list[str], special_tokens: dict[str, str]) -> str:
-    query = _sanitize(query, special_tokens)
-    docs = [_sanitize(d, special_tokens) for d in docs]
-
-    doc_token = special_tokens["doc_embed_token"]
-    query_token = special_tokens["query_embed_token"]
-
-    prefix = (
-        "<|im_start|>system\n"
-        "You are a search relevance expert who can determine a ranking of the passages based on how relevant they are to the query. "
-        "If the query is a question, how relevant a passage is depends on how well it answers the question. "
-        "If not, try to analyze the intent of the query and assess how well each passage satisfies the intent. "
-        "If an instruction is provided, you should follow the instruction when determining the ranking."
-        "<|im_end|>\n<|im_start|>user\n"
-    )
-    suffix = "<|im_end|>\n<|im_start|>assistant\n<think>\n\n</think>\n\n"
-
-    body = (
-        f"I will provide you with {len(docs)} passages, each indicated by a numerical identifier. "
-        f"Rank the passages based on their relevance to query: {query}\n"
-    )
-    body += "\n".join(f'<passage id="{i}">\n{doc}{doc_token}\n</passage>' for i, doc in enumerate(docs))
-    body += f"\n<query>\n{query}{query_token}\n</query>"
-    return prefix + body + suffix
-
-
-class MLXReranker:
-    """
-    MLX-accelerated jina-reranker-v3 for Apple Silicon.
-
-    Loads the model from a local directory (use huggingface_hub.snapshot_download
-    to fetch jinaai/jina-reranker-v3-mlx if you don't have it already).
-    """
-
-    _SPECIAL_TOKENS = {
-        "query_embed_token": "<|rerank_token|>",
-        "doc_embed_token": "<|embed_token|>",
-    }
-    _DOC_TOKEN_ID = 151670
-    _QUERY_TOKEN_ID = 151671
-
-    def __init__(self, model_path: str, projector_path: str):
-        from mlx_lm import load
-
-        self.model, self.tokenizer = load(model_path)
-        self.model.eval()
-        self.projector = _load_projector(projector_path)
-
-    def rerank(self, query: str, documents: list[str], top_n: int | None = None) -> list[dict]:
-        """
-        Rank documents by relevance to a query.
-
-        Returns a list of dicts with keys: document, relevance_score, index.
-        Sorted by descending relevance_score.
-        """
-        import mlx.core as mx
-
-        prompt = _format_prompt(query, documents, self._SPECIAL_TOKENS)
-        input_ids = self.tokenizer.encode(prompt)
-        hidden_states = self.model.model([input_ids])[0]  # [seq_len, hidden_size]
-
-        input_ids_np = np.array(input_ids)
-        query_positions = np.where(input_ids_np == self._QUERY_TOKEN_ID)[0]
-        doc_positions = np.where(input_ids_np == self._DOC_TOKEN_ID)[0]
-
-        if len(query_positions) == 0:
-            raise ValueError("Query embed token not found in prompt")
-        if len(doc_positions) == 0:
-            raise ValueError("Document embed tokens not found in prompt")
-
-        query_hidden = mx.expand_dims(hidden_states[int(query_positions[0])], axis=0)
-        doc_hidden = mx.stack([hidden_states[int(p)] for p in doc_positions])
-
-        query_emb = self.projector(query_hidden)  # [1, 512]
-        doc_emb = self.projector(doc_hidden)  # [num_docs, 512]
-
-        query_exp = mx.broadcast_to(mx.expand_dims(query_emb, 0), (1, len(documents), 512))
-        doc_exp = mx.expand_dims(doc_emb, 0)
-
-        scores = mx.sum(doc_exp * query_exp, axis=-1) / (
-            mx.sqrt(mx.sum(doc_exp * doc_exp, axis=-1)) * mx.sqrt(mx.sum(query_exp * query_exp, axis=-1))
-        )  # [1, num_docs]
-        scores_np = np.array(scores[0])
-
-        order = np.argsort(scores_np)[::-1]
-        n = min(top_n, len(documents)) if top_n is not None else len(documents)
-        return [
-            {
-                "document": documents[order[i]],
-                "relevance_score": float(scores_np[order[i]]),
-                "index": int(order[i]),
-            }
-            for i in range(n)
-        ]

hindsight-api-slim/hindsight_api/engine/operation_metadata.py

@@ -1,69 +0,0 @@
-"""
-Typed metadata models for async operations.
-
-These dataclasses define the structure of result_metadata for different operation types.
-The metadata is exposed in the API for debugging purposes and may change without notice.
-"""
-
-from dataclasses import asdict, dataclass
-from typing import Any
-
-
-@dataclass
-class BatchRetainParentMetadata:
-    """Metadata for parent batch_retain operations (when split into sub-batches)."""
-
-    items_count: int
-    total_tokens: int
-    num_sub_batches: int
-    is_parent: bool = True
-
-    def to_dict(self) -> dict[str, Any]:
-        """Convert to dict for JSON serialization."""
-        return asdict(self)
-
-
-@dataclass
-class BatchRetainChildMetadata:
-    """Metadata for child batch_retain operations (individual sub-batches)."""
-
-    items_count: int
-    parent_operation_id: str
-    sub_batch_index: int
-    total_sub_batches: int
-
-    def to_dict(self) -> dict[str, Any]:
-        """Convert to dict for JSON serialization."""
-        return asdict(self)
-
-
-@dataclass
-class RetainMetadata:
-    """Metadata for regular retain operations (non-batched, deprecated async path)."""
-
-    items_count: int
-
-    def to_dict(self) -> dict[str, Any]:
-        """Convert to dict for JSON serialization."""
-        return asdict(self)
-
-
-@dataclass
-class ConsolidationMetadata:
-    """Metadata for consolidation operations."""
-
-    # Currently empty, but structure for future fields
-    def to_dict(self) -> dict[str, Any]:
-        """Convert to dict for JSON serialization."""
-        return asdict(self)
-
-
-@dataclass
-class RefreshMentalModelMetadata:
-    """Metadata for mental model refresh operations."""
-
-    mental_model_id: str
-
-    def to_dict(self) -> dict[str, Any]:
-        """Convert to dict for JSON serialization."""
-        return asdict(self)

hindsight-api-slim/hindsight_api/engine/parsers/__init__.py

@@ -1,128 +0,0 @@
-"""File parser implementations."""
-
-import logging
-from dataclasses import dataclass
-
-from .base import FileParser, UnsupportedFileTypeError
-from .iris import IrisParser
-from .markitdown import MarkitdownParser
-
-__all__ = [
-    "FileParser",
-    "UnsupportedFileTypeError",
-    "IrisParser",
-    "MarkitdownParser",
-    "FileParserRegistry",
-    "ConvertResult",
-]
-
-
-@dataclass
-class ConvertResult:
-    """Result of a successful file conversion."""
-
-    content: str
-    parser_name: str
-
-
-logger = logging.getLogger(__name__)
-
-
-class FileParserRegistry:
-    """Registry for file parsers with auto-detection."""
-
-    def __init__(self):
-        """Initialize empty parser registry."""
-        self._parsers: dict[str, FileParser] = {}
-
-    def register(self, parser: FileParser):
-        """
-        Register a parser.
-
-        Args:
-            parser: FileParser instance
-        """
-        self._parsers[parser.name()] = parser
-
-    def get_parser(
-        self,
-        name: str | None,
-        filename: str,
-        content_type: str | None = None,
-    ) -> FileParser:
-        """
-        Get parser by name or auto-detect.
-
-        Args:
-            name: Parser name (e.g., "markitdown") or None for auto-detect
-            filename: File name for auto-detection
-            content_type: MIME type (optional)
-
-        Returns:
-            FileParser instance
-
-        Raises:
-            ValueError: If no suitable parser found
-        """
-        if name:
-            # Explicit parser requested — return it directly, let the parser
-            # raise UnsupportedFileTypeError from convert() if needed
-            if name not in self._parsers:
-                raise ValueError(f"Parser '{name}' not found. Available: {list(self._parsers.keys())}")
-            return self._parsers[name]
-
-        # Auto-detect parser
-        for parser in self._parsers.values():
-            if parser.supports(filename, content_type):
-                return parser
-
-        raise ValueError(f"No parser found for {filename}. Available parsers: {list(self._parsers.keys())}")
-
-    async def convert_with_fallback(
-        self,
-        parsers: list[str],
-        file_data: bytes,
-        filename: str,
-        content_type: str | None = None,
-    ) -> ConvertResult:
-        """
-        Try each parser in order, falling back on failure or empty content.
-
-        Moves to the next parser if the current one raises UnsupportedFileTypeError
-        or returns empty content. Any other exception (RuntimeError, network error,
-        etc.) also triggers a fallback so the chain is exhausted before failing.
-
-        Args:
-            parsers: Ordered list of parser names to try
-            file_data: Raw file bytes
-            filename: Original filename
-            content_type: MIME type (optional)
-
-        Returns:
-            ConvertResult with the parsed content and the name of the parser that succeeded
-
-        Raises:
-            ValueError: If a parser name is not registered
-            RuntimeError: If all parsers fail or return empty content
-        """
-        last_error: Exception | None = None
-        for name in parsers:
-            parser = self.get_parser(name, filename, content_type)
-            try:
-                content = await parser.convert(file_data, filename)
-                if content and content.strip():
-                    return ConvertResult(content=content, parser_name=name)
-                logger.warning(f"Parser '{name}' returned empty content for '{filename}', trying next")
-                last_error = RuntimeError(f"Parser '{name}' returned no content for '{filename}'")
-            except UnsupportedFileTypeError as e:
-                logger.warning(f"Parser '{name}' does not support '{filename}', trying next: {e}")
-                last_error = e
-            except Exception as e:
-                logger.warning(f"Parser '{name}' failed for '{filename}', trying next: {e}")
-                last_error = e
-
-        raise last_error or RuntimeError(f"No parsers available for '{filename}'")
-
-    def list_parsers(self) -> list[str]:
-        """Get list of registered parser names."""
-        return list(self._parsers.keys())

hindsight-api-slim/hindsight_api/engine/parsers/base.py

@@ -1,58 +0,0 @@
-"""Abstract base class for file parsers."""
-
-from abc import ABC, abstractmethod
-
-
-class UnsupportedFileTypeError(Exception):
-    """Raised by a parser when it does not support the given file type."""
-
-    pass
-
-
-class FileParser(ABC):
-    """Abstract base for file to markdown parsers."""
-
-    @abstractmethod
-    async def convert(self, file_data: bytes, filename: str) -> str:
-        """
-        Parse file to markdown.
-
-        Args:
-            file_data: Raw file bytes
-            filename: Original filename (used for format detection)
-
-        Returns:
-            Markdown content as string
-
-        Raises:
-            UnsupportedFileTypeError: If the file type is not supported by this parser
-            RuntimeError: If parsing fails for another reason
-        """
-        pass
-
-    def supports(self, filename: str, content_type: str | None = None) -> bool:
-        """
-        Check if parser supports this file type.
-
-        Override this for local/static extension-based filtering.
-        Parsers that delegate to a remote service should leave this as True
-        and raise UnsupportedFileTypeError from convert() instead.
-
-        Args:
-            filename: File name (used for extension check)
-            content_type: MIME type (optional)
-
-        Returns:
-            True if this parser can handle the file (default: True)
-        """
-        return True
-
-    @abstractmethod
-    def name(self) -> str:
-        """
-        Get parser name.
-
-        Returns:
-            Parser name (e.g., "markitdown")
-        """
-        pass

hindsight-api-slim/hindsight_api/engine/parsers/iris.py

@@ -1,138 +0,0 @@
-"""Iris parser implementation using the Vectorize Iris HTTP API."""
-
-import asyncio
-import logging
-import mimetypes
-import time
-
-import httpx
-
-from .base import FileParser, UnsupportedFileTypeError
-
-logger = logging.getLogger(__name__)
-
-_IRIS_BASE_URL = "https://api.vectorize.io/v1"
-_DEFAULT_POLL_INTERVAL = 2.0  # seconds
-_DEFAULT_TIMEOUT = 300.0  # seconds
-
-
-class IrisParser(FileParser):
-    """
-    Iris file parser using the Vectorize Iris cloud extraction service.
-
-    Uploads files to the Vectorize Iris API, starts an extraction job,
-    and polls until the text is ready. The API determines which file types
-    are supported — UnsupportedFileTypeError is raised if the file is rejected.
-
-    Authentication:
-        Requires HINDSIGHT_API_FILE_PARSER_IRIS_TOKEN and
-        HINDSIGHT_API_FILE_PARSER_IRIS_ORG_ID environment variables,
-        or pass them explicitly via the constructor.
-    """
-
-    def __init__(
-        self,
-        token: str,
-        org_id: str,
-        poll_interval: float = _DEFAULT_POLL_INTERVAL,
-        timeout: float = _DEFAULT_TIMEOUT,
-    ):
-        """
-        Initialize iris parser.
-
-        Args:
-            token: Vectorize API token
-            org_id: Vectorize organization ID
-            poll_interval: Seconds between status poll requests (default: 2)
-            timeout: Maximum seconds to wait for extraction (default: 300)
-        """
-        self._token = token
-        self._org_id = org_id
-        self._poll_interval = poll_interval
-        self._timeout = timeout
-        self._auth_headers = {"Authorization": f"Bearer {token}"}
-
-    async def convert(self, file_data: bytes, filename: str) -> str:
-        """
-        Parse file to text using the Vectorize Iris API.
-
-        Raises:
-            UnsupportedFileTypeError: If the Iris API rejects the file type (4xx)
-            RuntimeError: If extraction fails for another reason
-        """
-        content_type = mimetypes.guess_type(filename)[0] or "application/octet-stream"
-
-        async with httpx.AsyncClient(timeout=httpx.Timeout(30.0, read=120.0)) as client:
-            # Step 1: Request a presigned upload URL
-            init_resp = await client.post(
-                f"{_IRIS_BASE_URL}/org/{self._org_id}/files",
-                headers=self._auth_headers,
-                json={"name": filename, "contentType": content_type},
-            )
-            _raise_for_status(init_resp, filename, "file upload init")
-            init_data = init_resp.json()
-            file_id: str = init_data["fileId"]
-            upload_url: str = init_data["uploadUrl"]
-
-            # Step 2: Upload the file bytes to the presigned URL (no auth header)
-            # Ensure file_data is plain bytes (GCS storage may return obstore.Bytes)
-            upload_resp = await client.put(
-                upload_url,
-                content=bytes(file_data),
-                headers={"Content-Type": content_type},
-            )
-            _raise_for_status(upload_resp, filename, "file upload")
-
-            # Step 3: Start extraction
-            extract_resp = await client.post(
-                f"{_IRIS_BASE_URL}/org/{self._org_id}/extraction",
-                headers=self._auth_headers,
-                json={"fileId": file_id},
-            )
-            _raise_for_status(extract_resp, filename, "start extraction")
-            extraction_id: str = extract_resp.json()["extractionId"]
-
-            # Step 4: Poll until ready or timeout
-            deadline = time.monotonic() + self._timeout
-            while True:
-                status_resp = await client.get(
-                    f"{_IRIS_BASE_URL}/org/{self._org_id}/extraction/{extraction_id}",
-                    headers=self._auth_headers,
-                )
-                _raise_for_status(status_resp, filename, "poll extraction status")
-                status_data = status_resp.json()
-
-                if status_data.get("ready"):
-                    data = status_data.get("data", {})
-                    if not data.get("success"):
-                        error = data.get("error", "unknown error")
-                        raise RuntimeError(f"Iris extraction failed for '{filename}': {error}")
-                    text = data.get("text")
-                    if not text:
-                        raise RuntimeError(f"No content extracted from '{filename}'")
-                    return text
-
-                if time.monotonic() >= deadline:
-                    raise RuntimeError(f"Iris extraction timed out after {self._timeout}s for '{filename}'")
-
-                await asyncio.sleep(self._poll_interval)
-
-    def name(self) -> str:
-        """Get parser name."""
-        return "iris"
-
-
-def _raise_for_status(response: httpx.Response, filename: str, step: str) -> None:
-    """
-    Raise an appropriate error including the response body on HTTP errors.
-
-    Raises UnsupportedFileTypeError for 4xx responses (file rejected by the API),
-    RuntimeError for other HTTP errors.
-    """
-    if not response.is_error:
-        return
-    body = response.text or "<empty>"
-    msg = f"Iris API error during {step} for '{filename}': {response.status_code} {response.reason_phrase} — {body}"
-    if response.is_client_error:
-        raise UnsupportedFileTypeError(msg)
-    raise RuntimeError(msg)

hindsight-api-slim/hindsight_api/engine/parsers/markitdown.py

@@ -1,109 +0,0 @@
-"""Markitdown parser implementation."""
-
-import asyncio
-import logging
-import tempfile
-from pathlib import Path
-
-from .base import FileParser
-
-logger = logging.getLogger(__name__)
-
-
-class MarkitdownParser(FileParser):
-    """
-    Markitdown file parser.
-
-    Uses Microsoft's markitdown library to convert various file formats
-    to markdown including PDF, Office docs, images (via OCR), audio, HTML.
-
-    Supported formats:
-    - PDF (.pdf)
-    - Word (.docx, .doc)
-    - PowerPoint (.pptx, .ppt)
-    - Excel (.xlsx, .xls)
-    - Images (.jpg, .jpeg, .png) - with OCR
-    - HTML (.html, .htm)
-    - Text (.txt, .md)
-    - Audio (.mp3, .wav) - with transcription
-    """
-
-    def __init__(self):
-        """Initialize markitdown parser."""
-        # Lazy import to avoid requiring markitdown for all users
-        try:
-            from markitdown import MarkItDown
-
-            self._markitdown = MarkItDown()
-        except ImportError as e:
-            raise ImportError(
-                "markitdown package is required for file parsing. Install with: pip install markitdown"
-            ) from e
-
-    async def convert(self, file_data: bytes, filename: str) -> str:
-        """Parse file to markdown using markitdown."""
-        # markitdown is synchronous, so we run it in executor to avoid blocking
-        loop = asyncio.get_event_loop()
-        return await loop.run_in_executor(None, self._convert_sync, file_data, filename)
-
-    def _convert_sync(self, file_data: bytes, filename: str) -> str:
-        """Synchronous parsing (runs in thread pool)."""
-        # Write to temp file (markitdown requires file path)
-        with tempfile.NamedTemporaryFile(suffix=Path(filename).suffix, delete=False) as tmp:
-            tmp.write(file_data)
-            tmp_path = tmp.name
-
-        try:
-            # Parse using markitdown
-            result = self._markitdown.convert(tmp_path)
-
-            if not result or not result.text_content:
-                raise RuntimeError(f"No content extracted from '{filename}'")
-
-            return result.text_content
-
-        except Exception as e:
-            logger.error(f"Markitdown parsing failed for {filename}: {e}")
-            raise RuntimeError(f"Failed to parse '{filename}': {e}") from e
-
-        finally:
-            # Clean up temp file
-            try:
-                Path(tmp_path).unlink()
-            except Exception:
-                pass
-
-    def supports(self, filename: str, content_type: str | None = None) -> bool:
-        """Check if markitdown supports this file type."""
-        # Supported extensions (from markitdown docs)
-        supported_extensions = {
-            # Documents
-            ".pdf",
-            ".docx",
-            ".doc",
-            ".pptx",
-            ".ppt",
-            ".xlsx",
-            ".xls",
-            # Images (with OCR)
-            ".jpg",
-            ".jpeg",
-            ".png",
-            # Web
-            ".html",
-            ".htm",
-            # Text
-            ".txt",
-            ".md",
-            ".csv",
-            # Audio (with transcription)
-            ".mp3",
-            ".wav",
-        }
-
-        ext = Path(filename).suffix.lower()
-        return ext in supported_extensions
-
-    def name(self) -> str:
-        """Get parser name."""
-        return "markitdown"

hindsight-api-slim/hindsight_api/engine/providers/litellm_llm.py

@@ -1,380 +0,0 @@
-"""
-LiteLLM LLM provider for universal model support.
-
-This provider enables using 100+ LLM providers via the LiteLLM SDK, including:
-- AWS Bedrock (bedrock/anthropic.claude-3-5-sonnet-...)
-- Azure OpenAI (azure/gpt-4o)
-- Together AI (together_ai/meta-llama/...)
-- Any other LiteLLM-supported provider
-
-Uses litellm.acompletion() for async chat completions.
-Authentication for cloud providers (e.g., AWS Bedrock via boto3 credential chain)
-is handled automatically by LiteLLM.
-"""
-
-import asyncio
-import json
-import logging
-import time
-from typing import Any
-
-from hindsight_api.engine.llm_interface import LLMInterface, OutputTooLongError
-from hindsight_api.engine.response_models import LLMToolCall, LLMToolCallResult, TokenUsage
-from hindsight_api.metrics import get_metrics_collector
-
-logger = logging.getLogger(__name__)
-
-
-class LiteLLMLLM(LLMInterface):
-    """
-    LLM provider using the LiteLLM SDK for universal model support.
-
-    Supports any model accessible via litellm.acompletion(), including AWS Bedrock,
-    Azure OpenAI, Together AI, Fireworks AI, and more.
-
-    Model names follow LiteLLM conventions with provider prefixes:
-    - bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0
-    - azure/gpt-4o
-    - together_ai/meta-llama/Llama-3-70b-chat-hf
-    - fireworks_ai/accounts/fireworks/models/llama-v3p1-70b-instruct
-    """
-
-    def __init__(
-        self,
-        provider: str,
-        api_key: str,
-        base_url: str,
-        model: str,
-        reasoning_effort: str = "low",
-        timeout: float = 300.0,
-        **kwargs: Any,
-    ):
-        super().__init__(provider, api_key, base_url, model, reasoning_effort, **kwargs)
-        self.timeout = timeout
-        self._litellm: Any = None
-
-        try:
-            import litellm
-
-            self._litellm = litellm
-            # Suppress LiteLLM's verbose logging
-            litellm.suppress_debug_info = True  # type: ignore[assignment]
-            # Drop unsupported params instead of raising errors (e.g. tool_choice on some Bedrock models)
-            litellm.drop_params = True  # type: ignore[assignment]
-            logging.getLogger("LiteLLM").setLevel(logging.WARNING)
-            logger.info(f"LiteLLM SDK initialized for model: {self.model}")
-        except ImportError as e:
-            raise RuntimeError("LiteLLM SDK not installed. Run: uv add litellm or pip install litellm") from e
-
-    async def verify_connection(self) -> None:
-        try:
-            test_messages = [{"role": "user", "content": "test"}]
-            await self.call(
-                messages=test_messages,
-                max_completion_tokens=50,
-                temperature=0.0,
-                scope="verification",
-                max_retries=0,
-            )
-            logger.info("LiteLLM connection verified successfully")
-        except OutputTooLongError:
-            # Truncation is fine for verification — it means the connection works
-            logger.info("LiteLLM connection verified successfully (response truncated)")
-        except Exception as e:
-            logger.error(f"LiteLLM connection verification failed: {e}")
-            raise RuntimeError(f"Failed to verify LiteLLM connection: {e}") from e
-
-    def _build_common_kwargs(
-        self,
-        messages: list[dict[str, Any]],
-        max_completion_tokens: int | None = None,
-        temperature: float | None = None,
-    ) -> dict[str, Any]:
-        """Build common kwargs for litellm calls."""
-        kwargs: dict[str, Any] = {
-            "model": self.model,
-            "messages": messages,
-            "timeout": self.timeout,
-        }
-
-        if self.api_key:
-            kwargs["api_key"] = self.api_key
-        if self.base_url:
-            kwargs["api_base"] = self.base_url
-        if max_completion_tokens is not None:
-            kwargs["max_completion_tokens"] = max_completion_tokens
-        if temperature is not None:
-            kwargs["temperature"] = temperature
-
-        return kwargs
-
-    async def call(
-        self,
-        messages: list[dict[str, str]],
-        response_format: Any | None = None,
-        max_completion_tokens: int | None = None,
-        temperature: float | None = None,
-        scope: str = "memory",
-        max_retries: int = 10,
-        initial_backoff: float = 1.0,
-        max_backoff: float = 60.0,
-        skip_validation: bool = False,
-        strict_schema: bool = False,
-        return_usage: bool = False,
-    ) -> Any:
-        start_time = time.time()
-
-        call_kwargs = self._build_common_kwargs(messages, max_completion_tokens, temperature)
-
-        # Add JSON schema response format if provided
-        if response_format is not None and hasattr(response_format, "model_json_schema"):
-            schema = response_format.model_json_schema()
-            call_kwargs["response_format"] = {
-                "type": "json_schema",
-                "json_schema": {
-                    "name": response_format.__name__ if hasattr(response_format, "__name__") else "response",
-                    "schema": schema,
-                    "strict": strict_schema,
-                },
-            }
-
-        last_exception = None
-
-        for attempt in range(max_retries + 1):
-            try:
-                response = await self._litellm.acompletion(**call_kwargs)
-
-                content = response.choices[0].message.content or ""
-                finish_reason = response.choices[0].finish_reason
-
-                # Check for length-limited output
-                if finish_reason == "length":
-                    raise OutputTooLongError("LiteLLM response was truncated due to token limit")
-
-                if response_format is not None:
-                    # Strip markdown code fences if present
-                    clean_content = content
-                    if "```json" in content:
-                        clean_content = content.split("```json")[1].split("```")[0].strip()
-                    elif "```" in content:
-                        clean_content = content.split("```")[1].split("```")[0].strip()
-
-                    try:
-                        json_data = json.loads(clean_content)
-                    except json.JSONDecodeError:
-                        json_data = json.loads(content)
-
-                    if skip_validation:
-                        result = json_data
-                    else:
-                        result = response_format.model_validate(json_data)
-                else:
-                    result = content
-
-                # Extract usage
-                input_tokens = getattr(response.usage, "prompt_tokens", 0) or 0
-                output_tokens = getattr(response.usage, "completion_tokens", 0) or 0
-                total_tokens = input_tokens + output_tokens
-
-                # Record metrics
-                duration = time.time() - start_time
-                metrics = get_metrics_collector()
-                metrics.record_llm_call(
-                    provider=self.provider,
-                    model=self.model,
-                    scope=scope,
-                    duration=duration,
-                    input_tokens=input_tokens,
-                    output_tokens=output_tokens,
-                    success=True,
-                )
-
-                # Record trace span
-                from hindsight_api.tracing import _serialize_for_span, get_span_recorder
-
-                span_recorder = get_span_recorder()
-                span_recorder.record_llm_call(
-                    provider=self.provider,
-                    model=self.model,
-                    scope=scope,
-                    messages=messages,
-                    response_content=_serialize_for_span(result),
-                    input_tokens=input_tokens,
-                    output_tokens=output_tokens,
-                    duration=duration,
-                    finish_reason=finish_reason,
-                    error=None,
-                )
-
-                if duration > 10.0:
-                    logger.info(
-                        f"slow llm call: scope={scope}, model={self.provider}/{self.model}, "
-                        f"input_tokens={input_tokens}, output_tokens={output_tokens}, "
-                        f"time={duration:.3f}s"
-                    )
-
-                if return_usage:
-                    token_usage = TokenUsage(
-                        input_tokens=input_tokens,
-                        output_tokens=output_tokens,
-                        total_tokens=total_tokens,
-                    )
-                    return result, token_usage
-                return result
-
-            except OutputTooLongError:
-                raise
-
-            except json.JSONDecodeError as e:
-                last_exception = e
-                if attempt < max_retries:
-                    logger.warning("LiteLLM returned invalid JSON, retrying...")
-                    backoff = min(initial_backoff * (2**attempt), max_backoff)
-                    await asyncio.sleep(backoff)
-                    continue
-                else:
-                    logger.error(f"LiteLLM returned invalid JSON after {max_retries + 1} attempts")
-                    raise
-
-            except Exception as e:
-                error_str = str(e).lower()
-                # Fast fail on auth errors
-                if "401" in error_str or "403" in error_str or "unauthorized" in error_str:
-                    logger.error(f"LiteLLM auth error, not retrying: {e}")
-                    raise
-
-                last_exception = e
-                if attempt < max_retries:
-                    # Retry on rate limits, connection errors, server errors
-                    is_retryable = any(
-                        keyword in error_str
-                        for keyword in ("rate", "limit", "timeout", "connection", "500", "502", "503", "529")
-                    )
-                    if is_retryable:
-                        backoff = min(initial_backoff * (2**attempt), max_backoff)
-                        jitter = backoff * 0.2 * (2 * (time.time() % 1) - 1)
-                        await asyncio.sleep(backoff + jitter)
-                        continue
-
-                logger.error(f"LiteLLM API error after {attempt + 1} attempts: {e}")
-                raise
-
-        if last_exception:
-            raise last_exception
-        raise RuntimeError("LiteLLM call failed after all retries")
-
-    async def call_with_tools(
-        self,
-        messages: list[dict[str, Any]],
-        tools: list[dict[str, Any]],
-        max_completion_tokens: int | None = None,
-        temperature: float | None = None,
-        scope: str = "tools",
-        max_retries: int = 5,
-        initial_backoff: float = 1.0,
-        max_backoff: float = 30.0,
-        tool_choice: str | dict[str, Any] = "auto",
-    ) -> LLMToolCallResult:
-        start_time = time.time()
-
-        call_kwargs = self._build_common_kwargs(messages, max_completion_tokens, temperature)
-        call_kwargs["tools"] = tools
-        call_kwargs["tool_choice"] = tool_choice
-
-        last_exception = None
-        for attempt in range(max_retries + 1):
-            try:
-                response = await self._litellm.acompletion(**call_kwargs)
-
-                message = response.choices[0].message
-                content = message.content
-                finish_reason = response.choices[0].finish_reason
-
-                # Extract tool calls
-                tool_calls: list[LLMToolCall] = []
-                if message.tool_calls:
-                    for tc in message.tool_calls:
-                        arguments = tc.function.arguments
-                        if isinstance(arguments, str):
-                            arguments = json.loads(arguments)
-                        tool_calls.append(
-                            LLMToolCall(
-                                id=tc.id,
-                                name=tc.function.name,
-                                arguments=arguments,
-                            )
-                        )
-
-                # Extract usage
-                input_tokens = getattr(response.usage, "prompt_tokens", 0) or 0
-                output_tokens = getattr(response.usage, "completion_tokens", 0) or 0
-
-                # Record metrics
-                duration = time.time() - start_time
-                metrics = get_metrics_collector()
-                metrics.record_llm_call(
-                    provider=self.provider,
-                    model=self.model,
-                    scope=scope,
-                    duration=duration,
-                    input_tokens=input_tokens,
-                    output_tokens=output_tokens,
-                    success=True,
-                )
-
-                # Record trace span
-                from hindsight_api.tracing import get_span_recorder
-
-                span_recorder = get_span_recorder()
-                tool_calls_dict = (
-                    [{"id": tc.id, "name": tc.name, "arguments": tc.arguments} for tc in tool_calls]
-                    if tool_calls
-                    else None
-                )
-                span_recorder.record_llm_call(
-                    provider=self.provider,
-                    model=self.model,
-                    scope=scope,
-                    messages=messages,
-                    response_content=content,
-                    input_tokens=input_tokens,
-                    output_tokens=output_tokens,
-                    duration=duration,
-                    finish_reason=finish_reason,
-                    error=None,
-                    tool_calls=tool_calls_dict,
-                )
-
-                return LLMToolCallResult(
-                    content=content,
-                    tool_calls=tool_calls,
-                    finish_reason=finish_reason or ("tool_calls" if tool_calls else "stop"),
-                    input_tokens=input_tokens,
-                    output_tokens=output_tokens,
-                )
-
-            except Exception as e:
-                error_str = str(e).lower()
-                if "401" in error_str or "403" in error_str or "unauthorized" in error_str:
-                    raise
-
-                last_exception = e
-                if attempt < max_retries:
-                    is_retryable = any(
-                        keyword in error_str
-                        for keyword in ("rate", "limit", "timeout", "connection", "500", "502", "503", "529")
-                    )
-                    if is_retryable:
-                        await asyncio.sleep(min(initial_backoff * (2**attempt), max_backoff))
-                        continue
-
-                logger.error(f"LiteLLM tool call error after {attempt + 1} attempts: {e}")
-                raise
-
-        if last_exception:
-            raise last_exception
-        raise RuntimeError("LiteLLM tool call failed after all retries")
-
-    async def cleanup(self) -> None:
-        """Clean up resources."""
-        pass

hindsight-api-slim/hindsight_api/engine/providers/none_llm.py

@@ -1,78 +0,0 @@
-"""
-No-op LLM provider for chunk-only storage mode.
-
-When the LLM provider is set to "none", the system operates without any LLM dependency.
-Retain uses chunks mode (no fact extraction), and reflect/consolidation are disabled.
-This provider acts as a safety net — if any code path unexpectedly tries to call the LLM,
-it raises a clear error instead of a confusing connection failure.
-"""
-
-import logging
-from typing import Any
-
-from ..llm_interface import LLMInterface
-from ..response_models import LLMToolCallResult
-
-logger = logging.getLogger(__name__)
-
-
-class LLMNotAvailableError(Exception):
-    """Raised when an operation requires an LLM but the provider is set to 'none'."""
-
-    pass
-
-
-class NoneLLM(LLMInterface):
-    """
-    No-op LLM provider that rejects all LLM calls.
-
-    Used when HINDSIGHT_API_LLM_PROVIDER=none to run Hindsight as a chunk store
-    with semantic search but without LLM-based features (fact extraction, reflect,
-    consolidation).
-    """
-
-    async def verify_connection(self) -> None:
-        """No-op — no LLM connection to verify."""
-        logger.debug("NoneLLM: no LLM connection to verify (provider=none)")
-
-    async def call(
-        self,
-        messages: list[dict[str, str]],
-        response_format: Any | None = None,
-        max_completion_tokens: int | None = None,
-        temperature: float | None = None,
-        scope: str = "memory",
-        max_retries: int = 10,
-        initial_backoff: float = 1.0,
-        max_backoff: float = 60.0,
-        skip_validation: bool = False,
-        strict_schema: bool = False,
-        return_usage: bool = False,
-    ) -> Any:
-        """Raise LLMNotAvailableError — no LLM is configured."""
-        raise LLMNotAvailableError(
-            "LLM provider is set to 'none'. This operation requires an LLM. "
-            "Set HINDSIGHT_API_LLM_PROVIDER to a real provider (e.g., openai, anthropic, gemini)."
-        )
-
-    async def call_with_tools(
-        self,
-        messages: list[dict[str, Any]],
-        tools: list[dict[str, Any]],
-        max_completion_tokens: int | None = None,
-        temperature: float | None = None,
-        scope: str = "tools",
-        max_retries: int = 5,
-        initial_backoff: float = 1.0,
-        max_backoff: float = 30.0,
-        tool_choice: str | dict[str, Any] = "auto",
-    ) -> LLMToolCallResult:
-        """Raise LLMNotAvailableError — no LLM is configured."""
-        raise LLMNotAvailableError(
-            "LLM provider is set to 'none'. This operation requires an LLM. "
-            "Set HINDSIGHT_API_LLM_PROVIDER to a real provider (e.g., openai, anthropic, gemini)."
-        )
-
-    async def cleanup(self) -> None:
-        """No-op — nothing to clean up."""
-        pass

hindsight-api-slim/hindsight_api/engine/retain/entity_labels.py

@@ -1,194 +0,0 @@
-"""
-Entity labels models and helpers for retain pipeline.
-
-Defines a controlled vocabulary of key:value classification labels
-(e.g., 'pedagogy:scaffolding', 'interest:active') that are extracted
-at retain time and stored as entities.
-"""
-
-from typing import Literal
-
-from pydantic import BaseModel, Field, create_model
-
-
-class LabelValue(BaseModel):
-    """A single allowed value for a label group."""
-
-    value: str
-    description: str = ""
-
-
-class LabelGroup(BaseModel):
-    """A label group (dimension) with its type and allowed values."""
-
-    key: str
-    description: str = ""
-    type: Literal["value", "multi-values", "text"] = "value"
-    optional: bool = True
-    tag: bool = False
-    values: list[LabelValue] = []
-
-
-class EntityLabelsConfig(BaseModel):
-    """Entity labels configuration for a bank (controlled vocabulary)."""
-
-    attributes: list[LabelGroup] = []
-
-
-def parse_entity_labels(raw: dict | list | None) -> EntityLabelsConfig | None:
-    """
-    Parse raw entity labels config into EntityLabelsConfig.
-
-    Accepts:
-    - None → returns None
-    - list → list of attribute dicts (each may use legacy free_values/multi_value or new type field)
-    - dict → {attributes: [...]}
-
-    Legacy migration (backward-compat):
-    - free_values=True          → type="text"
-    - multi_value=True          → type="multi-values"
-    - neither / free_values=False → type="value"
-
-    Args:
-        raw: Raw entity labels config from bank config
-
-    Returns:
-        EntityLabelsConfig or None if raw is None/empty
-    """
-    if raw is None:
-        return None
-
-    if isinstance(raw, list):
-        if not raw:
-            return None
-        attributes = [LabelGroup.model_validate(_migrate_label_group(a)) for a in raw]
-        return EntityLabelsConfig(attributes=attributes)
-
-    if isinstance(raw, dict):
-        attrs_raw = raw.get("attributes", [])
-        if not attrs_raw:
-            return None
-        attributes = [LabelGroup.model_validate(_migrate_label_group(a)) for a in attrs_raw]
-        return EntityLabelsConfig(attributes=attributes)
-
-    return None
-
-
-def _migrate_label_group(raw: dict) -> dict:
-    """Migrate legacy free_values/multi_value fields to the new type field."""
-    if not isinstance(raw, dict) or "type" in raw:
-        return raw
-    patched = dict(raw)
-    if patched.get("free_values"):
-        patched["type"] = "text"
-    elif patched.get("multi_value"):
-        patched["type"] = "multi-values"
-    else:
-        patched["type"] = "value"
-    # Remove legacy keys so Pydantic doesn't error on unknown fields
-    patched.pop("free_values", None)
-    patched.pop("multi_value", None)
-    return patched
-
-
-def build_labels_model(labels_cfg: EntityLabelsConfig) -> type[BaseModel] | None:
-    """
-    Build a dynamic Pydantic model for structured label extraction.
-
-    Each LabelGroup becomes a typed field based on its type:
-    - type="text"                        → str | None  (always optional)
-    - type="value",   optional=True      → Literal["v1","v2"] | None
-    - type="value",   optional=False     → Literal["v1","v2"]  (required)
-    - type="multi-values"                → list[Literal["v1","v2"]]
-
-    Args:
-        labels_cfg: Parsed EntityLabelsConfig
-
-    Returns:
-        Dynamic Pydantic model class, or None if no groups defined
-    """
-    fields: dict = {}
-    for group in labels_cfg.attributes:
-        if not group.key:
-            continue
-        description = group.description or group.key
-
-        if group.type == "text":
-            # Free-form: any string value accepted, always optional
-            fields[group.key] = (str | None, Field(default=None, description=description))
-        else:
-            # Enum-constrained: must have defined values
-            if not group.values:
-                continue
-            values = tuple(v.value for v in group.values if v.value)
-            if not values:
-                continue
-            # Literal[("v1", "v2")] is equivalent to Literal["v1", "v2"] in Python 3.11+
-            literal_type = Literal[values]  # type: ignore[valid-type]
-            if group.type == "multi-values":
-                fields[group.key] = (
-                    list[literal_type],  # type: ignore[valid-type]
-                    Field(default_factory=list, description=description),
-                )
-            elif group.optional:
-                fields[group.key] = (
-                    literal_type | None,  # type: ignore[valid-type]
-                    Field(default=None, description=description),
-                )
-            else:
-                fields[group.key] = (
-                    literal_type,  # type: ignore[valid-type]
-                    Field(description=description),
-                )
-
-    if not fields:
-        return None
-
-    return create_model("Labels", **fields)
-
-
-def is_label_entity(text: str, labels_cfg: EntityLabelsConfig, labels_lookup: set[str]) -> bool:
-    """
-    Return True if entity text belongs to any configured label group.
-
-    For enum groups: checks the pre-built lookup set.
-    For text groups: checks that the text starts with a known key prefix.
-    """
-    if text.lower() in labels_lookup:
-        return True
-    for group in labels_cfg.attributes:
-        if group.type == "text" and group.key and text.lower().startswith(f"{group.key.lower()}:"):
-            return True
-    return False
-
-
-def build_labels_lookup(labels_cfg: EntityLabelsConfig | list | None) -> set[str]:
-    """
-    Build a set of valid 'key:value' label strings (lowercase) for fast lookup.
-
-    Accepts either EntityLabelsConfig or raw list/None for backwards compatibility.
-
-    Args:
-        labels_cfg: EntityLabelsConfig, raw list of attribute dicts, or None
-
-    Returns:
-        Set of lowercase 'key:value' strings
-    """
-    if labels_cfg is None:
-        return set()
-
-    # Accept raw list/dict for backwards compatibility
-    if not isinstance(labels_cfg, EntityLabelsConfig):
-        parsed = parse_entity_labels(labels_cfg)
-        if parsed is None:
-            return set()
-        labels_cfg = parsed
-
-    valid = set()
-    for group in labels_cfg.attributes:
-        if group.type == "text":
-            continue  # No fixed vocabulary — all values accepted in post-processing
-        for v in group.values:
-            if group.key and v.value:
-                valid.add(f"{group.key}:{v.value}".lower())
-    return valid

hindsight-api-slim/hindsight_api/engine/retain/orchestrator.py

@@ -1,548 +0,0 @@
-"""
-Main orchestrator for the retain pipeline.
-
-Coordinates all retain pipeline modules to store memories efficiently.
-"""
-
-import logging
-import time
-import uuid
-from collections.abc import Awaitable, Callable
-from datetime import UTC, datetime
-from typing import Any
-
-from ..db_utils import acquire_with_retry, retry_with_backoff
-from . import bank_utils
-
-
-def utcnow():
-    """Get current UTC time."""
-    return datetime.now(UTC)
-
-
-def parse_datetime_flexible(value: Any) -> datetime:
-    """
-    Parse a datetime value that could be either a datetime object or an ISO string.
-
-    This handles datetime values from both direct Python calls and deserialized JSON
-    (where datetime objects are serialized as ISO strings).
-
-    Args:
-        value: Either a datetime object or an ISO format string
-
-    Returns:
-        datetime object (timezone-aware)
-
-    Raises:
-        TypeError: If value is neither datetime nor string
-        ValueError: If string is not a valid ISO datetime
-    """
-    if isinstance(value, datetime):
-        # Ensure timezone-aware
-        if value.tzinfo is None:
-            return value.replace(tzinfo=UTC)
-        return value
-    elif isinstance(value, str):
-        # Parse ISO format string (handles both 'Z' and '+00:00' timezone formats)
-        dt = datetime.fromisoformat(value.replace("Z", "+00:00"))
-        # Ensure timezone-aware
-        if dt.tzinfo is None:
-            return dt.replace(tzinfo=UTC)
-        return dt
-    else:
-        raise TypeError(f"Expected datetime or string, got {type(value).__name__}")
-
-
-import asyncpg
-
-from ..response_models import TokenUsage
-from . import (
-    chunk_storage,
-    embedding_processing,
-    entity_processing,
-    fact_extraction,
-    fact_storage,
-    link_creation,
-)
-from .types import EntityLink, ExtractedFact, ProcessedFact, RetainContent, RetainContentDict
-
-logger = logging.getLogger(__name__)
-
-
-async def retain_batch(
-    pool,
-    embeddings_model,
-    llm_config,
-    entity_resolver,
-    format_date_fn,
-    bank_id: str,
-    contents_dicts: list[RetainContentDict],
-    config,
-    document_id: str | None = None,
-    is_first_batch: bool = True,
-    fact_type_override: str | None = None,
-    confidence_score: float | None = None,
-    document_tags: list[str] | None = None,
-    operation_id: str | None = None,
-    schema: str | None = None,
-    outbox_callback: Callable[["asyncpg.Connection"], Awaitable[None]] | None = None,
-) -> tuple[list[list[str]], TokenUsage]:
-    """
-    Process a batch of content through the retain pipeline.
-
-    Args:
-        pool: Database connection pool
-        embeddings_model: Embeddings model for generating embeddings
-        llm_config: LLM configuration for fact extraction
-        entity_resolver: Entity resolver for entity processing
-        format_date_fn: Function to format datetime to readable string
-        bank_id: Bank identifier
-        contents_dicts: List of content dictionaries
-        config: Resolved HindsightConfig for this bank
-        document_id: Optional document ID
-        is_first_batch: Whether this is the first batch
-        fact_type_override: Override fact type for all facts
-        confidence_score: Confidence score for opinions
-        document_tags: Tags applied to all items in this batch
-
-    Returns:
-        Tuple of (unit ID lists, token usage for fact extraction)
-    """
-    start_time = time.time()
-    total_chars = sum(len(item.get("content", "")) for item in contents_dicts)
-
-    # Buffer all logs
-    log_buffer = []
-    log_buffer.append(f"{'=' * 60}")
-    log_buffer.append(f"RETAIN_BATCH START: {bank_id}")
-    log_buffer.append(f"Batch size: {len(contents_dicts)} content items, {total_chars:,} chars")
-    log_buffer.append(f"{'=' * 60}")
-
-    # Get bank profile
-    profile = await bank_utils.get_bank_profile(pool, bank_id)
-    agent_name = profile["name"]
-
-    # Convert dicts to RetainContent objects
-    contents = []
-    for item in contents_dicts:
-        # Merge item-level tags with document-level tags
-        item_tags = item.get("tags", []) or []
-        merged_tags = list(set(item_tags + (document_tags or [])))
-
-        # Handle event_date: distinguish "not provided" (default to now) from
-        # "explicitly None" (caller opted into no timestamp).
-        if "event_date" in item and item["event_date"] is None:
-            event_date_value = None  # Caller explicitly signalled "unknown date"
-        elif item.get("event_date"):
-            event_date_value = parse_datetime_flexible(item["event_date"])
-        else:
-            event_date_value = utcnow()  # Backward-compatible default
-
-        content = RetainContent(
-            content=item["content"],
-            context=item.get("context", ""),
-            event_date=event_date_value,
-            metadata=item.get("metadata", {}),
-            entities=item.get("entities", []),
-            tags=merged_tags,
-            observation_scopes=item.get("observation_scopes"),
-        )
-        contents.append(content)
-
-    # Step 1: Extract facts from all contents
-    step_start = time.time()
-
-    extracted_facts, chunks, usage = await fact_extraction.extract_facts_from_contents(
-        contents, llm_config, agent_name, config, pool, operation_id, schema
-    )
-    log_buffer.append(
-        f"[1] Extract facts: {len(extracted_facts)} facts, {len(chunks)} chunks from {len(contents)} contents in {time.time() - step_start:.3f}s"
-    )
-
-    if not extracted_facts:
-        # Still need to create document if document_id was provided or chunks exist
-        from collections import defaultdict
-
-        docs_tracked = 0
-        async with acquire_with_retry(pool) as conn:
-            async with conn.transaction():
-                # Group contents by document_id (consistent with normal path)
-                contents_by_doc_early = defaultdict(list)
-                for idx, content_dict in enumerate(contents_dicts):
-                    doc_id = content_dict.get("document_id")
-                    contents_by_doc_early[doc_id].append((idx, content_dict))
-
-                if document_id:
-                    # Legacy: single document_id parameter
-                    combined_content = "\n".join([c.get("content", "") for c in contents_dicts])
-                    # Collect tags from all content items and merge with document_tags
-                    all_tags = set(document_tags or [])
-                    for item in contents_dicts:
-                        item_tags = item.get("tags", []) or []
-                        all_tags.update(item_tags)
-                    merged_tags = list(all_tags)
-
-                    retain_params = {}
-                    if contents_dicts:
-                        first_item = contents_dicts[0]
-                        if first_item.get("context"):
-                            retain_params["context"] = first_item["context"]
-                        if first_item.get("event_date"):
-                            retain_params["event_date"] = (
-                                first_item["event_date"].isoformat()
-                                if hasattr(first_item["event_date"], "isoformat")
-                                else str(first_item["event_date"])
-                            )
-                        if first_item.get("metadata"):
-                            retain_params["metadata"] = first_item["metadata"]
-                    await fact_storage.handle_document_tracking(
-                        conn, bank_id, document_id, combined_content, is_first_batch, retain_params, merged_tags
-                    )
-                    docs_tracked += 1
-                else:
-                    # Handle per-item document_ids and/or chunks (mirrors normal path logic)
-                    has_any_doc_ids = any(item.get("document_id") for item in contents_dicts)
-
-                    if has_any_doc_ids or chunks:
-                        for original_doc_id, doc_contents in contents_by_doc_early.items():
-                            should_create_doc = (original_doc_id is not None) or chunks
-                            if not should_create_doc:
-                                continue
-
-                            actual_doc_id = original_doc_id
-                            if actual_doc_id is None:
-                                # No document_id but have chunks - generate one
-                                actual_doc_id = str(uuid.uuid4())
-
-                            combined_content = "\n".join([c.get("content", "") for _, c in doc_contents])
-                            all_tags = set(document_tags or [])
-                            for _, item in doc_contents:
-                                item_tags = item.get("tags", []) or []
-                                all_tags.update(item_tags)
-                            merged_tags = list(all_tags)
-
-                            retain_params = {}
-                            if doc_contents:
-                                first_item = doc_contents[0][1]
-                                if first_item.get("context"):
-                                    retain_params["context"] = first_item["context"]
-                                if first_item.get("event_date"):
-                                    retain_params["event_date"] = (
-                                        first_item["event_date"].isoformat()
-                                        if hasattr(first_item["event_date"], "isoformat")
-                                        else str(first_item["event_date"])
-                                    )
-                                if first_item.get("metadata"):
-                                    retain_params["metadata"] = first_item["metadata"]
-                            await fact_storage.handle_document_tracking(
-                                conn,
-                                bank_id,
-                                actual_doc_id,
-                                combined_content,
-                                is_first_batch,
-                                retain_params,
-                                merged_tags,
-                            )
-                            docs_tracked += 1
-
-        total_time = time.time() - start_time
-        doc_status = f"{docs_tracked} document(s) tracked" if docs_tracked > 0 else "no document tracked"
-        logger.info(
-            f"RETAIN_BATCH COMPLETE: 0 facts extracted from {len(contents)} contents in {total_time:.3f}s ({doc_status}, no facts)"
-        )
-        return [[] for _ in contents], usage
-
-    # Apply fact_type_override if provided
-    if fact_type_override:
-        for fact in extracted_facts:
-            fact.fact_type = fact_type_override
-
-    # Step 2: Augment texts and generate embeddings
-    step_start = time.time()
-    augmented_texts = embedding_processing.augment_texts_with_dates(extracted_facts, format_date_fn)
-    embeddings = await embedding_processing.generate_embeddings_batch(embeddings_model, augmented_texts)
-    log_buffer.append(f"[2] Generate embeddings: {len(embeddings)} embeddings in {time.time() - step_start:.3f}s")
-
-    # Step 3: Convert to ProcessedFact objects (without chunk_ids yet)
-    processed_facts = [
-        ProcessedFact.from_extracted_fact(extracted_fact, embedding)
-        for extracted_fact, embedding in zip(extracted_facts, embeddings)
-    ]
-
-    # Group contents by document_id for document tracking and chunk storage
-    from collections import defaultdict
-
-    contents_by_doc = defaultdict(list)
-    for idx, content_dict in enumerate(contents_dicts):
-        doc_id = content_dict.get("document_id")
-        contents_by_doc[doc_id].append((idx, content_dict))
-
-    # Step 4: Database transaction (retried on deadlock)
-    result_unit_ids: list[list[str]] = []
-
-    log_buffer_pre_db = len(log_buffer)
-
-    async def _run_db_work() -> None:
-        nonlocal result_unit_ids
-
-        # Reset per-fact mutations and log buffer so each retry attempt starts clean
-        del log_buffer[log_buffer_pre_db:]
-        document_ids_added: list[str] = []
-        for pf in processed_facts:
-            pf.document_id = None
-            pf.chunk_id = None
-
-        # Discard any leftover pending stats from a previous failed attempt so
-        # retries don't double-count or accumulate unbounded state.
-        entity_resolver.discard_pending_stats()
-
-        async with acquire_with_retry(pool) as conn:
-            async with conn.transaction():
-                # Handle document tracking for all documents
-                step_start = time.time()
-                # Map None document_id to generated UUIDs
-                doc_id_mapping = {}  # Maps original doc_id (including None) to actual doc_id used
-
-                if document_id:
-                    # Legacy: single document_id parameter
-                    combined_content = "\n".join([c.get("content", "") for c in contents_dicts])
-                    retain_params = {}
-                    # Collect tags from all content items and merge with document_tags
-                    all_tags = set(document_tags or [])
-                    for item in contents_dicts:
-                        item_tags = item.get("tags", []) or []
-                        all_tags.update(item_tags)
-                    merged_tags = list(all_tags)
-
-                    if contents_dicts:
-                        first_item = contents_dicts[0]
-                        if first_item.get("context"):
-                            retain_params["context"] = first_item["context"]
-                        if first_item.get("event_date"):
-                            retain_params["event_date"] = (
-                                first_item["event_date"].isoformat()
-                                if hasattr(first_item["event_date"], "isoformat")
-                                else str(first_item["event_date"])
-                            )
-                        if first_item.get("metadata"):
-                            retain_params["metadata"] = first_item["metadata"]
-
-                    await fact_storage.handle_document_tracking(
-                        conn, bank_id, document_id, combined_content, is_first_batch, retain_params, merged_tags
-                    )
-                    document_ids_added.append(document_id)
-                    doc_id_mapping[None] = document_id  # For backwards compatibility
-                else:
-                    # Handle per-item document_ids (create documents if any item has document_id or if chunks exist)
-                    has_any_doc_ids = any(item.get("document_id") for item in contents_dicts)
-
-                    if has_any_doc_ids or chunks:
-                        for original_doc_id, doc_contents in contents_by_doc.items():
-                            actual_doc_id = original_doc_id
-
-                            # Only create document record if:
-                            # 1. Item has explicit document_id, OR
-                            # 2. There are chunks (need document for chunk storage)
-                            should_create_doc = (original_doc_id is not None) or chunks
-
-                            if should_create_doc:
-                                if actual_doc_id is None:
-                                    # No document_id but have chunks - generate one
-                                    actual_doc_id = str(uuid.uuid4())
-
-                                # Store mapping for later use
-                                doc_id_mapping[original_doc_id] = actual_doc_id
-
-                                # Combine content for this document
-                                combined_content = "\n".join([c.get("content", "") for _, c in doc_contents])
-
-                                # Collect tags from all content items for this document and merge with document_tags
-                                all_tags = set(document_tags or [])
-                                for _, item in doc_contents:
-                                    item_tags = item.get("tags", []) or []
-                                    all_tags.update(item_tags)
-                                merged_tags = list(all_tags)
-
-                                # Extract retain params from first content item
-                                retain_params = {}
-                                if doc_contents:
-                                    first_item = doc_contents[0][1]
-                                    if first_item.get("context"):
-                                        retain_params["context"] = first_item["context"]
-                                    if first_item.get("event_date"):
-                                        retain_params["event_date"] = (
-                                            first_item["event_date"].isoformat()
-                                            if hasattr(first_item["event_date"], "isoformat")
-                                            else str(first_item["event_date"])
-                                        )
-                                    if first_item.get("metadata"):
-                                        retain_params["metadata"] = first_item["metadata"]
-
-                                await fact_storage.handle_document_tracking(
-                                    conn,
-                                    bank_id,
-                                    actual_doc_id,
-                                    combined_content,
-                                    is_first_batch,
-                                    retain_params,
-                                    merged_tags,
-                                )
-                                document_ids_added.append(actual_doc_id)
-
-                if document_ids_added:
-                    log_buffer.append(
-                        f"[2.5] Document tracking: {len(document_ids_added)} documents in {time.time() - step_start:.3f}s"
-                    )
-
-                # Store chunks and map to facts for all documents
-                step_start = time.time()
-                chunk_id_map_by_doc = {}  # Maps (doc_id, chunk_index) -> chunk_id
-
-                if chunks:
-                    # Group chunks by their source document
-                    chunks_by_doc = defaultdict(list)
-                    for chunk in chunks:
-                        # chunk.content_index tells us which content this chunk came from
-                        original_doc_id = contents_dicts[chunk.content_index].get("document_id")
-                        # Map to actual document_id (handles None -> generated UUID mapping)
-                        actual_doc_id = doc_id_mapping.get(original_doc_id, original_doc_id)
-                        if actual_doc_id is None and document_id:
-                            actual_doc_id = document_id
-                        chunks_by_doc[actual_doc_id].append(chunk)
-
-                    # Store chunks for each document
-                    for doc_id, doc_chunks in chunks_by_doc.items():
-                        chunk_id_map = await chunk_storage.store_chunks_batch(conn, bank_id, doc_id, doc_chunks)
-                        # Store mapping with document context
-                        for chunk_idx, chunk_id in chunk_id_map.items():
-                            chunk_id_map_by_doc[(doc_id, chunk_idx)] = chunk_id
-
-                    log_buffer.append(
-                        f"[3] Store chunks: {len(chunks)} chunks for {len(chunks_by_doc)} documents in {time.time() - step_start:.3f}s"
-                    )
-
-                    # Map chunk_ids and document_ids to facts
-                    for fact, processed_fact in zip(extracted_facts, processed_facts):
-                        # Get the original document_id for this fact's source content
-                        original_doc_id = contents_dicts[fact.content_index].get("document_id")
-                        # Map to actual document_id (handles None -> generated UUID mapping)
-                        actual_doc_id = doc_id_mapping.get(original_doc_id, original_doc_id)
-                        if actual_doc_id is None and document_id:
-                            actual_doc_id = document_id
-
-                        # Set document_id on the fact
-                        processed_fact.document_id = actual_doc_id
-
-                        # Map chunk_id if this fact came from a chunk
-                        if fact.chunk_index is not None:
-                            # Look up chunk_id using (doc_id, chunk_index)
-                            chunk_id = chunk_id_map_by_doc.get((actual_doc_id, fact.chunk_index))
-                            if chunk_id:
-                                processed_fact.chunk_id = chunk_id
-                else:
-                    # No chunks - still need to set document_id on facts
-                    for fact, processed_fact in zip(extracted_facts, processed_facts):
-                        original_doc_id = contents_dicts[fact.content_index].get("document_id")
-                        # Map to actual document_id (handles None -> generated UUID mapping)
-                        actual_doc_id = doc_id_mapping.get(original_doc_id, original_doc_id)
-                        if actual_doc_id is None and document_id:
-                            actual_doc_id = document_id
-                        processed_fact.document_id = actual_doc_id
-
-                non_duplicate_facts = processed_facts
-
-                # Insert facts (document_id is now stored per-fact)
-                step_start = time.time()
-                unit_ids = await fact_storage.insert_facts_batch(conn, bank_id, non_duplicate_facts)
-                log_buffer.append(f"[5] Insert facts: {len(unit_ids)} units in {time.time() - step_start:.3f}s")
-
-                # Process entities
-                step_start = time.time()
-                # Build map of content_index -> user entities for merging
-                user_entities_per_content = {
-                    idx: content.entities for idx, content in enumerate(contents) if content.entities
-                }
-                entity_links = await entity_processing.process_entities_batch(
-                    entity_resolver,
-                    conn,
-                    bank_id,
-                    unit_ids,
-                    non_duplicate_facts,
-                    log_buffer,
-                    user_entities_per_content=user_entities_per_content,
-                    entity_labels=getattr(config, "entity_labels", None),
-                )
-                log_buffer.append(f"[6] Process entities: {len(entity_links)} links in {time.time() - step_start:.3f}s")
-
-                # Create temporal links
-                step_start = time.time()
-                temporal_link_count = await link_creation.create_temporal_links_batch(conn, bank_id, unit_ids)
-                log_buffer.append(f"[7] Temporal links: {temporal_link_count} links in {time.time() - step_start:.3f}s")
-
-                # Create semantic links
-                step_start = time.time()
-                embeddings_for_links = [fact.embedding for fact in non_duplicate_facts]
-                semantic_link_count = await link_creation.create_semantic_links_batch(
-                    conn, bank_id, unit_ids, embeddings_for_links
-                )
-                log_buffer.append(f"[8] Semantic links: {semantic_link_count} links in {time.time() - step_start:.3f}s")
-
-                # Insert entity links
-                step_start = time.time()
-                if entity_links:
-                    await entity_processing.insert_entity_links_batch(conn, entity_links)
-                log_buffer.append(
-                    f"[9] Entity links: {len(entity_links) if entity_links else 0} links in {time.time() - step_start:.3f}s"
-                )
-
-                # Create causal links
-                step_start = time.time()
-                causal_link_count = await link_creation.create_causal_links_batch(conn, unit_ids, non_duplicate_facts)
-                log_buffer.append(f"[10] Causal links: {causal_link_count} links in {time.time() - step_start:.3f}s")
-
-                # Map results back to original content items
-                result_unit_ids = _map_results_to_contents(contents, extracted_facts, unit_ids)
-
-                # Transactional outbox: queue any side-effect tasks (e.g. webhook deliveries)
-                # inside the same transaction so they are atomically committed with the retain data.
-                if outbox_callback:
-                    await outbox_callback(conn)
-
-            # Flush entity stats (mention_count / last_seen) now that the transaction
-            # has committed.  Uses a fresh pool connection — no locks held.
-            await entity_resolver.flush_pending_stats()
-
-            # Log final summary
-            total_time = time.time() - start_time
-            log_buffer.append(f"{'=' * 60}")
-            log_buffer.append(f"RETAIN_BATCH COMPLETE: {len(unit_ids)} units in {total_time:.3f}s")
-            if document_ids_added:
-                log_buffer.append(f"Documents: {', '.join(document_ids_added)}")
-            log_buffer.append(f"{'=' * 60}")
-
-            logger.info("\n" + "\n".join(log_buffer) + "\n")
-
-    await retry_with_backoff(_run_db_work)
-    return result_unit_ids, usage
-
-
-def _map_results_to_contents(
-    contents: list[RetainContent],
-    extracted_facts: list[ExtractedFact],
-    unit_ids: list[str],
-) -> list[list[str]]:
-    """Map created unit IDs back to original content items."""
-    facts_by_content: dict[int, list[int]] = {i: [] for i in range(len(contents))}
-    for i, fact in enumerate(extracted_facts):
-        facts_by_content[fact.content_index].append(i)
-
-    result_unit_ids = []
-    unit_idx = 0
-    for content_index in range(len(contents)):
-        content_unit_ids = []
-        for _ in facts_by_content[content_index]:
-            content_unit_ids.append(unit_ids[unit_idx])
-            unit_idx += 1
-        result_unit_ids.append(content_unit_ids)
-
-    return result_unit_ids

hindsight-api-slim/hindsight_api/engine/search/link_expansion_retrieval.py

@@ -1,493 +0,0 @@
-"""
-Link Expansion graph retrieval.
-
-Expands from semantic/temporal seeds through three parallel, first-class signals
-stored in memory_links:
-
-1. Entity links  — precomputed co-occurrence graph (created at retain time, bounded to
-                   MAX_LINKS_PER_ENTITY per entity). Score = number of distinct shared
-                   entities between the seed set and each candidate.
-2. Semantic links — precomputed kNN graph (each new fact linked to its top-5 most
-                    similar existing facts at insert time, similarity >= 0.7). Checked
-                    in both directions since the graph is not symmetric. Score = weight.
-3. Causal links  — explicit causal chains (causes/caused_by/enables/prevents).
-                   Score = weight + 1.0 (boosted as highest-quality signal).
-
-All three signals are bounded at retain time, so no LATERAL fan-out caps are needed
-at query time. Each expansion is a simple aggregation over a small result set.
-
-For non-observation fact types the three expansions are issued as a single CTE query
-(one roundtrip, one connection) with a `source` discriminator column so the Python
-merge step can apply per-signal score transformations.
-"""
-
-import logging
-import math
-import time
-
-from ..db_utils import acquire_with_retry
-from ..memory_engine import fq_table
-from .graph_retrieval import GraphRetriever
-from .tags import TagGroup, TagsMatch, filter_results_by_tag_groups, filter_results_by_tags
-from .types import MPFPTimings, RetrievalResult
-
-logger = logging.getLogger(__name__)
-
-
-async def _find_semantic_seeds(
-    conn,
-    query_embedding_str: str,
-    bank_id: str,
-    fact_type: str,
-    limit: int = 20,
-    threshold: float = 0.3,
-    tags: list[str] | None = None,
-    tags_match: TagsMatch = "any",
-    tag_groups: list[TagGroup] | None = None,
-) -> list[RetrievalResult]:
-    """Find semantic seeds via embedding search."""
-    from .tags import build_tag_groups_where_clause, build_tags_where_clause_simple
-
-    tags_clause = build_tags_where_clause_simple(tags, 6, match=tags_match)
-    tag_groups_param_start = 6 + (1 if tags else 0)
-    groups_clause, groups_params, _ = build_tag_groups_where_clause(tag_groups, tag_groups_param_start)
-    params = [query_embedding_str, bank_id, fact_type, threshold, limit]
-    if tags:
-        params.append(tags)
-    params.extend(groups_params)
-
-    rows = await conn.fetch(
-        f"""
-        SELECT id, text, context, event_date, occurred_start, occurred_end,
-               mentioned_at, fact_type, document_id, chunk_id, tags,
-               1 - (embedding <=> $1::vector) AS similarity
-        FROM {fq_table("memory_units")}
-        WHERE bank_id = $2
-          AND embedding IS NOT NULL
-          AND fact_type = $3
-          AND (1 - (embedding <=> $1::vector)) >= $4
-          {tags_clause}
-          {groups_clause}
-        ORDER BY embedding <=> $1::vector
-        LIMIT $5
-        """,
-        *params,
-    )
-    return [RetrievalResult.from_db_row(dict(r)) for r in rows]
-
-
-class LinkExpansionRetriever(GraphRetriever):
-    """
-    Graph retrieval via direct link expansion from seeds.
-
-    Runs three expansions through precomputed memory_links: entity co-occurrence,
-    semantic kNN, and causal chains, all bounded at retain time.
-
-    For non-observation fact types the three expansions are issued as a single CTE
-    query (one roundtrip, one connection slot) with a `source` discriminator column.
-    The Python merge step applies per-signal score transformations.
-    """
-
-    def __init__(
-        self,
-        causal_weight_threshold: float = 0.3,
-    ):
-        """
-        Args:
-            causal_weight_threshold: Minimum weight for causal links to follow.
-        """
-        self.causal_weight_threshold = causal_weight_threshold
-
-    @property
-    def name(self) -> str:
-        return "link_expansion"
-
-    async def retrieve(
-        self,
-        pool,
-        query_embedding_str: str,
-        bank_id: str,
-        fact_type: str,
-        budget: int,
-        query_text: str | None = None,
-        semantic_seeds: list[RetrievalResult] | None = None,
-        temporal_seeds: list[RetrievalResult] | None = None,
-        adjacency=None,
-        tags: list[str] | None = None,
-        tags_match: TagsMatch = "any",
-        tag_groups: list[TagGroup] | None = None,
-    ) -> tuple[list[RetrievalResult], MPFPTimings | None]:
-        """
-        Retrieve facts by expanding links from seeds.
-
-        Args:
-            pool: Database connection pool
-            query_embedding_str: Query embedding as string
-            bank_id: Memory bank ID
-            fact_type: Fact type to filter
-            budget: Maximum results to return
-            query_text: Original query text (unused)
-            semantic_seeds: Pre-computed semantic entry points
-            temporal_seeds: Pre-computed temporal entry points
-            adjacency: Unused, kept for interface compatibility
-            tags: Optional list of tags for visibility filtering
-
-        Returns:
-            Tuple of (results, timings)
-        """
-        start_time = time.time()
-        timings = MPFPTimings(fact_type=fact_type)
-
-        async with acquire_with_retry(pool) as conn:
-            # Find seeds if not provided
-            if semantic_seeds:
-                all_seeds = list(semantic_seeds)
-            else:
-                seeds_start = time.time()
-                all_seeds = await _find_semantic_seeds(
-                    conn,
-                    query_embedding_str,
-                    bank_id,
-                    fact_type,
-                    limit=20,
-                    threshold=0.3,
-                    tags=tags,
-                    tags_match=tags_match,
-                    tag_groups=tag_groups,
-                )
-                timings.seeds_time = time.time() - seeds_start
-                logger.debug(
-                    f"[LinkExpansion] Found {len(all_seeds)} semantic seeds for fact_type={fact_type} "
-                    f"(tags={tags}, tags_match={tags_match})"
-                )
-
-            if temporal_seeds:
-                all_seeds.extend(temporal_seeds)
-
-            if not all_seeds:
-                return [], timings
-
-            seed_ids = list({s.id for s in all_seeds})
-            timings.pattern_count = len(seed_ids)
-
-            query_start = time.time()
-
-            if fact_type == "observation":
-                entity_rows, semantic_rows, causal_rows = await self._expand_observations(conn, seed_ids, budget)
-            else:
-                entity_rows, semantic_rows, causal_rows = await self._expand_combined(conn, seed_ids, fact_type, budget)
-
-            timings.edge_load_time = time.time() - query_start
-            timings.db_queries = 1
-            timings.edge_count = len(entity_rows) + len(semantic_rows) + len(causal_rows)
-
-        # Merge results with additive intra-score: entity + semantic + causal ∈ [0, 3].
-        #
-        # Entity score: tanh(count × 0.5) maps shared-entity count to [0, 1]:
-        #   1 entity → 0.46,  2 → 0.76,  3 → 0.91,  4 → 0.96  (saturates naturally)
-        # Semantic score: similarity weight, already ∈ [0.7, 1.0].
-        # Causal score:   link weight, already ∈ [0, 1].
-        #
-        # Facts appearing in multiple signals accumulate higher scores, rewarding
-        # convergent evidence. The outer RRF uses rank position from this sorted list.
-        entity_scores: dict[str, float] = {}
-        semantic_scores: dict[str, float] = {}
-        causal_scores: dict[str, float] = {}
-        row_map: dict[str, dict] = {}
-
-        for row in entity_rows:
-            fact_id = str(row["id"])
-            entity_scores[fact_id] = math.tanh(row["score"] * 0.5)
-            row_map[fact_id] = dict(row)
-
-        for row in semantic_rows:
-            fact_id = str(row["id"])
-            semantic_scores[fact_id] = max(semantic_scores.get(fact_id, 0.0), row["score"])
-            row_map.setdefault(fact_id, dict(row))
-
-        for row in causal_rows:
-            fact_id = str(row["id"])
-            causal_scores[fact_id] = max(causal_scores.get(fact_id, 0.0), row["score"])
-            row_map.setdefault(fact_id, dict(row))
-
-        all_ids = set(entity_scores) | set(semantic_scores) | set(causal_scores)
-        score_map = {
-            fid: entity_scores.get(fid, 0.0) + semantic_scores.get(fid, 0.0) + causal_scores.get(fid, 0.0)
-            for fid in all_ids
-        }
-
-        sorted_ids = sorted(score_map.keys(), key=lambda x: score_map[x], reverse=True)[:budget]
-        rows = [row_map[fact_id] for fact_id in sorted_ids]
-
-        results = []
-        for row in rows:
-            result = RetrievalResult.from_db_row(dict(row))
-            result.activation = row["score"]
-            results.append(result)
-
-        if tags:
-            results = filter_results_by_tags(results, tags, match=tags_match)
-
-        if tag_groups:
-            results = filter_results_by_tag_groups(results, tag_groups)
-
-        timings.result_count = len(results)
-        timings.traverse = time.time() - start_time
-
-        logger.debug(
-            f"LinkExpansion: {len(results)} results from {len(seed_ids)} seeds "
-            f"in {timings.traverse * 1000:.1f}ms (query: {timings.edge_load_time * 1000:.1f}ms)"
-        )
-
-        return results, timings
-
-    async def _expand_combined(
-        self,
-        conn,
-        seed_ids: list,
-        fact_type: str,
-        budget: int,
-    ) -> tuple[list, list, list]:
-        """
-        Single-roundtrip CTE query combining entity, semantic, and causal expansions.
-
-        Uses a `source` discriminator column so the caller can apply per-signal
-        score transformations.  The three CTEs share one connection slot — important
-        for asyncpg which does not allow concurrent queries on the same connection.
-
-        Index coverage (requires migration d2e3f4a5b6c7):
-          entity:   idx_memory_links_entity_covering (from_unit_id) INCLUDE (to_unit_id, entity_id)
-                    WHERE link_type = 'entity'  → index-only scan, no heap reads
-          semantic incoming:
-                    idx_memory_links_to_type_weight (to_unit_id, link_type, weight DESC)
-                    → replaces costly BitmapAnd of two separate scans
-        """
-        ml = fq_table("memory_links")
-        mu = fq_table("memory_units")
-        all_rows = await conn.fetch(
-            f"""
-            WITH entity_expanded AS (
-                -- Entity co-occurrence: seeds → their precomputed entity-link neighbors.
-                -- Score = distinct shared entities (bounded at retain time to
-                -- MAX_LINKS_PER_ENTITY=50).  GROUP BY mu.id is sufficient because mu.id
-                -- is the primary key and functionally determines all other mu columns.
-                SELECT
-                    mu.id, mu.text, mu.context, mu.event_date, mu.occurred_start,
-                    mu.occurred_end, mu.mentioned_at,
-                    mu.fact_type, mu.document_id, mu.chunk_id, mu.tags,
-                    COUNT(DISTINCT ml.entity_id)::float AS score,
-                    'entity'::text AS source
-                FROM {ml} ml
-                JOIN {mu} mu ON mu.id = ml.to_unit_id
-                WHERE ml.from_unit_id = ANY($1::uuid[])
-                  AND ml.link_type = 'entity'
-                  AND mu.fact_type = $2
-                  AND mu.id != ALL($1::uuid[])
-                GROUP BY mu.id
-                ORDER BY score DESC
-                LIMIT $3
-            ),
-            semantic_expanded AS (
-                -- Semantic kNN: both outgoing (seeds → their kNN at insert time) and
-                -- incoming (facts inserted after seeds that found seeds as kNN).
-                -- Score = max similarity weight across both directions.
-                SELECT
-                    id, text, context, event_date, occurred_start,
-                    occurred_end, mentioned_at,
-                    fact_type, document_id, chunk_id, tags,
-                    MAX(weight) AS score,
-                    'semantic'::text AS source
-                FROM (
-                    SELECT
-                        mu.id, mu.text, mu.context, mu.event_date, mu.occurred_start,
-                        mu.occurred_end, mu.mentioned_at,
-                        mu.fact_type, mu.document_id, mu.chunk_id, mu.tags,
-                        ml.weight
-                    FROM {ml} ml
-                    JOIN {mu} mu ON mu.id = ml.to_unit_id
-                    WHERE ml.from_unit_id = ANY($1::uuid[])
-                      AND ml.link_type = 'semantic'
-                      AND mu.fact_type = $2
-                      AND mu.id != ALL($1::uuid[])
-                    UNION ALL
-                    SELECT
-                        mu.id, mu.text, mu.context, mu.event_date, mu.occurred_start,
-                        mu.occurred_end, mu.mentioned_at,
-                        mu.fact_type, mu.document_id, mu.chunk_id, mu.tags,
-                        ml.weight
-                    FROM {ml} ml
-                    JOIN {mu} mu ON mu.id = ml.from_unit_id
-                    WHERE ml.to_unit_id = ANY($1::uuid[])
-                      AND ml.link_type = 'semantic'
-                      AND mu.fact_type = $2
-                      AND mu.id != ALL($1::uuid[])
-                ) sem_raw
-                GROUP BY id, text, context, event_date, occurred_start,
-                         occurred_end, mentioned_at,
-                         fact_type, document_id, chunk_id, tags
-                ORDER BY score DESC
-                LIMIT $3
-            ),
-            causal_expanded AS (
-                -- Causal chains: explicit causes/enables/prevents links from seeds.
-                -- DISTINCT ON handles the case where a seed has multiple causal links
-                -- to the same target; best weight wins.
-                SELECT DISTINCT ON (mu.id)
-                    mu.id, mu.text, mu.context, mu.event_date, mu.occurred_start,
-                    mu.occurred_end, mu.mentioned_at,
-                    mu.fact_type, mu.document_id, mu.chunk_id, mu.tags,
-                    ml.weight AS score,
-                    'causal'::text AS source
-                FROM {ml} ml
-                JOIN {mu} mu ON ml.to_unit_id = mu.id
-                WHERE ml.from_unit_id = ANY($1::uuid[])
-                  AND ml.link_type IN ('causes', 'caused_by', 'enables', 'prevents')
-                  AND ml.weight >= $4
-                  AND mu.fact_type = $2
-                ORDER BY mu.id, ml.weight DESC
-                LIMIT $3
-            )
-            SELECT * FROM entity_expanded
-            UNION ALL
-            SELECT * FROM semantic_expanded
-            UNION ALL
-            SELECT * FROM causal_expanded
-            """,
-            seed_ids,
-            fact_type,
-            budget,
-            self.causal_weight_threshold,
-        )
-
-        entity_rows = [r for r in all_rows if r["source"] == "entity"]
-        semantic_rows = [r for r in all_rows if r["source"] == "semantic"]
-        causal_rows = [r for r in all_rows if r["source"] == "causal"]
-        return entity_rows, semantic_rows, causal_rows
-
-    async def _expand_observations(
-        self,
-        conn,
-        seed_ids: list,
-        budget: int,
-    ) -> tuple[list, list, list]:
-        """
-        Observation-specific expansion.
-
-        Observations don't have direct entity links in memory_links (they're created
-        by consolidation, not retain).  Instead, traverse source_memory_ids → world
-        facts → entities → other world facts → their observations.
-
-        Semantic and causal expansions run as a second combined CTE query.
-        """
-        source_ids_found: list = []
-        if logger.isEnabledFor(logging.DEBUG):
-            debug_rows = await conn.fetch(
-                f"""
-                SELECT id, source_memory_ids
-                FROM {fq_table("memory_units")}
-                WHERE id = ANY($1::uuid[])
-                """,
-                seed_ids,
-            )
-            for row in debug_rows:
-                if row["source_memory_ids"]:
-                    source_ids_found.extend(row["source_memory_ids"])
-            logger.debug(
-                f"[LinkExpansion] observation graph: {len(seed_ids)} seeds, "
-                f"{len(source_ids_found)} source_memory_ids found"
-            )
-
-        entity_rows = await conn.fetch(
-            f"""
-            WITH seed_sources AS (
-                SELECT DISTINCT unnest(source_memory_ids) AS source_id
-                FROM {fq_table("memory_units")}
-                WHERE id = ANY($1::uuid[])
-                  AND source_memory_ids IS NOT NULL
-            ),
-            connected_sources AS (
-                -- Mirror the non-observation entity expansion: follow pre-bounded entity
-                -- links in memory_links (capped to MAX_LINKS_PER_ENTITY=50 at retain time).
-                -- Score = number of distinct shared entities, same as the non-obs path.
-                SELECT DISTINCT ml.to_unit_id AS source_id
-                FROM seed_sources ss
-                JOIN {fq_table("memory_links")} ml ON ml.from_unit_id = ss.source_id
-                WHERE ml.link_type = 'entity'
-            ),
-            connected_array AS (
-                SELECT array_agg(source_id) AS source_ids FROM connected_sources
-            )
-            SELECT
-                mu.id, mu.text, mu.context, mu.event_date, mu.occurred_start,
-                mu.occurred_end, mu.mentioned_at,
-                mu.fact_type, mu.document_id, mu.chunk_id, mu.tags,
-                (SELECT COUNT(DISTINCT s) FROM unnest(mu.source_memory_ids) s WHERE s = ANY(ca.source_ids))::float AS score
-            FROM {fq_table("memory_units")} mu, connected_array ca
-            WHERE mu.fact_type = 'observation'
-              AND mu.id != ALL($1::uuid[])
-              AND ca.source_ids IS NOT NULL
-              AND mu.source_memory_ids && ca.source_ids
-            ORDER BY score DESC
-            LIMIT $2
-            """,
-            seed_ids,
-            budget,
-        )
-        logger.debug(f"[LinkExpansion] observation graph: found {len(entity_rows)} connected observations")
-
-        # Semantic + causal for observations in one query
-        ml = fq_table("memory_links")
-        mu = fq_table("memory_units")
-        sem_causal_rows = await conn.fetch(
-            f"""
-            WITH semantic_expanded AS (
-                SELECT
-                    id, text, context, event_date, occurred_start,
-                    occurred_end, mentioned_at,
-                    fact_type, document_id, chunk_id, tags,
-                    MAX(weight) AS score,
-                    'semantic'::text AS source
-                FROM (
-                    SELECT mu.id, mu.text, mu.context, mu.event_date, mu.occurred_start,
-                           mu.occurred_end, mu.mentioned_at, mu.fact_type, mu.document_id,
-                           mu.chunk_id, mu.tags, ml.weight
-                    FROM {ml} ml JOIN {mu} mu ON mu.id = ml.to_unit_id
-                    WHERE ml.from_unit_id = ANY($1::uuid[])
-                      AND ml.link_type = 'semantic' AND mu.fact_type = 'observation'
-                      AND mu.id != ALL($1::uuid[])
-                    UNION ALL
-                    SELECT mu.id, mu.text, mu.context, mu.event_date, mu.occurred_start,
-                           mu.occurred_end, mu.mentioned_at, mu.fact_type, mu.document_id,
-                           mu.chunk_id, mu.tags, ml.weight
-                    FROM {ml} ml JOIN {mu} mu ON mu.id = ml.from_unit_id
-                    WHERE ml.to_unit_id = ANY($1::uuid[])
-                      AND ml.link_type = 'semantic' AND mu.fact_type = 'observation'
-                      AND mu.id != ALL($1::uuid[])
-                ) sem_raw
-                GROUP BY id, text, context, event_date, occurred_start, occurred_end,
-                         mentioned_at, fact_type, document_id, chunk_id, tags
-                ORDER BY score DESC LIMIT $2
-            ),
-            causal_expanded AS (
-                SELECT DISTINCT ON (mu.id)
-                    mu.id, mu.text, mu.context, mu.event_date, mu.occurred_start,
-                    mu.occurred_end, mu.mentioned_at, mu.fact_type, mu.document_id,
-                    mu.chunk_id, mu.tags, ml.weight AS score, 'causal'::text AS source
-                FROM {ml} ml JOIN {mu} mu ON ml.to_unit_id = mu.id
-                WHERE ml.from_unit_id = ANY($1::uuid[])
-                  AND ml.link_type IN ('causes', 'caused_by', 'enables', 'prevents')
-                  AND ml.weight >= $3 AND mu.fact_type = 'observation'
-                ORDER BY mu.id, ml.weight DESC LIMIT $2
-            )
-            SELECT * FROM semantic_expanded
-            UNION ALL
-            SELECT * FROM causal_expanded
-            """,
-            seed_ids,
-            budget,
-            self.causal_weight_threshold,
-        )
-
-        semantic_rows = [r for r in sem_causal_rows if r["source"] == "semantic"]
-        causal_rows = [r for r in sem_causal_rows if r["source"] == "causal"]
-        return entity_rows, semantic_rows, causal_rows

hindsight-api-slim/hindsight_api/engine/search/tags.py

@@ -1,390 +0,0 @@
-"""
-Tags filtering utilities for retrieval.
-
-Provides SQL building functions for filtering memories by tags.
-Supports four matching modes via TagsMatch enum:
-- "any": OR matching, includes untagged memories (default, backward compatible)
-- "all": AND matching, includes untagged memories
-- "any_strict": OR matching, excludes untagged memories
-- "all_strict": AND matching, excludes untagged memories
-
-OR matching (any/any_strict): Memory matches if ANY of its tags overlap with request tags
-AND matching (all/all_strict): Memory matches if ALL request tags are present in its tags
-"""
-
-from __future__ import annotations
-
-from typing import Annotated, Literal
-
-from pydantic import BaseModel, ConfigDict, Field
-
-TagsMatch = Literal["any", "all", "any_strict", "all_strict"]
-
-
-def _parse_tags_match(match: TagsMatch) -> tuple[str, bool]:
-    """
-    Parse TagsMatch into operator and include_untagged flag.
-
-    Returns:
-        Tuple of (operator, include_untagged)
-        - operator: "&&" for any/any_strict, "@>" for all/all_strict
-        - include_untagged: True for any/all, False for any_strict/all_strict
-    """
-    if match == "any":
-        return "&&", True
-    elif match == "all":
-        return "@>", True
-    elif match == "any_strict":
-        return "&&", False
-    elif match == "all_strict":
-        return "@>", False
-    else:
-        # Default to "any" behavior
-        return "&&", True
-
-
-def build_tags_where_clause(
-    tags: list[str] | None,
-    param_offset: int = 1,
-    table_alias: str = "",
-    match: TagsMatch = "any",
-) -> tuple[str, list, int]:
-    """
-    Build a SQL WHERE clause for filtering by tags.
-
-    Supports four matching modes:
-    - "any" (default): OR matching, includes untagged memories
-    - "all": AND matching, includes untagged memories
-    - "any_strict": OR matching, excludes untagged memories
-    - "all_strict": AND matching, excludes untagged memories
-
-    Args:
-        tags: List of tags to filter by. If None or empty, returns empty clause (no filtering).
-        param_offset: Starting parameter number for SQL placeholders (default 1).
-        table_alias: Optional table alias prefix (e.g., "mu." for "memory_units mu").
-        match: Matching mode. Defaults to "any".
-
-    Returns:
-        Tuple of (sql_clause, params, next_param_offset):
-        - sql_clause: SQL WHERE clause string
-        - params: List of parameter values to bind
-        - next_param_offset: Next available parameter number
-
-    Example:
-        >>> clause, params, next_offset = build_tags_where_clause(['user_a'], 3, 'mu.', 'any_strict')
-        >>> print(clause)  # "AND mu.tags IS NOT NULL AND mu.tags != '{}' AND mu.tags && $3"
-    """
-    if not tags:
-        return "", [], param_offset
-
-    column = f"{table_alias}tags" if table_alias else "tags"
-    operator, include_untagged = _parse_tags_match(match)
-
-    if include_untagged:
-        # Include untagged memories (NULL or empty array) OR matching tags
-        clause = f"AND ({column} IS NULL OR {column} = '{{}}' OR {column} {operator} ${param_offset})"
-    else:
-        # Strict: only memories with matching tags (exclude NULL and empty)
-        clause = f"AND {column} IS NOT NULL AND {column} != '{{}}' AND {column} {operator} ${param_offset}"
-
-    return clause, [tags], param_offset + 1
-
-
-def build_tags_where_clause_simple(
-    tags: list[str] | None,
-    param_num: int,
-    table_alias: str = "",
-    match: TagsMatch = "any",
-) -> str:
-    """
-    Build a simple SQL WHERE clause for tags filtering.
-
-    This is a convenience version that returns just the clause string,
-    assuming the caller will add the tags array to their params list.
-
-    Args:
-        tags: List of tags to filter by. If None or empty, returns empty string.
-        param_num: Parameter number to use in the clause.
-        table_alias: Optional table alias prefix.
-        match: Matching mode. Defaults to "any".
-
-    Returns:
-        SQL clause string or empty string.
-    """
-    if not tags:
-        return ""
-
-    column = f"{table_alias}tags" if table_alias else "tags"
-    operator, include_untagged = _parse_tags_match(match)
-
-    if include_untagged:
-        # Include untagged memories (NULL or empty array) OR matching tags
-        return f"AND ({column} IS NULL OR {column} = '{{}}' OR {column} {operator} ${param_num})"
-    else:
-        # Strict: only memories with matching tags (exclude NULL and empty)
-        return f"AND {column} IS NOT NULL AND {column} != '{{}}' AND {column} {operator} ${param_num}"
-
-
-def filter_results_by_tags(
-    results: list,
-    tags: list[str] | None,
-    match: TagsMatch = "any",
-) -> list:
-    """
-    Filter retrieval results by tags in Python (for post-processing).
-
-    Used when SQL filtering isn't possible (e.g., graph traversal results).
-
-    Args:
-        results: List of RetrievalResult objects with a 'tags' attribute.
-        tags: List of tags to filter by. If None or empty, returns all results.
-        match: Matching mode. Defaults to "any".
-
-    Returns:
-        Filtered list of results.
-    """
-    if not tags:
-        return results
-
-    _, include_untagged = _parse_tags_match(match)
-    is_any_match = match in ("any", "any_strict")
-
-    tags_set = set(tags)
-    filtered = []
-
-    for result in results:
-        result_tags = getattr(result, "tags", None)
-
-        # Check if untagged
-        is_untagged = result_tags is None or len(result_tags) == 0
-
-        if is_untagged:
-            if include_untagged:
-                filtered.append(result)
-            # else: skip untagged
-        else:
-            result_tags_set = set(result_tags)
-            if is_any_match:
-                # Any overlap
-                if result_tags_set & tags_set:
-                    filtered.append(result)
-            else:
-                # All tags must be present
-                if tags_set <= result_tags_set:
-                    filtered.append(result)
-
-    return filtered
-
-
-# =============================================================================
-# Compound tag group models (recursive boolean expressions)
-# =============================================================================
-
-
-class TagGroupLeaf(BaseModel):
-    """A leaf tag filter: matches memories by tag list and match mode."""
-
-    tags: list[str]
-    match: TagsMatch = "any_strict"
-
-
-class TagGroupAnd(BaseModel):
-    """Compound AND group: all child filters must match."""
-
-    model_config = ConfigDict(populate_by_name=True)
-    filters: list[TagGroup] = Field(alias="and")
-
-
-class TagGroupOr(BaseModel):
-    """Compound OR group: at least one child filter must match."""
-
-    model_config = ConfigDict(populate_by_name=True)
-    filters: list[TagGroup] = Field(alias="or")
-
-
-class TagGroupNot(BaseModel):
-    """Compound NOT group: child filter must NOT match."""
-
-    model_config = ConfigDict(populate_by_name=True)
-    filter: TagGroup = Field(alias="not")
-
-
-# TagGroup is a discriminated union; Pydantic will try left-to-right.
-# TagGroupLeaf is identified by the presence of 'tags'.
-# TagGroupAnd / TagGroupOr / TagGroupNot are compound (no 'tags' key).
-TagGroup = Annotated[
-    TagGroupLeaf | TagGroupAnd | TagGroupOr | TagGroupNot,
-    Field(union_mode="left_to_right"),
-]
-
-# Rebuild forward-reference models so recursive TagGroup is resolved.
-TagGroupAnd.model_rebuild()
-TagGroupOr.model_rebuild()
-TagGroupNot.model_rebuild()
-
-
-# =============================================================================
-# SQL builder for compound tag groups
-# =============================================================================
-
-
-def _build_group_clause(
-    group: TagGroup,
-    param_offset: int,
-    table_alias: str,
-) -> tuple[str, list, int]:
-    """
-    Recursively build an inner SQL clause (no leading AND/OR) for a single TagGroup.
-
-    Returns:
-        (inner_clause, params, next_param_offset)
-    """
-    if isinstance(group, TagGroupLeaf):
-        column = f"{table_alias}tags" if table_alias else "tags"
-        operator, include_untagged = _parse_tags_match(group.match)
-        if include_untagged:
-            clause = f"({column} IS NULL OR {column} = '{{}}' OR {column} {operator} ${param_offset})"
-        else:
-            clause = f"({column} IS NOT NULL AND {column} != '{{}}' AND {column} {operator} ${param_offset})"
-        return clause, [group.tags], param_offset + 1
-
-    elif isinstance(group, TagGroupAnd):
-        parts = []
-        params: list = []
-        offset = param_offset
-        for child in group.filters:
-            child_clause, child_params, offset = _build_group_clause(child, offset, table_alias)
-            parts.append(child_clause)
-            params.extend(child_params)
-        inner = " AND ".join(parts)
-        return f"({inner})", params, offset
-
-    elif isinstance(group, TagGroupOr):
-        parts = []
-        params = []
-        offset = param_offset
-        for child in group.filters:
-            child_clause, child_params, offset = _build_group_clause(child, offset, table_alias)
-            parts.append(child_clause)
-            params.extend(child_params)
-        inner = " OR ".join(parts)
-        return f"({inner})", params, offset
-
-    elif isinstance(group, TagGroupNot):
-        child_clause, child_params, next_offset = _build_group_clause(group.filter, param_offset, table_alias)
-        return f"NOT {child_clause}", child_params, next_offset
-
-    else:
-        # Should never happen with proper Pydantic validation
-        return "", [], param_offset
-
-
-def build_tag_groups_where_clause(
-    tag_groups: list[TagGroup] | None,
-    param_offset: int,
-    table_alias: str = "",
-) -> tuple[str, list, int]:
-    """
-    Build a SQL WHERE clause for compound tag group filtering.
-
-    Top-level groups are AND-ed together. Each group is a recursive boolean
-    expression (leaf, and, or, not).
-
-    Args:
-        tag_groups: List of TagGroup objects. If None or empty, returns empty clause.
-        param_offset: Starting parameter number for SQL placeholders.
-        table_alias: Optional table alias prefix (e.g., "mu." for "memory_units mu").
-
-    Returns:
-        Tuple of (sql_clause, params, next_param_offset):
-        - sql_clause: SQL WHERE clause string starting with "AND" (or empty string)
-        - params: List of parameter values to bind (one per leaf node)
-        - next_param_offset: Next available parameter number
-
-    Example:
-        >>> groups = [TagGroupLeaf(tags=["user:alice"], match="all_strict")]
-        >>> clause, params, next_offset = build_tag_groups_where_clause(groups, 3)
-        >>> print(clause)  # "AND (tags IS NOT NULL AND tags != '{}' AND tags @> $3)"
-    """
-    if not tag_groups:
-        return "", [], param_offset
-
-    all_params: list = []
-    all_clauses: list[str] = []
-    offset = param_offset
-
-    for group in tag_groups:
-        inner_clause, group_params, offset = _build_group_clause(group, offset, table_alias)
-        all_clauses.append(inner_clause)
-        all_params.extend(group_params)
-
-    combined = " AND ".join(all_clauses)
-    return f"AND {combined}", all_params, offset
-
-
-# =============================================================================
-# Python-side filter for compound tag groups (post-retrieval filtering)
-# =============================================================================
-
-
-def _match_group(result: object, group: TagGroup) -> bool:
-    """
-    Recursively evaluate a TagGroup against a retrieval result.
-
-    Args:
-        result: Any object with a 'tags' attribute (list[str] or None).
-        group: The TagGroup to evaluate.
-
-    Returns:
-        True if the result matches the group, False otherwise.
-    """
-    if isinstance(group, TagGroupLeaf):
-        result_tags = getattr(result, "tags", None)
-        is_untagged = result_tags is None or len(result_tags) == 0
-        _, include_untagged = _parse_tags_match(group.match)
-        is_any_match = group.match in ("any", "any_strict")
-        tags_set = set(group.tags)
-
-        if is_untagged:
-            return include_untagged
-        else:
-            result_tags_set = set(result_tags)
-            if is_any_match:
-                return bool(result_tags_set & tags_set)
-            else:
-                return tags_set <= result_tags_set
-
-    elif isinstance(group, TagGroupAnd):
-        return all(_match_group(result, child) for child in group.filters)
-
-    elif isinstance(group, TagGroupOr):
-        return any(_match_group(result, child) for child in group.filters)
-
-    elif isinstance(group, TagGroupNot):
-        return not _match_group(result, group.filter)
-
-    else:
-        return True
-
-
-def filter_results_by_tag_groups(
-    results: list,
-    tag_groups: list[TagGroup] | None,
-) -> list:
-    """
-    Filter retrieval results by compound tag groups in Python (for post-processing).
-
-    Used when SQL filtering isn't possible (e.g., graph traversal results).
-    Top-level groups are AND-ed together.
-
-    Args:
-        results: List of RetrievalResult objects with a 'tags' attribute.
-        tag_groups: List of TagGroup objects. If None or empty, returns all results.
-
-    Returns:
-        Filtered list of results where ALL top-level groups match.
-    """
-    if not tag_groups:
-        return results
-
-    return [r for r in results if all(_match_group(r, group) for group in tag_groups)]

hindsight-api-slim/hindsight_api/engine/storage/__init__.py

@@ -1,79 +0,0 @@
-"""File storage backends for uploaded files."""
-
-from collections.abc import Callable
-
-from .base import FileStorage
-from .postgresql import PostgreSQLFileStorage
-
-__all__ = ["FileStorage", "PostgreSQLFileStorage", "create_file_storage"]
-
-
-def create_file_storage(
-    storage_type: str,
-    pool_getter: Callable | None = None,
-    schema: str | None = None,
-    schema_getter: Callable | None = None,
-    **kwargs,
-) -> FileStorage:
-    """
-    Create file storage backend based on configuration.
-
-    Args:
-        storage_type: "native" (PostgreSQL BYTEA) or "s3" (S3-compatible object storage)
-        pool_getter: Database pool getter (required for native)
-        schema: Static database schema (for native single-tenant)
-        schema_getter: Callable returning current schema at query time (for native multi-tenant)
-        **kwargs: Additional args passed to storage backend
-
-    Returns:
-        FileStorage instance
-
-    Raises:
-        ValueError: If storage_type is unknown or required args are missing
-    """
-    if storage_type == "native":
-        if not pool_getter:
-            raise ValueError("pool_getter required for native (PostgreSQL) storage")
-        return PostgreSQLFileStorage(pool_getter=pool_getter, schema=schema, schema_getter=schema_getter)
-    elif storage_type == "s3":
-        from ...config import get_config
-        from .s3 import S3FileStorage
-
-        config = get_config()
-        bucket = config.file_storage_s3_bucket
-        if not bucket:
-            raise ValueError("HINDSIGHT_API_FILE_STORAGE_S3_BUCKET is required for S3 storage")
-        return S3FileStorage(
-            bucket=bucket,
-            region=config.file_storage_s3_region,
-            endpoint=config.file_storage_s3_endpoint,
-            access_key_id=config.file_storage_s3_access_key_id,
-            secret_access_key=config.file_storage_s3_secret_access_key,
-        )
-    elif storage_type == "gcs":
-        from ...config import get_config
-        from .gcs import GCSFileStorage
-
-        config = get_config()
-        bucket = config.file_storage_gcs_bucket
-        if not bucket:
-            raise ValueError("HINDSIGHT_API_FILE_STORAGE_GCS_BUCKET is required for GCS storage")
-        return GCSFileStorage(
-            bucket=bucket,
-            service_account_key=config.file_storage_gcs_service_account_key,
-        )
-    elif storage_type == "azure":
-        from ...config import get_config
-        from .azure import AzureFileStorage
-
-        config = get_config()
-        container = config.file_storage_azure_container
-        if not container:
-            raise ValueError("HINDSIGHT_API_FILE_STORAGE_AZURE_CONTAINER is required for Azure storage")
-        return AzureFileStorage(
-            container_name=container,
-            account_name=config.file_storage_azure_account_name,
-            account_key=config.file_storage_azure_account_key,
-        )
-    else:
-        raise ValueError(f"Unknown storage type: {storage_type}. Supported: 'native', 's3', 'gcs', 'azure'.")

hindsight-api-slim/hindsight_api/engine/storage/azure.py

@@ -1,62 +0,0 @@
-"""Azure Blob Storage backend using obstore."""
-
-import logging
-from datetime import timedelta
-
-import obstore as obs
-from obstore.store import AzureStore
-
-from .base import FileStorage
-
-logger = logging.getLogger(__name__)
-
-
-class AzureFileStorage(FileStorage):
-    """
-    Azure Blob Storage backend.
-
-    Uses obstore (Rust-backed) for high-throughput async access to Azure Blob Storage.
-    Supports account key, SAS token, and default Azure credentials.
-    """
-
-    def __init__(
-        self,
-        container_name: str,
-        account_name: str | None = None,
-        account_key: str | None = None,
-    ):
-        kwargs: dict = {}
-        if account_name:
-            kwargs["account_name"] = account_name
-        if account_key:
-            kwargs["account_key"] = account_key
-
-        self._store = AzureStore(container_name, **kwargs)
-        logger.info(f"Initialized Azure file storage: container={container_name}, account={account_name}")
-
-    async def store(self, file_data: bytes, key: str, metadata: dict[str, str] | None = None) -> str:
-        await obs.put_async(self._store, key, file_data)
-        logger.debug(f"Stored file {key} ({len(file_data)} bytes) in Azure")
-        return key
-
-    async def retrieve(self, key: str) -> bytes:
-        try:
-            response = await obs.get_async(self._store, key)
-            return await response.bytes_async()
-        except Exception as e:
-            if "not found" in str(e).lower() or "BlobNotFound" in str(e):
-                raise FileNotFoundError(f"File not found: {key}") from e
-            raise
-
-    async def delete(self, key: str) -> None:
-        await obs.delete_async(self._store, key)
-
-    async def exists(self, key: str) -> bool:
-        try:
-            await obs.head_async(self._store, key)
-            return True
-        except Exception:
-            return False
-
-    async def get_download_url(self, key: str, expires_in: int = 3600) -> str:
-        return await obs.sign_async(self._store, "GET", key, timedelta(seconds=expires_in))

hindsight-api-slim/hindsight_api/engine/storage/base.py

@@ -1,83 +0,0 @@
-"""Abstract base class for file storage backends."""
-
-from abc import ABC, abstractmethod
-
-
-class FileStorage(ABC):
-    """Abstract base for file storage backends."""
-
-    @abstractmethod
-    async def store(
-        self,
-        file_data: bytes,
-        key: str,
-        metadata: dict[str, str] | None = None,
-    ) -> str:
-        """
-        Store file and return storage key.
-
-        Args:
-            file_data: Raw file bytes
-            key: Storage key (e.g., "banks/{bank_id}/files/{file_id}.pdf")
-            metadata: Optional metadata to store with file
-
-        Returns:
-            Storage key that can be used to retrieve the file
-        """
-        pass
-
-    @abstractmethod
-    async def retrieve(self, key: str) -> bytes:
-        """
-        Retrieve file by storage key.
-
-        Args:
-            key: Storage key
-
-        Returns:
-            File data as bytes
-
-        Raises:
-            FileNotFoundError: If file does not exist
-        """
-        pass
-
-    @abstractmethod
-    async def delete(self, key: str) -> None:
-        """
-        Delete file by storage key.
-
-        Args:
-            key: Storage key
-        """
-        pass
-
-    @abstractmethod
-    async def exists(self, key: str) -> bool:
-        """
-        Check if file exists.
-
-        Args:
-            key: Storage key
-
-        Returns:
-            True if file exists, False otherwise
-        """
-        pass
-
-    @abstractmethod
-    async def get_download_url(self, key: str, expires_in: int = 3600) -> str:
-        """
-        Get a URL for downloading the file.
-
-        For PostgreSQL storage, this might be a relative API path.
-        For S3, this would be a pre-signed URL.
-
-        Args:
-            key: Storage key
-            expires_in: Expiration time in seconds (may be ignored for some backends)
-
-        Returns:
-            Download URL or path
-        """
-        pass

hindsight-api-slim/hindsight_api/engine/storage/gcs.py

@@ -1,105 +0,0 @@
-"""Google Cloud Storage backend using obstore."""
-
-import logging
-import os
-from datetime import datetime, timedelta, timezone
-
-import obstore as obs
-from obstore.store import GCSStore
-
-from .base import FileStorage
-
-logger = logging.getLogger(__name__)
-
-
-def _make_google_auth_credential_provider():
-    """Create a credential provider using google.auth (supports all credential types).
-
-    obstore's built-in credential parsing only supports service_account and
-    authorized_user JSON types.  This provider uses the google-auth library
-    which additionally handles external_account (Workload Identity Federation),
-    impersonated credentials, and metadata-server credentials.
-    """
-    import google.auth
-    import google.auth.transport.requests
-
-    credentials, _ = google.auth.default(scopes=["https://www.googleapis.com/auth/cloud-platform"])
-    request = google.auth.transport.requests.Request()
-
-    def _provide():
-        credentials.refresh(request)
-        expiry = credentials.expiry
-        if expiry and expiry.tzinfo is None:
-            expiry = expiry.replace(tzinfo=timezone.utc)
-        return {"token": credentials.token, "expires_at": expiry}
-
-    return _provide
-
-
-class GCSFileStorage(FileStorage):
-    """
-    Google Cloud Storage backend.
-
-    Uses obstore (Rust-backed) for high-throughput async access to GCS.
-    Supports Application Default Credentials, service account keys, and explicit credentials.
-    """
-
-    def __init__(
-        self,
-        bucket: str,
-        service_account_key: str | None = None,
-    ):
-        kwargs: dict = {}
-        if service_account_key:
-            kwargs["service_account_key"] = service_account_key
-        else:
-            # Use google.auth credential provider for broad credential type support
-            # (service_account, authorized_user, external_account, metadata server, etc.)
-            try:
-                kwargs["credential_provider"] = _make_google_auth_credential_provider()
-                logger.info("Using google.auth credential provider for GCS")
-            except Exception as e:
-                logger.warning(
-                    f"Failed to create google.auth credential provider, falling back to obstore defaults: {e}"
-                )
-
-        # Workaround for https://github.com/developmentseed/obstore/issues/605
-        # obstore's Rust layer doesn't support external_account credentials (Workload
-        # Identity Federation) and eagerly parses GOOGLE_APPLICATION_CREDENTIALS even
-        # when credential_provider is given. Per the obstore maintainer's guidance,
-        # remove env vars so the Rust code doesn't try to authenticate itself.
-        # google.auth (used by credential_provider above) has already loaded credentials.
-        gac = os.environ.pop("GOOGLE_APPLICATION_CREDENTIALS", None)
-        try:
-            self._store = GCSStore(bucket, **kwargs)
-        finally:
-            if gac is not None:
-                os.environ["GOOGLE_APPLICATION_CREDENTIALS"] = gac
-        logger.info(f"Initialized GCS file storage: bucket={bucket}")
-
-    async def store(self, file_data: bytes, key: str, metadata: dict[str, str] | None = None) -> str:
-        await obs.put_async(self._store, key, file_data)
-        logger.debug(f"Stored file {key} ({len(file_data)} bytes) in GCS")
-        return key
-
-    async def retrieve(self, key: str) -> bytes:
-        try:
-            response = await obs.get_async(self._store, key)
-            return await response.bytes_async()
-        except Exception as e:
-            if "not found" in str(e).lower():
-                raise FileNotFoundError(f"File not found: {key}") from e
-            raise
-
-    async def delete(self, key: str) -> None:
-        await obs.delete_async(self._store, key)
-
-    async def exists(self, key: str) -> bool:
-        try:
-            await obs.head_async(self._store, key)
-            return True
-        except Exception:
-            return False
-
-    async def get_download_url(self, key: str, expires_in: int = 3600) -> str:
-        return await obs.sign_async(self._store, "GET", key, timedelta(seconds=expires_in))

hindsight-api-slim/hindsight_api/engine/storage/postgresql.py

@@ -1,153 +0,0 @@
-"""PostgreSQL BYTEA-based file storage (default, zero-config)."""
-
-import logging
-from collections.abc import Callable
-from typing import TYPE_CHECKING
-
-if TYPE_CHECKING:
-    import asyncpg
-
-from .base import FileStorage
-
-logger = logging.getLogger(__name__)
-
-
-def fq_table(table: str, schema: str | None = None) -> str:
-    """Get fully-qualified table name with optional schema prefix."""
-    if schema:
-        return f'"{schema}".{table}'
-    return table
-
-
-class PostgreSQLFileStorage(FileStorage):
-    """
-    PostgreSQL BYTEA-based file storage.
-
-    Stores files directly in PostgreSQL using BYTEA columns.
-    This is the default storage backend - zero configuration required!
-
-    Pros:
-    - Works out of the box (no external dependencies)
-    - Transactional consistency with database
-    - Simple backups (included in pg_dump)
-    - Good performance for <10MB files
-
-    Cons:
-    - Database bloat for large/many files
-    - Not ideal for distributed deployments
-    - Higher cost than object storage at scale
-
-    For production/scale, consider S3FileStorage instead.
-    """
-
-    def __init__(
-        self,
-        pool_getter: Callable[[], "asyncpg.Pool"],
-        schema: str | None = None,
-        schema_getter: Callable[[], str] | None = None,
-    ):
-        """
-        Initialize PostgreSQL file storage.
-
-        Args:
-            pool_getter: Function that returns asyncpg connection pool
-            schema: Static database schema (fallback for single-tenant / tests)
-            schema_getter: Callable returning current schema at query time (for multi-tenant)
-        """
-        self._pool_getter = pool_getter
-        self._static_schema = schema
-        self._schema_getter = schema_getter
-
-    @property
-    def _schema(self) -> str | None:
-        """Resolve schema dynamically per-request when schema_getter is provided."""
-        if self._schema_getter:
-            return self._schema_getter()
-        return self._static_schema
-
-    async def store(
-        self,
-        file_data: bytes,
-        key: str,
-        metadata: dict[str, str] | None = None,
-    ) -> str:
-        """Store file in PostgreSQL."""
-        pool = self._pool_getter()
-
-        async with pool.acquire() as conn:
-            await conn.execute(
-                f"""
-                INSERT INTO {fq_table("file_storage", self._schema)}
-                (storage_key, data)
-                VALUES ($1, $2)
-                ON CONFLICT (storage_key) DO UPDATE SET
-                    data = EXCLUDED.data
-                """,
-                key,
-                file_data,
-            )
-
-        logger.debug(f"Stored file {key} ({len(file_data)} bytes) in PostgreSQL")
-        return key
-
-    async def retrieve(self, key: str) -> bytes:
-        """Retrieve file from PostgreSQL."""
-        pool = self._pool_getter()
-
-        async with pool.acquire() as conn:
-            row = await conn.fetchrow(
-                f"""
-                SELECT data FROM {fq_table("file_storage", self._schema)}
-                WHERE storage_key = $1
-                """,
-                key,
-            )
-
-            if not row:
-                raise FileNotFoundError(f"File not found: {key}")
-
-            return bytes(row["data"])
-
-    async def delete(self, key: str) -> None:
-        """Delete file from PostgreSQL."""
-        pool = self._pool_getter()
-
-        async with pool.acquire() as conn:
-            result = await conn.execute(
-                f"""
-                DELETE FROM {fq_table("file_storage", self._schema)}
-                WHERE storage_key = $1
-                """,
-                key,
-            )
-
-            # Check if anything was deleted
-            if result == "DELETE 0":
-                logger.warning(f"Attempted to delete non-existent file: {key}")
-
-    async def exists(self, key: str) -> bool:
-        """Check if file exists in PostgreSQL."""
-        pool = self._pool_getter()
-
-        async with pool.acquire() as conn:
-            row = await conn.fetchrow(
-                f"""
-                SELECT 1 FROM {fq_table("file_storage", self._schema)}
-                WHERE storage_key = $1
-                """,
-                key,
-            )
-
-            return row is not None
-
-    async def get_download_url(self, key: str, expires_in: int = 3600) -> str:
-        """
-        Get download URL for PostgreSQL-stored file.
-
-        Returns an API endpoint path (not a pre-signed URL since the file
-        is stored in the database). The expires_in parameter is ignored
-        for PostgreSQL storage.
-        """
-        # Return API path for download endpoint
-        # (expires_in ignored for database storage - auth handled at API level)
-        return f"/v1/default/files/download/{key}"

hindsight-api-slim/hindsight_api/engine/storage/s3.py

@@ -1,71 +0,0 @@
-"""S3 object storage backend using obstore."""
-
-import logging
-from datetime import timedelta
-
-import obstore as obs
-from obstore.store import S3Store
-
-from .base import FileStorage
-
-logger = logging.getLogger(__name__)
-
-
-class S3FileStorage(FileStorage):
-    """
-    S3-compatible object storage backend.
-
-    Uses obstore (Rust-backed) for high-throughput async access to
-    Amazon S3, MinIO, Cloudflare R2, and other S3-compliant APIs.
-    """
-
-    def __init__(
-        self,
-        bucket: str,
-        region: str | None = None,
-        endpoint: str | None = None,
-        access_key_id: str | None = None,
-        secret_access_key: str | None = None,
-    ):
-        kwargs: dict = {}
-        if region:
-            kwargs["region"] = region
-        if endpoint:
-            kwargs["endpoint"] = endpoint
-            # Allow plain HTTP for local S3-compatible services (MinIO, LocalStack, etc.)
-            if endpoint.startswith("http://"):
-                kwargs["allow_http"] = True
-        if access_key_id:
-            kwargs["access_key_id"] = access_key_id
-        if secret_access_key:
-            kwargs["secret_access_key"] = secret_access_key
-
-        self._store = S3Store(bucket, **kwargs)
-        logger.info(f"Initialized S3 file storage: bucket={bucket}, region={region}, endpoint={endpoint}")
-
-    async def store(self, file_data: bytes, key: str, metadata: dict[str, str] | None = None) -> str:
-        await obs.put_async(self._store, key, file_data)
-        logger.debug(f"Stored file {key} ({len(file_data)} bytes) in S3")
-        return key
-
-    async def retrieve(self, key: str) -> bytes:
-        try:
-            response = await obs.get_async(self._store, key)
-            return await response.bytes_async()
-        except Exception as e:
-            if "not found" in str(e).lower() or "NoSuchKey" in str(e):
-                raise FileNotFoundError(f"File not found: {key}") from e
-            raise
-
-    async def delete(self, key: str) -> None:
-        await obs.delete_async(self._store, key)
-
-    async def exists(self, key: str) -> bool:
-        try:
-            await obs.head_async(self._store, key)
-            return True
-        except Exception:
-            return False
-
-    async def get_download_url(self, key: str, expires_in: int = 3600) -> str:
-        return await obs.sign_async(self._store, "GET", key, timedelta(seconds=expires_in))

hindsight-api-slim/hindsight_api/mcp_local.py

@@ -1,40 +0,0 @@
-"""
-Local MCP server entry point for use with Claude Code (HTTP transport).
-
-This is a thin wrapper around the main hindsight-api server that pre-configures
-sensible defaults for local use (embedded PostgreSQL via pg0, warning log level).
-
-The full API runs on localhost:8888. Configure Claude Code's MCP settings:
-    claude mcp add --transport http hindsight http://localhost:8888/mcp/
-
-Or pinned to a specific bank (single-bank mode):
-    claude mcp add --transport http hindsight http://localhost:8888/mcp/default/
-
-Run with:
-    hindsight-local-mcp
-
-Or with uvx:
-    uvx hindsight-api@latest hindsight-local-mcp
-
-Environment variables:
-    HINDSIGHT_API_LLM_API_KEY: Required. API key for LLM provider.
-    HINDSIGHT_API_LLM_PROVIDER: Optional. LLM provider (default: "openai").
-    HINDSIGHT_API_LLM_MODEL: Optional. LLM model (default: "gpt-4o-mini").
-    HINDSIGHT_API_DATABASE_URL: Optional. Override database URL (default: pg0://hindsight-mcp).
-"""
-
-import os
-
-
-def main() -> None:
-    """Start the Hindsight API server with local defaults."""
-    # Set local defaults (only if not already configured by the user)
-    os.environ.setdefault("HINDSIGHT_API_DATABASE_URL", "pg0://hindsight-mcp")
-
-    from hindsight_api.main import main as api_main
-
-    api_main()
-
-
-if __name__ == "__main__":
-    main()

hindsight-api-slim/hindsight_api/webhooks/__init__.py

@@ -1,13 +0,0 @@
-"""Webhook system for Hindsight API event notifications."""
-
-from .manager import WebhookManager
-from .models import ConsolidationEventData, RetainEventData, WebhookConfig, WebhookEvent, WebhookEventType
-
-__all__ = [
-    "WebhookManager",
-    "WebhookConfig",
-    "WebhookEvent",
-    "WebhookEventType",
-    "ConsolidationEventData",
-    "RetainEventData",
-]

hindsight-api-slim/hindsight_api/webhooks/manager.py

@@ -1,242 +0,0 @@
-"""Webhook manager for delivering event notifications."""
-
-import hashlib
-import hmac
-import json
-import logging
-import uuid
-from datetime import datetime, timezone
-from typing import TYPE_CHECKING
-
-import asyncpg
-
-from .models import WebhookConfig, WebhookEvent, WebhookHttpConfig
-
-if TYPE_CHECKING:
-    from hindsight_api.extensions.tenant import TenantExtension
-
-logger = logging.getLogger(__name__)
-
-# Retry delay schedule in seconds: 5 retries after the first attempt.
-# Fast early retries catch transient failures; later retries handle longer outages.
-RETRY_DELAYS = [5, 300, 1800, 7200, 18000]
-MAX_ATTEMPTS = len(RETRY_DELAYS) + 1  # first attempt + len(RETRY_DELAYS) retries
-
-
-def _fq_table(table: str, schema: str | None = None) -> str:
-    """Get fully-qualified table name with optional schema prefix."""
-    if schema:
-        return f'"{schema}".{table}'
-    return table
-
-
-def _parse_http_config(value: str | dict | None) -> WebhookHttpConfig:
-    """Parse http_config column value (JSONB returned as text or dict) into a model."""
-    if value is None:
-        return WebhookHttpConfig()
-    if isinstance(value, str):
-        return WebhookHttpConfig.model_validate_json(value)
-    return WebhookHttpConfig.model_validate(value)
-
-
-class WebhookManager:
-    """
-    Manages webhook registration and event firing.
-
-    Supports both global webhooks (configured via env vars) and per-bank
-    webhooks stored in the database. Deliveries are queued as async_operations
-    tasks (operation_type='webhook_delivery') and picked up by the worker poller.
-    """
-
-    def __init__(
-        self,
-        pool: asyncpg.Pool,
-        global_webhooks: list[WebhookConfig],
-        tenant_extension: "TenantExtension | None" = None,
-    ):
-        self._pool = pool
-        self._global_webhooks = global_webhooks
-        self._tenant_extension = tenant_extension
-
-    def _sign_payload(self, secret: str, payload_bytes: bytes) -> str:
-        """Compute HMAC-SHA256 signature for a payload."""
-        return "sha256=" + hmac.new(secret.encode(), payload_bytes, hashlib.sha256).hexdigest()
-
-    async def fire_event(self, event: WebhookEvent, schema: str | None = None) -> None:
-        """
-        Queue webhook deliveries for an event as async_operations tasks.
-
-        Loads per-bank and global webhooks, inserts pending webhook_delivery tasks for
-        any webhook whose event_types list matches the fired event type. The worker
-        poller picks these up and calls MemoryEngine._handle_webhook_delivery().
-
-        Args:
-            event: The event to deliver.
-            schema: Database schema (for multi-tenant). None = default schema.
-        """
-        webhook_table = _fq_table("webhooks", schema)
-        ops_table = _fq_table("async_operations", schema)
-        now = datetime.now(timezone.utc)
-        payload_str = event.model_dump_json()
-
-        try:
-            # Load per-bank webhooks from DB (bank-specific + global NULL rows)
-            rows = await self._pool.fetch(
-                f"""
-                SELECT id, bank_id, url, secret, event_types, enabled, http_config::text
-                FROM {webhook_table}
-                WHERE (bank_id = $1 OR bank_id IS NULL) AND enabled = true
-                """,
-                event.bank_id,
-            )
-
-            db_webhooks = [
-                WebhookConfig(
-                    id=str(row["id"]),
-                    bank_id=row["bank_id"],
-                    url=row["url"],
-                    secret=row["secret"],
-                    event_types=list(row["event_types"]) if row["event_types"] else [],
-                    enabled=row["enabled"],
-                    http_config=_parse_http_config(row["http_config"]),
-                )
-                for row in rows
-            ]
-
-            # Merge with global webhooks from env config
-            all_webhooks = self._global_webhooks + db_webhooks
-            matched = 0
-
-            for webhook in all_webhooks:
-                if not webhook.enabled:
-                    continue
-                if event.event.value not in webhook.event_types:
-                    continue
-
-                operation_id = uuid.uuid4()
-                webhook_id = webhook.id if webhook.id else None
-
-                task_payload = json.dumps(
-                    {
-                        "type": "webhook_delivery",
-                        "operation_id": str(operation_id),
-                        "bank_id": event.bank_id,
-                        "url": webhook.url,
-                        "secret": webhook.secret,
-                        "event_type": event.event.value,
-                        "payload": payload_str,
-                        "webhook_id": webhook_id,
-                        "http_config": webhook.http_config.model_dump(),
-                    }
-                )
-
-                await self._pool.execute(
-                    f"""
-                    INSERT INTO {ops_table}
-                      (operation_id, bank_id, operation_type, status, task_payload, result_metadata, created_at, updated_at)
-                    VALUES ($1, $2, 'webhook_delivery', 'pending', $3::jsonb, '{{}}'::jsonb, $4, $4)
-                    """,
-                    operation_id,
-                    event.bank_id,
-                    task_payload,
-                    now,
-                )
-                matched += 1
-
-            logger.debug(f"Fired webhook event {event.event} for bank {event.bank_id}: {matched} delivery(ies) queued")
-
-        except Exception as e:
-            logger.error(f"Failed to queue webhook deliveries for event {event.event}: {e}")
-
-    async def fire_event_with_conn(
-        self, event: WebhookEvent, conn: asyncpg.Connection, schema: str | None = None
-    ) -> None:
-        """
-        Queue webhook deliveries within an existing database connection/transaction.
-
-        Identical to fire_event() but uses the provided connection instead of acquiring
-        one from the pool. Use this to atomically insert delivery tasks in the same
-        transaction as the primary operation (transactional outbox pattern).
-
-        Args:
-            event: The event to deliver.
-            conn: Existing asyncpg connection (may be inside an active transaction).
-            schema: Database schema (for multi-tenant). None = default schema.
-        """
-        webhook_table = _fq_table("webhooks", schema)
-        ops_table = _fq_table("async_operations", schema)
-        now = datetime.now(timezone.utc)
-        payload_str = event.model_dump_json()
-
-        try:
-            rows = await conn.fetch(
-                f"""
-                SELECT id, bank_id, url, secret, event_types, enabled, http_config::text
-                FROM {webhook_table}
-                WHERE (bank_id = $1 OR bank_id IS NULL) AND enabled = true
-                """,
-                event.bank_id,
-            )
-
-            db_webhooks = [
-                WebhookConfig(
-                    id=str(row["id"]),
-                    bank_id=row["bank_id"],
-                    url=row["url"],
-                    secret=row["secret"],
-                    event_types=list(row["event_types"]) if row["event_types"] else [],
-                    enabled=row["enabled"],
-                    http_config=_parse_http_config(row["http_config"]),
-                )
-                for row in rows
-            ]
-
-            all_webhooks = self._global_webhooks + db_webhooks
-            matched = 0
-
-            for webhook in all_webhooks:
-                if not webhook.enabled:
-                    continue
-                if event.event.value not in webhook.event_types:
-                    continue
-
-                operation_id = uuid.uuid4()
-                webhook_id = webhook.id if webhook.id else None
-
-                task_payload = json.dumps(
-                    {
-                        "type": "webhook_delivery",
-                        "operation_id": str(operation_id),
-                        "bank_id": event.bank_id,
-                        "url": webhook.url,
-                        "secret": webhook.secret,
-                        "event_type": event.event.value,
-                        "payload": payload_str,
-                        "webhook_id": webhook_id,
-                        "http_config": webhook.http_config.model_dump(),
-                    }
-                )
-
-                await conn.execute(
-                    f"""
-                    INSERT INTO {ops_table}
-                      (operation_id, bank_id, operation_type, status, task_payload, result_metadata, created_at, updated_at)
-                    VALUES ($1, $2, 'webhook_delivery', 'pending', $3::jsonb, '{{}}'::jsonb, $4, $4)
-                    """,
-                    operation_id,
-                    event.bank_id,
-                    task_payload,
-                    now,
-                )
-                matched += 1
-
-            logger.debug(
-                f"Fired webhook event {event.event} for bank {event.bank_id}: {matched} delivery(ies) queued (in-transaction)"
-            )
-
-        except Exception as e:
-            logger.error(
-                f"Failed to queue webhook deliveries (in-transaction) for event {event.event}: {e}. "
-                "CRITICAL: The enclosing database transaction is now aborted and will roll back all changes."
-            )
-            raise

hindsight-api-slim/hindsight_api/webhooks/models.py

@@ -1,51 +0,0 @@
-"""Pydantic models for the webhook system."""
-
-from datetime import datetime
-from enum import StrEnum
-
-from pydantic import BaseModel, Field
-
-
-class WebhookEventType(StrEnum):
-    CONSOLIDATION_COMPLETED = "consolidation.completed"
-    RETAIN_COMPLETED = "retain.completed"
-
-
-class ConsolidationEventData(BaseModel):
-    observations_created: int | None = None
-    observations_updated: int | None = None
-    observations_deleted: int | None = None
-    error_message: str | None = None
-
-
-class RetainEventData(BaseModel):
-    document_id: str | None = None
-    tags: list[str] | None = None
-
-
-class WebhookEvent(BaseModel):
-    event: WebhookEventType
-    bank_id: str
-    operation_id: str
-    status: str  # "completed" or "failed"
-    timestamp: datetime
-    data: ConsolidationEventData | RetainEventData
-
-
-class WebhookHttpConfig(BaseModel):
-    """HTTP delivery configuration for a webhook."""
-
-    method: str = Field(default="POST", description="HTTP method: GET or POST")
-    timeout_seconds: int = Field(default=30, description="HTTP request timeout in seconds")
-    headers: dict[str, str] = Field(default_factory=dict, description="Custom HTTP headers")
-    params: dict[str, str] = Field(default_factory=dict, description="Custom HTTP query parameters")
-
-
-class WebhookConfig(BaseModel):
-    id: str
-    bank_id: str | None
-    url: str
-    secret: str | None
-    event_types: list[str]
-    enabled: bool
-    http_config: WebhookHttpConfig = Field(default_factory=WebhookHttpConfig)

hindsight-api-slim/hindsight_api/worker/exceptions.py

@@ -1,9 +0,0 @@
-from datetime import datetime
-
-
-class RetryTaskAt(Exception):
-    """Raise from a task handler to schedule a retry at a specific time."""
-
-    def __init__(self, retry_at: datetime, message: str = ""):
-        self.retry_at = retry_at
-        super().__init__(message)

hindsight-api-slim/pyproject.toml

@@ -1,209 +0,0 @@
-[build-system]
-requires = ["hatchling"]
-build-backend = "hatchling.build"
-
-[project]
-name = "hindsight-api-slim"
-version = "0.4.20"
-description = "Hindsight: Agent Memory That Works Like Human Memory"
-readme = "README.md"
-requires-python = ">=3.11"
-dependencies = [
-    "asyncpg>=0.29.0",
-    "python-dotenv>=1.0.0",
-    "openai>=1.0.0",
-    "pydantic>=2.0.0",
-    "rich>=13.0.0",
-    "langchain-text-splitters>=0.3.0",
-    "fastapi[standard]>=0.120.3",
-    "uvicorn>=0.38.0",
-    "wsproto>=1.0.0",
-    "sqlalchemy>=2.0.44",
-    "alembic>=1.17.1",
-    "pgvector>=0.4.1",
-    "greenlet>=3.2.4",
-    "psycopg2-binary>=2.9.11",
-    "tiktoken>=0.12.0",
-    "httpx>=0.27.0",
-    "PyJWT[crypto]>=2.8.0",
-    "fastmcp>=2.14.0", # CVE-2025-66416
-    "python-dateutil>=2.8.0",
-    "opentelemetry-api>=1.20.0",
-    "opentelemetry-sdk>=1.20.0",
-    "opentelemetry-instrumentation-fastapi>=0.41b0",
-    "opentelemetry-exporter-prometheus>=0.41b0",
-    "opentelemetry-exporter-otlp-proto-http>=1.20.0",
-    "opentelemetry-semantic-conventions>=0.41b0",
-    "dateparser>=1.2.2",
-    "google-genai>=1.0.0",
-    "google-auth>=2.0.0",
-    "anthropic>=0.40.0",
-    "typer>=0.9.0",
-    "cohere>=5.0.0",
-    "litellm>=1.0.0,<=1.82.6", # 1.82.7+ contains a supply chain attack (malicious .pth credential stealer)
-    "markitdown[pdf,docx,pptx,xlsx,xls]>=0.1.4", # File to markdown conversion
-    "obstore>=0.4.0", # S3/GCS/Azure object storage client (Rust-backed)
-    "uvloop>=0.22.1",
-    # Transitive dependency security fixes
-    "pyasn1>=0.6.3", # DoS vulnerability fix
-    "urllib3>=2.6.3", # Decompression-bomb safeguards bypass fix
-    "langchain-core>=1.2.11", # Serialization injection + SSRF vulnerability fix
-    "langsmith>=0.6.3", # SSRF via tracing header injection fix
-    "protobuf>=6.33.5", # JSON recursion depth bypass fix
-    "pillow>=12.1.1", # Out-of-bounds write in PSD image loading fix
-    "cryptography>=46.0.5", # Subgroup attack vulnerability fix
-    "filelock>=3.20.1", # TOCTOU race condition fix
-    "authlib>=1.6.9", # Account takeover/JWS header injection vulnerability fix
-    "pyjwt>=2.12.0", # Accepts unknown crit header extensions fix
-    "orjson>=3.11.6", # Unbounded recursion DoS fix
-    "tornado>=6.5.5", # DoS multipart/incomplete cookie validation fix
-    "aiohttp>=3.13.3", # Multiple DoS vulnerabilities
-    "claude-agent-sdk>=0.1.27",
-    "boto3>=1.42.74",
-]
-
-[project.optional-dependencies]
-local-ml = [
-    # Local ML models for embeddings/reranking
-    "sentence-transformers>=3.3.0",
-    "transformers>=4.53.0", # Security fixes for ReDoS vulnerabilities
-    "torch>=2.6.0", # CVE fix for remote code execution
-    "einops>=0.8.2",
-    "flashrank>=0.2.0",
-    # Apple Silicon local inference
-    "mlx>=0.31.0",
-    "mlx-lm>=0.31.1",
-    "safetensors>=0.6.2",
-]
-embedded-db = [
-    "pg0-embedded>=0.11.0",
-]
-all = [
-    "hindsight-api-slim[local-ml,embedded-db]",
-]
-test = [
-    "pytest>=7.0.0",
-    "pytest-asyncio>=0.21.0",
-    "pytest-timeout>=2.4.0",
-    "pytest-xdist>=3.0.0",
-    "filelock>=3.20.1",  # TOCTOU race condition fix
-    "testcontainers>=4.0.0",
-]
-
-[project.scripts]
-hindsight-api = "hindsight_api.main:main"
-hindsight-worker = "hindsight_api.worker.main:main"
-hindsight-local-mcp = "hindsight_api.mcp_local:main"
-hindsight-admin = "hindsight_api.admin.cli:main"
-
-[tool.hatch.build.targets.wheel]
-packages = ["hindsight_api"]
-
-[tool.hatch.build.targets.wheel.sources]
-"hindsight_api" = "hindsight_api"
-
-[tool.hatch.build.targets.sdist]
-include = [
-    "hindsight_api/**/*",
-]
-
-[tool.hatch.build]
-include = [
-    "hindsight_api/**/*.py",
-    "hindsight_api/alembic/**/*",
-]
-
-[tool.pytest.ini_options]
-log_cli = true
-log_cli_level = "INFO"
-log_cli_format = "%(asctime)s - %(levelname)s - %(name)s - %(message)s"
-log_cli_date_format = "%Y-%m-%d %H:%M:%S"
-addopts = "--timeout 300 -n 8 --dist loadgroup --durations=10 -v"
-asyncio_mode = "auto"
-asyncio_default_fixture_loop_scope = "function"
-log_auto_indent = true
-filterwarnings = [
-    "ignore:The @wait_container_is_ready decorator is deprecated:DeprecationWarning",
-    "ignore::RuntimeWarning:asyncio",
-]
-
-[dependency-groups]
-dev = [
-    "pytest>=9.0.0",
-    "pytest-asyncio>=1.3.0",
-    "pytest-timeout>=2.4.0",
-    "pytest-xdist>=3.8.0",
-    "python-dotenv>=1.2.1",
-    "filelock>=3.20.1",  # TOCTOU race condition fix
-    "ruff>=0.8.0",
-    "ty>=0.0.1",
-    "testcontainers>=4.0.0",
-]
-
-[tool.ruff]
-line-length = 120
-target-version = "py311"
-exclude = [
-    "tests/",
-    "**/tests/",
-]
-
-[tool.ruff.lint]
-select = [
-    "E",      # pycodestyle errors
-    "W",      # pycodestyle warnings
-    "F",      # Pyflakes
-    "I",      # isort
-]
-ignore = [
-    "E501",   # line too long (handled by formatter)
-    "E402",   # module import not at top of file
-    "F401",   # unused import (too noisy during development)
-    "F841",   # unused variable (too noisy during development)
-    "F811",   # redefined while unused
-    "F821",   # undefined name (forward references in type hints)
-]
-
-[tool.ruff.lint.isort]
-known-third-party = ["alembic"]
-
-[tool.ruff.format]
-quote-style = "double"
-indent-style = "space"
-
-[tool.uv]
-# Use explicit index for PyTorch to prevent the pytorch index from serving
-# non-pytorch packages (e.g. markupsafe) with incompatible wheels
-[[tool.uv.index]]
-name = "pytorch-cpu"
-url = "https://download.pytorch.org/whl/cpu"
-explicit = true
-
-[tool.uv.sources]
-# Route torch to the CPU-only PyTorch index; everything else uses PyPI
-torch = { index = "pytorch-cpu" }
-
-[tool.ty]
-# Type checking configuration
-# ty is an extremely fast Python type checker from Astral (same team as ruff/uv)
-
-[tool.ty.environment]
-python-version = "3.11"
-
-[tool.ty.src]
-exclude = [
-    "tests/",
-    "hindsight_api/alembic/",
-]
-
-[tool.ty.rules]
-# Disable noisy rules while keeping important ones
-invalid-argument-type = "ignore"    # False positives with **kwargs patterns
-invalid-return-type = "ignore"      # Often intentional in async code
-invalid-parameter-default = "ignore" # Optional params with None default
-possibly-missing-attribute = "ignore" # Common with Optional types
-invalid-raise = "ignore"            # False positives with exception tracking
-call-non-callable = "ignore"        # False positives with Optional types
-invalid-key = "ignore"              # Pydantic ConfigDict not understood
-invalid-method-override = "ignore"  # Intentional signature differences
-unresolved-reference = "ignore"     # Forward references not always resolved

hindsight-api-slim/tests/test_async_batch_retain.py

@@ -1,435 +0,0 @@
-"""Test async batch retain with smart batching and parent-child operations."""
-
-import asyncio
-import json
-import uuid
-
-import pytest
-
-from hindsight_api.extensions import RequestContext
-
-
-async def _ensure_bank(pool, bank_id: str) -> None:
-    """Upsert a minimal bank row so FK on async_operations passes."""
-    await pool.execute(
-        "INSERT INTO banks (bank_id, name) VALUES ($1, $2) ON CONFLICT DO NOTHING",
-        bank_id,
-        bank_id,
-    )
-
-
-@pytest.mark.asyncio
-async def test_duplicate_document_ids_rejected_async(memory, request_context):
-    """Test that async retain rejects batches with duplicate document_ids."""
-    bank_id = "test_duplicate_async"
-    contents = [
-        {"content": "First item", "document_id": "doc1"},
-        {"content": "Second item", "document_id": "doc2"},
-        {"content": "Third item", "document_id": "doc1"},  # Duplicate!
-    ]
-
-    # Should raise ValueError due to duplicate document_ids
-    with pytest.raises(ValueError, match="duplicate document_ids.*doc1"):
-        await memory.submit_async_retain(
-            bank_id=bank_id,
-            contents=contents,
-            request_context=request_context,
-        )
-
-
-@pytest.mark.asyncio
-async def test_duplicate_document_ids_rejected_sync(memory, request_context):
-    """Test that sync retain also rejects batches with duplicate document_ids."""
-    bank_id = "test_duplicate_sync"
-    contents = [
-        {"content": "First item", "document_id": "doc1"},
-        {"content": "Second item", "document_id": "doc1"},  # Duplicate!
-    ]
-
-    # Should raise ValueError due to duplicate document_ids
-    with pytest.raises(ValueError, match="duplicate document_ids.*doc1"):
-        await memory.retain_batch_async(
-            bank_id=bank_id,
-            contents=contents,
-            request_context=request_context,
-        )
-
-
-@pytest.mark.asyncio
-async def test_small_async_batch_no_splitting(memory, request_context):
-    """Test that small async batches create parent with single child (simplified code path)."""
-    bank_id = "test_small_async"
-    contents = [{"content": "Alice works at Google", "document_id": f"doc{i}"} for i in range(5)]
-
-    # Calculate total chars (should be well under threshold)
-    total_chars = sum(len(item["content"]) for item in contents)
-    assert total_chars < 10_000, "Test batch should be small"
-
-    # Submit async retain
-    result = await memory.submit_async_retain(
-        bank_id=bank_id,
-        contents=contents,
-        request_context=request_context,
-    )
-
-    # Verify we got an operation_id back
-    assert "operation_id" in result
-    assert "items_count" in result
-    assert result["items_count"] == 5
-
-    operation_id = result["operation_id"]
-
-    # Wait for task to complete (SyncTaskBackend executes immediately)
-    await asyncio.sleep(0.1)
-
-    # Check operation status
-    status = await memory.get_operation_status(
-        bank_id=bank_id,
-        operation_id=operation_id,
-        request_context=request_context,
-    )
-
-    # Should be a parent operation with single child (simplified code path)
-    assert status["status"] == "completed"
-    assert status["operation_type"] == "batch_retain"
-    assert "child_operations" in status
-    assert status["result_metadata"]["num_sub_batches"] == 1  # Single sub-batch
-    assert len(status["child_operations"]) == 1
-    assert status["child_operations"][0]["status"] == "completed"
-
-
-@pytest.mark.asyncio
-async def test_large_async_batch_auto_splits(memory, request_context):
-    """Test that large async batches automatically split into sub-batches with parent operation."""
-    from hindsight_api.engine.memory_engine import count_tokens
-
-    bank_id = "test_large_async"
-
-    # Create a large batch that exceeds the threshold (10k tokens default)
-    # Repeating "A"s gets heavily compressed by tokenizer, use varied content
-    # Use ~22k chars per item = ~5.5k tokens per item, 2 items = ~11k tokens total (exceeds 10k)
-    large_content = "The quick brown fox jumps over the lazy dog. " * 500  # ~22k chars = ~5.5k tokens
-    contents = [{"content": large_content + f" item {i}", "document_id": f"doc{i}"} for i in range(2)]
-
-    # Calculate total tokens (should exceed threshold)
-    total_tokens = sum(count_tokens(item["content"]) for item in contents)
-    assert total_tokens > 10_000, "Test batch should exceed threshold"
-
-    # Submit async retain
-    result = await memory.submit_async_retain(
-        bank_id=bank_id,
-        contents=contents,
-        request_context=request_context,
-    )
-
-    # Verify we got an operation_id back
-    assert "operation_id" in result
-    assert "items_count" in result
-    assert result["items_count"] == 2
-
-    parent_operation_id = result["operation_id"]
-
-    # Wait for tasks to complete
-    await asyncio.sleep(0.5)
-
-    # Check parent operation status
-    parent_status = await memory.get_operation_status(
-        bank_id=bank_id,
-        operation_id=parent_operation_id,
-        request_context=request_context,
-    )
-
-    # Should be a parent operation with children
-    assert parent_status["operation_type"] == "batch_retain"
-    assert "child_operations" in parent_status
-    assert "num_sub_batches" in parent_status["result_metadata"]
-    assert parent_status["result_metadata"]["num_sub_batches"] >= 2  # Should split into at least 2 batches
-    assert parent_status["result_metadata"]["items_count"] == 2
-
-    # Verify child operations
-    child_ops = parent_status["child_operations"]
-    assert len(child_ops) >= 2, "Should have at least 2 child operations"
-
-    # All children should be completed (SyncTaskBackend executes immediately)
-    for child in child_ops:
-        assert child["status"] == "completed"
-        assert child["sub_batch_index"] is not None
-        assert child["items_count"] > 0
-
-    # Parent status should be aggregated as "completed"
-    assert parent_status["status"] == "completed"
-
-
-@pytest.mark.asyncio
-async def test_parent_operation_status_aggregation_pending(memory, request_context):
-    """Test that parent operation shows 'pending' when children are pending."""
-    bank_id = "test_parent_pending"
-    pool = await memory._get_pool()
-    await _ensure_bank(pool, bank_id)
-
-    # Manually create a parent operation
-    parent_id = uuid.uuid4()
-    async with pool.acquire() as conn:
-        await conn.execute(
-            """
-            INSERT INTO async_operations (operation_id, bank_id, operation_type, result_metadata, status)
-            VALUES ($1, $2, $3, $4, $5)
-            """,
-            parent_id,
-            bank_id,
-            "batch_retain",
-            json.dumps({"items_count": 20, "num_sub_batches": 2, "is_parent": True}),
-            "pending",
-        )
-
-        # Create 2 child operations - one completed, one pending
-        child1_id = uuid.uuid4()
-        await conn.execute(
-            """
-            INSERT INTO async_operations (operation_id, bank_id, operation_type, result_metadata, status)
-            VALUES ($1, $2, $3, $4, $5)
-            """,
-            child1_id,
-            bank_id,
-            "retain",
-            json.dumps(
-                {
-                    "items_count": 10,
-                    "parent_operation_id": str(parent_id),
-                    "sub_batch_index": 1,
-                    "total_sub_batches": 2,
-                }
-            ),
-            "completed",
-        )
-
-        child2_id = uuid.uuid4()
-        await conn.execute(
-            """
-            INSERT INTO async_operations (operation_id, bank_id, operation_type, result_metadata, status)
-            VALUES ($1, $2, $3, $4, $5)
-            """,
-            child2_id,
-            bank_id,
-            "retain",
-            json.dumps(
-                {
-                    "items_count": 10,
-                    "parent_operation_id": str(parent_id),
-                    "sub_batch_index": 2,
-                    "total_sub_batches": 2,
-                }
-            ),
-            "pending",
-        )
-
-    # Check parent status
-    parent_status = await memory.get_operation_status(
-        bank_id=bank_id,
-        operation_id=str(parent_id),
-        request_context=request_context,
-    )
-
-    # Parent should aggregate as "pending" since one child is still pending
-    assert parent_status["status"] == "pending"
-    assert len(parent_status["child_operations"]) == 2
-
-
-@pytest.mark.asyncio
-async def test_parent_operation_status_aggregation_failed(memory, request_context):
-    """Test that parent operation shows 'failed' when any child fails."""
-    bank_id = "test_parent_failed"
-    pool = await memory._get_pool()
-    await _ensure_bank(pool, bank_id)
-
-    # Manually create a parent operation
-    parent_id = uuid.uuid4()
-    async with pool.acquire() as conn:
-        await conn.execute(
-            """
-            INSERT INTO async_operations (operation_id, bank_id, operation_type, result_metadata, status)
-            VALUES ($1, $2, $3, $4, $5)
-            """,
-            parent_id,
-            bank_id,
-            "batch_retain",
-            json.dumps({"items_count": 20, "num_sub_batches": 2, "is_parent": True}),
-            "pending",
-        )
-
-        # Create 2 child operations - one completed, one failed
-        child1_id = uuid.uuid4()
-        await conn.execute(
-            """
-            INSERT INTO async_operations (operation_id, bank_id, operation_type, result_metadata, status)
-            VALUES ($1, $2, $3, $4, $5)
-            """,
-            child1_id,
-            bank_id,
-            "retain",
-            json.dumps(
-                {
-                    "items_count": 10,
-                    "parent_operation_id": str(parent_id),
-                    "sub_batch_index": 1,
-                    "total_sub_batches": 2,
-                }
-            ),
-            "completed",
-        )
-
-        child2_id = uuid.uuid4()
-        await conn.execute(
-            """
-            INSERT INTO async_operations (operation_id, bank_id, operation_type, result_metadata, status, error_message)
-            VALUES ($1, $2, $3, $4, $5, $6)
-            """,
-            child2_id,
-            bank_id,
-            "retain",
-            json.dumps(
-                {
-                    "items_count": 10,
-                    "parent_operation_id": str(parent_id),
-                    "sub_batch_index": 2,
-                    "total_sub_batches": 2,
-                }
-            ),
-            "failed",
-            "Test error",
-        )
-
-    # Check parent status
-    parent_status = await memory.get_operation_status(
-        bank_id=bank_id,
-        operation_id=str(parent_id),
-        request_context=request_context,
-    )
-
-    # Parent should aggregate as "failed" since one child failed
-    assert parent_status["status"] == "failed"
-    assert len(parent_status["child_operations"]) == 2
-
-    # Verify child with error is included
-    failed_child = [c for c in parent_status["child_operations"] if c["status"] == "failed"][0]
-    assert failed_child["error_message"] == "Test error"
-
-
-@pytest.mark.asyncio
-async def test_parent_operation_status_aggregation_completed(memory, request_context):
-    """Test that parent operation shows 'completed' when all children are completed."""
-    bank_id = "test_parent_completed"
-    pool = await memory._get_pool()
-    await _ensure_bank(pool, bank_id)
-
-    # Manually create a parent operation
-    parent_id = uuid.uuid4()
-    async with pool.acquire() as conn:
-        await conn.execute(
-            """
-            INSERT INTO async_operations (operation_id, bank_id, operation_type, result_metadata, status)
-            VALUES ($1, $2, $3, $4, $5)
-            """,
-            parent_id,
-            bank_id,
-            "batch_retain",
-            json.dumps({"items_count": 20, "num_sub_batches": 2, "is_parent": True}),
-            "pending",
-        )
-
-        # Create 2 child operations - both completed
-        child1_id = uuid.uuid4()
-        await conn.execute(
-            """
-            INSERT INTO async_operations (operation_id, bank_id, operation_type, result_metadata, status)
-            VALUES ($1, $2, $3, $4, $5)
-            """,
-            child1_id,
-            bank_id,
-            "retain",
-            json.dumps(
-                {
-                    "items_count": 10,
-                    "parent_operation_id": str(parent_id),
-                    "sub_batch_index": 1,
-                    "total_sub_batches": 2,
-                }
-            ),
-            "completed",
-        )
-
-        child2_id = uuid.uuid4()
-        await conn.execute(
-            """
-            INSERT INTO async_operations (operation_id, bank_id, operation_type, result_metadata, status)
-            VALUES ($1, $2, $3, $4, $5)
-            """,
-            child2_id,
-            bank_id,
-            "retain",
-            json.dumps(
-                {
-                    "items_count": 10,
-                    "parent_operation_id": str(parent_id),
-                    "sub_batch_index": 2,
-                    "total_sub_batches": 2,
-                }
-            ),
-            "completed",
-        )
-
-    # Check parent status
-    parent_status = await memory.get_operation_status(
-        bank_id=bank_id,
-        operation_id=str(parent_id),
-        request_context=request_context,
-    )
-
-    # Parent should aggregate as "completed" since all children are completed
-    assert parent_status["status"] == "completed"
-    assert len(parent_status["child_operations"]) == 2
-    assert all(c["status"] == "completed" for c in parent_status["child_operations"])
-
-
-@pytest.mark.asyncio
-async def test_config_retain_batch_tokens_respected(memory, request_context):
-    """Test that the retain_batch_tokens config setting is respected."""
-    from hindsight_api.config import get_config
-    from hindsight_api.engine.memory_engine import count_tokens
-
-    bank_id = "test_config_batch_tokens"
-    config = get_config()
-
-    # Check that config has the retain_batch_tokens setting
-    assert hasattr(config, "retain_batch_tokens")
-    assert config.retain_batch_tokens > 0
-
-    # Create a batch that's just under the threshold
-    # Use content that produces roughly half the token limit per item
-    content_size = config.retain_batch_tokens * 2  # chars (rough estimate: 1 token ~= 4 chars)
-    contents = [{"content": "A" * content_size, "document_id": f"doc{i}"} for i in range(2)]
-
-    total_tokens = sum(count_tokens(item["content"]) for item in contents)
-    # Should be equal to threshold (boundary case, no splitting since we use > not >=)
-    assert total_tokens <= config.retain_batch_tokens
-
-    # Submit - should NOT split
-    result = await memory.submit_async_retain(
-        bank_id=bank_id,
-        contents=contents,
-        request_context=request_context,
-    )
-
-    # Wait for completion
-    await asyncio.sleep(0.1)
-
-    # Check status - should be a parent with single child (even for small batches)
-    status = await memory.get_operation_status(
-        bank_id=bank_id,
-        operation_id=result["operation_id"],
-        request_context=request_context,
-    )
-
-    # Even small batches use parent-child pattern now (simpler code path)
-    assert "child_operations" in status
-    assert status["result_metadata"]["num_sub_batches"] == 1

hindsight-api-slim/tests/test_async_retain_tags.py

@@ -1,95 +0,0 @@
-"""Unit tests for async retain tag propagation."""
-
-from unittest.mock import AsyncMock, MagicMock, patch
-
-import pytest
-
-from hindsight_api.engine.memory_engine import MemoryEngine
-from hindsight_api.models import RequestContext
-
-
-@pytest.mark.asyncio
-async def test_submit_async_retain_includes_document_tags_in_task_payload():
-    """submit_async_retain should include document_tags in queued task payload."""
-    engine = MemoryEngine.__new__(MemoryEngine)
-    engine._initialized = True
-    engine._authenticate_tenant = AsyncMock()
-    engine._operation_validator = None
-    engine._submit_async_operation = AsyncMock(return_value={"operation_id": "op-1"})
-
-    # Mock the pool and connection for parent operation creation
-    mock_conn = AsyncMock()
-    mock_conn.execute = AsyncMock()
-    mock_conn.transaction = MagicMock()
-    mock_conn.transaction.return_value.__aenter__ = AsyncMock()
-    mock_conn.transaction.return_value.__aexit__ = AsyncMock()
-
-    mock_pool = AsyncMock()
-    mock_pool.acquire = AsyncMock(return_value=mock_conn)
-    mock_pool.release = AsyncMock()
-
-    engine._get_pool = AsyncMock(return_value=mock_pool)
-
-    request_context = RequestContext(tenant_id="tenant-a", api_key_id="key-a")
-    contents = [{"content": "Async retain payload test."}]
-    document_tags = ["scope:tools", "user:alice"]
-
-    with patch("hindsight_api.engine.memory_engine.bank_utils.get_bank_profile", new_callable=AsyncMock):
-        result = await MemoryEngine.submit_async_retain(
-            engine,
-            bank_id="bank-1",
-            contents=contents,
-            document_tags=document_tags,
-            request_context=request_context,
-        )
-
-    # Check result structure
-    assert "operation_id" in result
-    assert "items_count" in result
-    assert result["items_count"] == 1
-
-    # Verify authentication was called
-    engine._authenticate_tenant.assert_awaited_once_with(request_context)
-
-    # Verify child operation was submitted
-    engine._submit_async_operation.assert_awaited_once()
-
-    # Verify child operation payload contains document_tags
-    kwargs = engine._submit_async_operation.await_args.kwargs
-    assert kwargs["bank_id"] == "bank-1"
-    assert kwargs["operation_type"] == "retain"
-    assert kwargs["task_type"] == "batch_retain"
-    assert kwargs["task_payload"]["contents"] == contents
-    assert kwargs["task_payload"]["document_tags"] == document_tags
-    assert kwargs["task_payload"]["_tenant_id"] == "tenant-a"
-    assert kwargs["task_payload"]["_api_key_id"] == "key-a"
-
-
-@pytest.mark.asyncio
-async def test_handle_batch_retain_forwards_document_tags_to_retain_batch_async():
-    """Worker handler should forward document_tags from task payload."""
-    engine = MemoryEngine.__new__(MemoryEngine)
-    engine._initialized = True
-    engine.retain_batch_async = AsyncMock(return_value={"items_count": 1})
-
-    task_dict = {
-        "bank_id": "bank-1",
-        "contents": [{"content": "Forward tags test."}],
-        "document_tags": ["scope:client"],
-        "_tenant_id": "tenant-a",
-        "_api_key_id": "key-a",
-    }
-
-    await MemoryEngine._handle_batch_retain(engine, task_dict)
-
-    engine.retain_batch_async.assert_awaited_once()
-    kwargs = engine.retain_batch_async.await_args.kwargs
-    assert kwargs["bank_id"] == "bank-1"
-    assert kwargs["contents"] == task_dict["contents"]
-    assert kwargs["document_tags"] == ["scope:client"]
-
-    request_context = kwargs["request_context"]
-    assert request_context.internal is True
-    assert request_context.user_initiated is True
-    assert request_context.tenant_id == "tenant-a"
-    assert request_context.api_key_id == "key-a"

hindsight-api-slim/tests/test_batch_api.py

@@ -1,507 +0,0 @@
-"""
-Test OpenAI Batch API integration for retain fact extraction.
-
-Tests cover:
-- Normal batch API flow (submit, poll, complete)
-- Crash recovery (resume from existing batch_id)
-- Provider fallback (when batch API not supported)
-- Worker recovery on restart
-"""
-import pytest
-import asyncio
-import logging
-import json
-import uuid
-from datetime import datetime, timezone
-from unittest.mock import AsyncMock, MagicMock, patch
-from hindsight_api import RequestContext
-from hindsight_api.engine.retain.fact_extraction import (
-    extract_facts_from_contents_batch_api,
-    extract_facts_from_contents,
-    RetainContent,
-)
-from hindsight_api.config import HindsightConfig
-from hindsight_api.engine.llm_wrapper import create_llm_provider
-from hindsight_api.worker.poller import WorkerPoller
-
-logger = logging.getLogger(__name__)
-
-
-@pytest.fixture
-def mock_llm_config():
-    """Create a mock LLM config with batch API support."""
-    mock = MagicMock()
-    mock.provider = "openai"
-    mock.model = "gpt-4o-mini"
-    mock._provider_impl = AsyncMock()
-    return mock
-
-
-@pytest.fixture
-def test_contents():
-    """Create test content for fact extraction."""
-    return [
-        RetainContent(
-            content="Alice is a senior software engineer at TechCorp. She specializes in distributed systems.",
-            event_date=datetime(2024, 1, 15, tzinfo=timezone.utc),
-            context="team overview",
-        ),
-        RetainContent(
-            content="Bob joined the team last month as a junior developer. He is learning React.",
-            event_date=datetime(2024, 1, 15, tzinfo=timezone.utc),
-            context="team overview",
-        ),
-    ]
-
-
-@pytest.fixture
-def hindsight_config():
-    """Create test config with batch API enabled."""
-    config = HindsightConfig.from_env()
-    config.retain_batch_enabled = True
-    config.retain_batch_poll_interval_seconds = 1  # Fast polling for tests
-    config.retain_chunk_size = 4000
-    config.retain_extraction_mode = "concise"
-    config.retain_extract_causal_links = False
-    return config
-
-
-@pytest.mark.asyncio
-async def test_batch_api_normal_flow(mock_llm_config, test_contents, hindsight_config, memory, request_context):
-    """Test normal batch API flow: submit, poll, complete."""
-    bank_id = f"test_batch_{datetime.now(timezone.utc).timestamp()}"
-
-    try:
-        # Mock batch API responses
-        batch_id = "batch_test123"
-
-        # Mock supports_batch_api
-        mock_llm_config._provider_impl.supports_batch_api = AsyncMock(return_value=True)
-
-        # Mock submit_batch - returns batch metadata
-        mock_llm_config._provider_impl.submit_batch = AsyncMock(
-            return_value={
-                "batch_id": batch_id,
-                "status": "validating",
-                "request_counts": {"total": 2, "completed": 0, "failed": 0},
-            }
-        )
-
-        # Mock get_batch_status - simulate polling sequence
-        status_sequence = [
-            {"status": "in_progress", "request_counts": {"total": 2, "completed": 1, "failed": 0}},
-            {"status": "completed", "request_counts": {"total": 2, "completed": 2, "failed": 0}},
-        ]
-        mock_llm_config._provider_impl.get_batch_status = AsyncMock(side_effect=status_sequence)
-
-        # Mock retrieve_batch_results - returns fact extraction results
-        mock_results = [
-            {
-                "custom_id": "chunk_0",
-                "response": {
-                    "body": {
-                        "choices": [
-                            {
-                                "message": {
-                                    "content": json.dumps({
-                                        "facts": [
-                                            {
-                                                "what": "Alice is a senior software engineer at TechCorp",
-                                                "when": "present",
-                                                "where": "TechCorp",
-                                                "who": "Alice",
-                                                "why": "Professional background information",
-                                                "fact_type": "world",
-                                                "fact_kind": "conversation",
-                                            }
-                                        ]
-                                    })
-                                }
-                            }
-                        ],
-                        "usage": {"prompt_tokens": 100, "completion_tokens": 50, "total_tokens": 150},
-                    }
-                },
-            },
-            {
-                "custom_id": "chunk_1",
-                "response": {
-                    "body": {
-                        "choices": [
-                            {
-                                "message": {
-                                    "content": json.dumps({
-                                        "facts": [
-                                            {
-                                                "what": "Bob joined the team last month as a junior developer",
-                                                "when": "last month",
-                                                "where": "team",
-                                                "who": "Bob",
-                                                "why": "New team member information",
-                                                "fact_type": "world",
-                                                "fact_kind": "conversation",
-                                            }
-                                        ]
-                                    })
-                                }
-                            }
-                        ],
-                        "usage": {"prompt_tokens": 100, "completion_tokens": 50, "total_tokens": 150},
-                    }
-                },
-            },
-        ]
-        mock_llm_config._provider_impl.retrieve_batch_results = AsyncMock(return_value=mock_results)
-
-        # Call batch API extraction
-        facts, chunks, usage = await extract_facts_from_contents_batch_api(
-            contents=test_contents,
-            llm_config=mock_llm_config,
-            agent_name="test_agent",
-            config=hindsight_config,
-            pool=None,  # No DB pool for this test
-            operation_id=None,
-            schema=None,
-        )
-
-        # Verify results
-        assert len(facts) == 2, "Should extract 2 facts (one per chunk)"
-        # Facts are ExtractedFact objects with .fact_text field
-        assert "Alice" in facts[0].fact_text and "senior software engineer" in facts[0].fact_text
-        assert "Bob" in facts[1].fact_text and "junior developer" in facts[1].fact_text
-
-        # Verify chunks metadata
-        assert len(chunks) == 2, "Should have 2 chunks metadata"
-        assert chunks[0].fact_count == 1
-        assert chunks[1].fact_count == 1
-
-        # Verify token usage
-        assert usage.input_tokens == 200  # 100 per chunk
-        assert usage.output_tokens == 100  # 50 per chunk
-        assert usage.total_tokens == 300
-
-        # Verify API calls
-        mock_llm_config._provider_impl.submit_batch.assert_called_once()
-        assert mock_llm_config._provider_impl.get_batch_status.call_count == 2
-        mock_llm_config._provider_impl.retrieve_batch_results.assert_called_once_with(batch_id)
-
-        logger.info("✅ Normal batch API flow test passed")
-
-    finally:
-        # Cleanup
-        try:
-            await memory.delete_bank(bank_id, request_context=request_context)
-        except Exception:
-            pass
-
-
-@pytest.mark.asyncio
-async def test_batch_api_crash_recovery(mock_llm_config, test_contents, hindsight_config, memory, request_context):
-    """Test crash recovery: resume polling from existing batch_id."""
-    bank_id = f"test_crash_{datetime.now(timezone.utc).timestamp()}"
-    operation_id = str(uuid.uuid4())  # Must be UUID for async_operations table
-
-    try:
-        # Ensure bank exists
-        await memory.get_bank_profile(bank_id, request_context=request_context)
-
-        # Setup: Store batch_id in async_operations table (simulates partial execution)
-        batch_id = "batch_recovered_456"
-        pool = memory._pool
-        schema = request_context.tenant_id
-
-        from hindsight_api.engine.task_backend import fq_table
-        table = fq_table("async_operations", schema)
-
-        # Create operation with batch_id already stored
-        await pool.execute(
-            f"""
-            INSERT INTO {table} (operation_id, operation_type, bank_id, status, result_metadata)
-            VALUES ($1, 'retain', $2, 'processing', $3::jsonb)
-            """,
-            operation_id,
-            bank_id,
-            json.dumps({
-                "batch_id": batch_id,
-                "batch_provider": "openai",
-                "chunk_count": 2,
-            }),
-        )
-
-        # Mock batch API responses for resume scenario
-        mock_llm_config._provider_impl.supports_batch_api = AsyncMock(return_value=True)
-
-        # Mock get_batch_status - batch already in progress
-        mock_llm_config._provider_impl.get_batch_status = AsyncMock(
-            return_value={
-                "status": "completed",
-                "request_counts": {"total": 2, "completed": 2, "failed": 0},
-            }
-        )
-
-        # Mock retrieve_batch_results
-        mock_results = [
-            {
-                "custom_id": "chunk_0",
-                "response": {
-                    "body": {
-                        "choices": [
-                            {
-                                "message": {
-                                    "content": json.dumps({
-                                        "facts": [
-                                            {
-                                                "what": "Alice is a senior software engineer",
-                                                "when": "present",
-                                                "where": "TechCorp",
-                                                "who": "Alice",
-                                                "why": "Background",
-                                                "fact_type": "world",
-                                                "fact_kind": "conversation",
-                                            }
-                                        ]
-                                    })
-                                }
-                            }
-                        ],
-                        "usage": {"prompt_tokens": 100, "completion_tokens": 50, "total_tokens": 150},
-                    }
-                },
-            },
-            {
-                "custom_id": "chunk_1",
-                "response": {
-                    "body": {
-                        "choices": [
-                            {
-                                "message": {
-                                    "content": json.dumps({
-                                        "facts": [
-                                            {
-                                                "what": "Bob is a junior developer",
-                                                "when": "last month",
-                                                "where": "team",
-                                                "who": "Bob",
-                                                "why": "New member",
-                                                "fact_type": "world",
-                                                "fact_kind": "conversation",
-                                            }
-                                        ]
-                                    })
-                                }
-                            }
-                        ],
-                        "usage": {"prompt_tokens": 100, "completion_tokens": 50, "total_tokens": 150},
-                    }
-                },
-            },
-        ]
-        mock_llm_config._provider_impl.retrieve_batch_results = AsyncMock(return_value=mock_results)
-
-        # Call batch API extraction with operation_id (crash recovery scenario)
-        facts, chunks, usage = await extract_facts_from_contents_batch_api(
-            contents=test_contents,
-            llm_config=mock_llm_config,
-            agent_name="test_agent",
-            config=hindsight_config,
-            pool=pool,
-            operation_id=operation_id,  # Provides crash recovery context
-            schema=schema,
-        )
-
-        # Verify results
-        assert len(facts) == 2, "Should extract 2 facts after recovery"
-
-        # CRITICAL: Verify submit_batch was NOT called (because batch_id already exists)
-        mock_llm_config._provider_impl.submit_batch.assert_not_called()
-
-        # Verify get_batch_status WAS called (polling resumed)
-        mock_llm_config._provider_impl.get_batch_status.assert_called()
-
-        # Verify retrieve_batch_results was called with the recovered batch_id
-        mock_llm_config._provider_impl.retrieve_batch_results.assert_called_once_with(batch_id)
-
-        logger.info("✅ Crash recovery test passed - resumed polling without re-submission")
-
-    finally:
-        # Cleanup
-        try:
-            await memory.delete_bank(bank_id, request_context=request_context)
-        except Exception:
-            pass
-
-
-@pytest.mark.asyncio
-async def test_batch_api_fallback_unsupported_provider(mock_llm_config, test_contents, hindsight_config):
-    """Test fallback to sync mode when provider doesn't support batch API."""
-
-    # Mock provider that doesn't support batch API
-    mock_llm_config._provider_impl.supports_batch_api = AsyncMock(return_value=False)
-    mock_llm_config.provider = "groq"  # Example of provider
-
-    # Patch the sync mode function to verify it's called
-    with patch(
-        "hindsight_api.engine.retain.fact_extraction.extract_facts_from_contents"
-    ) as mock_sync_extract:
-        mock_sync_extract.return_value = ([], [], MagicMock())
-
-        # Call batch API extraction (should fallback to sync)
-        await extract_facts_from_contents_batch_api(
-            contents=test_contents,
-            llm_config=mock_llm_config,
-            agent_name="test_agent",
-            config=hindsight_config,
-            pool=None,
-            operation_id=None,
-            schema=None,
-        )
-
-        # Verify fallback occurred
-        mock_sync_extract.assert_called_once()
-
-        # Verify batch API methods were NOT called
-        mock_llm_config._provider_impl.submit_batch.assert_not_called()
-
-        logger.info("✅ Fallback to sync mode test passed")
-
-
-@pytest.mark.asyncio
-async def test_worker_batch_recovery(memory, request_context):
-    """Test that WorkerPoller._recover_batch_operations finds and resets orphaned batches."""
-    bank_id = f"test_worker_recovery_{datetime.now(timezone.utc).timestamp()}"
-    operation_id = str(uuid.uuid4())  # Must be UUID for async_operations table
-
-    try:
-        # Ensure bank exists
-        await memory.get_bank_profile(bank_id, request_context=request_context)
-
-        pool = memory._pool
-        schema = request_context.tenant_id
-
-        from hindsight_api.engine.task_backend import fq_table
-        table = fq_table("async_operations", schema)
-
-        # Create orphaned batch operation (simulates worker crash during polling)
-        batch_id = "batch_orphaned_999"
-        task_payload = {
-            "operation_type": "retain",
-            "bank_id": bank_id,
-            "contents": [{"content": "test", "event_date": "2024-01-15T00:00:00Z"}],
-        }
-
-        await pool.execute(
-            f"""
-            INSERT INTO {table} (operation_id, operation_type, bank_id, status, worker_id, result_metadata, task_payload)
-            VALUES ($1, 'retain', $2, 'processing', 'worker_crashed', $3::jsonb, $4::jsonb)
-            """,
-            operation_id,
-            bank_id,
-            json.dumps({
-                "batch_id": batch_id,
-                "batch_provider": "openai",
-                "chunk_count": 1,
-            }),
-            json.dumps(task_payload),
-        )
-
-        # Create WorkerPoller
-        from hindsight_api.extensions.builtin.tenant import DefaultTenantExtension
-        tenant_extension = DefaultTenantExtension(config={"schema": schema} if schema else {})
-
-        poller = WorkerPoller(
-            pool=pool,
-            worker_id="test_worker_recovery",
-            executor=memory,
-            poll_interval_ms=100,
-            schema=schema,
-            tenant_extension=tenant_extension,
-            max_slots=5,
-            consolidation_max_slots=2,
-        )
-
-        # Run recovery
-        recovered_count = await poller._recover_batch_operations(schema)
-
-        # Verify recovery
-        assert recovered_count == 1, "Should recover 1 batch operation"
-
-        # Verify operation was reset to pending
-        row = await pool.fetchrow(
-            f"SELECT status, worker_id FROM {table} WHERE operation_id = $1",
-            operation_id,
-        )
-
-        assert row["status"] == "pending", "Operation should be reset to pending"
-        assert row["worker_id"] is None, "Worker ID should be cleared"
-
-        logger.info("✅ Worker batch recovery test passed")
-
-    finally:
-        # Cleanup
-        try:
-            await memory.delete_bank(bank_id, request_context=request_context)
-        except Exception:
-            pass
-
-
-@pytest.mark.asyncio
-async def test_batch_api_via_extract_facts_from_contents(
-    mock_llm_config, test_contents, hindsight_config, memory, request_context
-):
-    """Test that extract_facts_from_contents routes to batch API when enabled."""
-    bank_id = f"test_routing_{datetime.now(timezone.utc).timestamp()}"
-
-    try:
-        # Enable batch API in config
-        hindsight_config.retain_batch_enabled = True
-
-        # Mock batch API support
-        mock_llm_config._provider_impl.supports_batch_api = AsyncMock(return_value=True)
-        mock_llm_config._provider_impl.submit_batch = AsyncMock(
-            return_value={"batch_id": "batch_123", "status": "validating", "request_counts": {}}
-        )
-        mock_llm_config._provider_impl.get_batch_status = AsyncMock(
-            return_value={"status": "completed", "request_counts": {"total": 1, "completed": 1, "failed": 0}}
-        )
-        mock_llm_config._provider_impl.retrieve_batch_results = AsyncMock(
-            return_value=[
-                {
-                    "custom_id": "chunk_0",
-                    "response": {
-                        "body": {
-                            "choices": [
-                                {
-                                    "message": {
-                                        "content": json.dumps({"facts": []})
-                                    }
-                                }
-                            ],
-                            "usage": {"prompt_tokens": 10, "completion_tokens": 5, "total_tokens": 15},
-                        }
-                    },
-                }
-            ]
-        )
-
-        # Call main extract_facts_from_contents (should route to batch API)
-        facts, chunks, usage = await extract_facts_from_contents(
-            contents=test_contents,
-            llm_config=mock_llm_config,
-            agent_name="test_agent",
-            config=hindsight_config,
-            pool=None,
-            operation_id=None,
-            schema=None,
-        )
-
-        # Verify batch API was called
-        mock_llm_config._provider_impl.submit_batch.assert_called_once()
-
-        logger.info("✅ Routing to batch API test passed")
-
-    finally:
-        # Cleanup
-        try:
-            await memory.delete_bank(bank_id, request_context=request_context)
-        except Exception:
-            pass

hindsight-api-slim/tests/test_batch_api_integration.py

@@ -1,263 +0,0 @@
-"""
-Real integration test for OpenAI Batch API.
-
-This test makes REAL API calls to OpenAI and measures actual timing.
-It will be slow (minutes to hours) depending on OpenAI's queue.
-
-To run:
-    pytest tests/test_batch_api_integration.py -v -s
-
-To skip in CI:
-    Add @pytest.mark.skip at the test level
-"""
-import pytest
-import os
-import asyncio
-import logging
-import time
-from datetime import datetime, timezone
-from dotenv import load_dotenv
-from hindsight_api import RequestContext
-from hindsight_api.engine.retain.fact_extraction import (
-    extract_facts_from_contents_batch_api,
-    RetainContent,
-)
-from hindsight_api.config import HindsightConfig
-from hindsight_api.engine.llm_wrapper import LLMProvider
-
-logger = logging.getLogger(__name__)
-
-# Load .env file for API keys
-load_dotenv()
-
-
-@pytest.fixture
-def openai_api_key():
-    """Get OpenAI API key from environment."""
-    # Try both current and commented keys from .env
-    api_key = os.getenv("HINDSIGHT_API_LLM_API_KEY")
-
-    # Check if it's an OpenAI key (starts with sk-proj- or sk-)
-    if not api_key or not api_key.startswith("sk-"):
-        # Try the OpenAI-specific env var (if set separately)
-        api_key = os.getenv("OPENAI_API_KEY")
-
-    if not api_key or not api_key.startswith("sk-"):
-        pytest.skip("OpenAI API key not found in environment. Set OPENAI_API_KEY or uncomment OpenAI config in .env")
-
-    return api_key
-
-
-@pytest.fixture
-def real_llm_config(openai_api_key):
-    """Create real LLM config for OpenAI."""
-    # Create config with OpenAI settings
-    config = HindsightConfig.from_env()
-
-    # Use LLMProvider wrapper (which creates _provider_impl internally)
-    llm_config = LLMProvider(
-        provider="openai",
-        api_key=openai_api_key,
-        base_url="https://api.openai.com/v1",
-        model="gpt-4o-mini",  # Fast, cheap model for testing
-        reasoning_effort="medium",  # Required parameter
-    )
-
-    return llm_config
-
-
-@pytest.fixture
-def test_contents_real():
-    """Create realistic test content for fact extraction."""
-    return [
-        RetainContent(
-            content="""
-            Alice is a senior software engineer at TechCorp, where she has been working for 5 years.
-            She specializes in distributed systems and microservices architecture. Alice graduated
-            from MIT with a degree in Computer Science in 2015. She is known for writing clean,
-            well-documented code and mentoring junior developers.
-            """,
-            event_date=datetime(2024, 1, 15, 10, 30, tzinfo=timezone.utc),
-            context="team member profile",
-        ),
-        RetainContent(
-            content="""
-            Bob joined TechCorp last month as a junior developer. He is learning React and Node.js
-            and recently completed his first feature, which was a user authentication flow. Bob
-            graduated from Berkeley with a degree in Computer Science in 2023. He is enthusiastic
-            and asks great questions during code reviews.
-            """,
-            event_date=datetime(2024, 1, 15, 10, 30, tzinfo=timezone.utc),
-            context="team member profile",
-        ),
-        RetainContent(
-            content="""
-            The team uses Kubernetes for container orchestration and deploys to AWS. They follow
-            agile methodologies with two-week sprints. Code reviews are mandatory before merging
-            any pull request. The team meets every morning for a 15-minute standup to discuss
-            progress and blockers.
-            """,
-            event_date=datetime(2024, 1, 15, 10, 30, tzinfo=timezone.utc),
-            context="team processes",
-        ),
-    ]
-
-
-@pytest.fixture
-def integration_config():
-    """Create config for integration test."""
-    config = HindsightConfig.from_env()
-    config.retain_batch_enabled = True
-    config.retain_batch_poll_interval_seconds = 30  # Poll every 30 seconds (reasonable for real API)
-    config.retain_chunk_size = 4000
-    config.retain_extraction_mode = "concise"
-    config.retain_extract_causal_links = False
-    return config
-
-
-@pytest.mark.skip(reason="Real API test - takes minutes and costs money. Run manually with: pytest tests/test_batch_api_integration.py::test_real_openai_batch_api -v -s")
-@pytest.mark.integration  # Mark as integration test
-@pytest.mark.slow  # Mark as slow test
-@pytest.mark.asyncio
-async def test_real_openai_batch_api(real_llm_config, test_contents_real, integration_config, memory, request_context):
-    """
-    REAL integration test: Submit actual batch to OpenAI and measure timing.
-
-    WARNING: This test:
-    - Makes real API calls to OpenAI
-    - Will take minutes to hours to complete
-    - Costs money (though very little with gpt-4o-mini)
-    - Requires valid OpenAI API key
-
-    To skip this test:
-        pytest tests/test_batch_api_integration.py --skip-integration
-    """
-    bank_id = f"test_real_batch_{datetime.now(timezone.utc).timestamp()}"
-
-    logger.info("=" * 80)
-    logger.info("STARTING REAL OPENAI BATCH API INTEGRATION TEST")
-    logger.info("=" * 80)
-    logger.info(f"Test contents: {len(test_contents_real)} items")
-    logger.info(f"Poll interval: {integration_config.retain_batch_poll_interval_seconds}s")
-    logger.info(f"Model: {real_llm_config.model}")
-    logger.info("This may take several minutes to hours depending on OpenAI's queue...")
-    logger.info("=" * 80)
-
-    try:
-        # Ensure bank exists
-        await memory.get_bank_profile(bank_id, request_context=request_context)
-
-        # Get database pool and schema for crash recovery testing
-        pool = memory._pool
-        schema = request_context.tenant_id
-
-        # Track overall timing
-        test_start_time = time.time()
-
-        # Call REAL batch API extraction
-        logger.info("\n📤 Submitting batch to OpenAI...")
-
-        facts, chunks, usage = await extract_facts_from_contents_batch_api(
-            contents=test_contents_real,
-            llm_config=real_llm_config,
-            agent_name="test_agent",
-            config=integration_config,
-            pool=pool,
-            operation_id=None,  # No crash recovery for this test
-            schema=schema,
-        )
-
-        test_end_time = time.time()
-        total_duration = test_end_time - test_start_time
-
-        # Log results
-        logger.info("\n" + "=" * 80)
-        logger.info("✅ BATCH COMPLETED SUCCESSFULLY")
-        logger.info("=" * 80)
-        logger.info(f"Total duration: {total_duration:.1f} seconds ({total_duration/60:.1f} minutes)")
-        logger.info(f"Facts extracted: {len(facts)}")
-        logger.info(f"Chunks processed: {len(chunks)}")
-        logger.info(f"Token usage: {usage.input_tokens} input + {usage.output_tokens} output = {usage.total_tokens} total")
-        logger.info(f"Estimated cost: ${(usage.input_tokens * 0.00015 / 1000 + usage.output_tokens * 0.0006 / 1000):.4f}")
-        logger.info("=" * 80)
-
-        # Log sample facts
-        logger.info("\n📋 Sample extracted facts:")
-        for i, fact in enumerate(facts[:5]):  # Show first 5 facts
-            logger.info(f"\nFact {i+1}:")
-            logger.info(f"  Type: {fact.fact_type}")
-            logger.info(f"  Text: {fact.fact_text[:100]}...")
-            logger.info(f"  Entities: {fact.entities}")
-
-        # Verify results
-        assert len(facts) > 0, "Should extract at least some facts"
-        assert len(chunks) == len(test_contents_real), f"Should have {len(test_contents_real)} chunks"
-        assert usage.total_tokens > 0, "Should have token usage"
-
-        # Verify fact structure
-        for fact in facts:
-            assert hasattr(fact, "fact_text"), "Fact should have fact_text"
-            assert hasattr(fact, "fact_type"), "Fact should have fact_type"
-            assert fact.fact_type in ["world", "experience", "opinion"], f"Invalid fact_type: {fact.fact_type}"
-
-        logger.info("\n✅ All assertions passed!")
-
-        # Write timing report to file for later analysis
-        report_path = "/tmp/openai_batch_api_timing_report.txt"
-        with open(report_path, "w") as f:
-            f.write(f"OpenAI Batch API Integration Test Report\n")
-            f.write(f"={'=' * 60}\n\n")
-            f.write(f"Test Date: {datetime.now(timezone.utc).isoformat()}\n")
-            f.write(f"Model: {real_llm_config.model}\n")
-            f.write(f"Contents: {len(test_contents_real)} items\n")
-            f.write(f"Poll Interval: {integration_config.retain_batch_poll_interval_seconds}s\n\n")
-            f.write(f"Results:\n")
-            f.write(f"  Total Duration: {total_duration:.1f}s ({total_duration/60:.1f} min)\n")
-            f.write(f"  Facts Extracted: {len(facts)}\n")
-            f.write(f"  Chunks Processed: {len(chunks)}\n")
-            f.write(f"  Token Usage: {usage.total_tokens} ({usage.input_tokens} in + {usage.output_tokens} out)\n")
-            f.write(f"  Estimated Cost: ${(usage.input_tokens * 0.00015 / 1000 + usage.output_tokens * 0.0006 / 1000):.4f}\n")
-
-        logger.info(f"\n📄 Timing report written to: {report_path}")
-
-    finally:
-        # Cleanup
-        try:
-            await memory.delete_bank(bank_id, request_context=request_context)
-            logger.info(f"\n🧹 Cleaned up test bank: {bank_id}")
-        except Exception as e:
-            logger.error(f"Failed to cleanup bank: {e}")
-
-
-@pytest.mark.skip(reason="Real API test - requires Groq API key. Run manually if needed.")
-@pytest.mark.integration
-@pytest.mark.slow
-@pytest.mark.asyncio
-async def test_real_batch_supports_groq(integration_config):
-    """
-    Test that Groq also supports batch API (if configured).
-
-    Groq has the same batch API interface as OpenAI.
-    """
-    groq_api_key = os.getenv("HINDSIGHT_API_LLM_API_KEY")
-
-    if not groq_api_key or not groq_api_key.startswith("gsk_"):
-        pytest.skip("Groq API key not found in environment")
-
-    llm_config = LLMProvider(
-        provider="groq",
-        api_key=groq_api_key,
-        base_url="https://api.groq.com/openai/v1",
-        model="llama-3.1-8b-instant",
-        reasoning_effort="medium",
-    )
-
-    # Check if Groq supports batch API
-    supports_batch = await llm_config._provider_impl.supports_batch_api()
-
-    logger.info(f"Groq batch API support: {supports_batch}")
-
-    # Groq should support batch API (same interface as OpenAI)
-    assert supports_batch, "Groq should support batch API"
-
-    logger.info("✅ Groq batch API support confirmed")

hindsight-api-slim/tests/test_batch_api_validation.py

@@ -1,38 +0,0 @@
-"""
-Test validation for batch API + synchronous retain.
-
-When HINDSIGHT_API_RETAIN_BATCH_ENABLED=true, synchronous retain operations
-should be rejected with a 400 error since they will timeout.
-"""
-
-import os
-import pytest
-from hindsight_api.engine.memory_engine import MemoryEngine
-from hindsight_api.config import HindsightConfig
-from hindsight_api import RequestContext
-
-
-@pytest.mark.asyncio
-async def test_batch_api_validation(memory, request_context):
-    """
-    Test that attempting synchronous retain with batch API enabled
-    raises an error at the HTTP layer.
-
-    This test verifies the validation logic exists - actual HTTP testing
-    would require full FastAPI app setup.
-    """
-    # Create config with batch API enabled
-    config = HindsightConfig.from_env()
-    config.retain_batch_enabled = True
-    config.retain_batch_poll_interval_seconds = 1
-
-    # Verify the validation exists in memory engine
-    # The actual HTTP validation happens in http.py api_retain()
-    # This test documents the expected behavior
-
-    assert config.retain_batch_enabled is True
-    assert config.retain_batch_poll_interval_seconds == 1
-
-    # When batch API is enabled and async=false, the HTTP endpoint
-    # should return 400 with message:
-    # "Batch API is enabled (HINDSIGHT_API_RETAIN_BATCH_ENABLED=true) but async=false"

hindsight-api-slim/tests/test_combined_scoring.py

@@ -1,171 +0,0 @@
-"""
-Tests for combined scoring (apply_combined_scoring).
-
-The function applies multiplicative recency/temporal boosts to the cross-encoder
-score so that the relative influence of these signals is proportional to the base
-relevance score, independent of the cross-encoder model's score calibration.
-"""
-
-from datetime import datetime, timedelta, timezone
-from unittest.mock import MagicMock
-
-import pytest
-
-from hindsight_api.engine.search.reranking import apply_combined_scoring, _RECENCY_ALPHA, _TEMPORAL_ALPHA
-from hindsight_api.engine.search.types import MergedCandidate, RetrievalResult, ScoredResult
-
-UTC = timezone.utc
-NOW = datetime(2024, 6, 1, tzinfo=UTC)
-
-
-def _make_result(
-    ce_norm: float,
-    occurred_start: datetime | None = None,
-    temporal_proximity: float | None = None,
-) -> ScoredResult:
-    retrieval = MagicMock(spec=RetrievalResult)
-    retrieval.occurred_start = occurred_start
-    retrieval.temporal_proximity = temporal_proximity
-
-    candidate = MagicMock(spec=MergedCandidate)
-    candidate.retrieval = retrieval
-    candidate.rrf_score = 0.05
-
-    return ScoredResult(
-        candidate=candidate,
-        cross_encoder_score=1.0,
-        cross_encoder_score_normalized=ce_norm,
-        weight=ce_norm,
-    )
-
-
-class TestBoostFormula:
-    def test_neutral_signals_leave_score_unchanged(self):
-        """recency=0.5 and temporal=0.5 both produce boost=1.0, so weight == ce."""
-        sr = _make_result(ce_norm=0.6)
-        apply_combined_scoring([sr], now=NOW)
-        assert abs(sr.weight - 0.6) < 1e-9
-
-    def test_max_recency_boost(self):
-        """A memory from today (recency≈1.0) should boost by (1 + alpha*0.5)."""
-        sr = _make_result(ce_norm=0.5, occurred_start=NOW)
-        apply_combined_scoring([sr], now=NOW)
-        expected = 0.5 * (1.0 + _RECENCY_ALPHA * 0.5) * 1.0  # temporal neutral
-        assert abs(sr.weight - expected) < 1e-6
-
-    def test_min_recency_penalty(self):
-        """A memory from >365 days ago (recency=0.1) should penalise score."""
-        old = NOW - timedelta(days=400)
-        sr = _make_result(ce_norm=0.5, occurred_start=old)
-        apply_combined_scoring([sr], now=NOW)
-        expected = 0.5 * (1.0 + _RECENCY_ALPHA * (0.1 - 0.5)) * 1.0
-        assert abs(sr.weight - expected) < 1e-6
-
-    def test_max_temporal_boost(self):
-        """temporal_proximity=1.0 should boost by (1 + alpha*0.5)."""
-        sr = _make_result(ce_norm=0.5, temporal_proximity=1.0)
-        apply_combined_scoring([sr], now=NOW)
-        expected = 0.5 * 1.0 * (1.0 + _TEMPORAL_ALPHA * 0.5)  # recency neutral
-        assert abs(sr.weight - expected) < 1e-6
-
-    def test_temporal_none_is_neutral(self):
-        """temporal_proximity=None must be treated as 0.5 (no boost/penalty)."""
-        sr_none = _make_result(ce_norm=0.5, temporal_proximity=None)
-        sr_half = _make_result(ce_norm=0.5, temporal_proximity=0.5)
-        apply_combined_scoring([sr_none], now=NOW)
-        apply_combined_scoring([sr_half], now=NOW)
-        assert abs(sr_none.weight - sr_half.weight) < 1e-9
-
-    def test_both_signals_combined(self):
-        """Both boosts are applied multiplicatively."""
-        sr = _make_result(ce_norm=0.5, occurred_start=NOW, temporal_proximity=1.0)
-        apply_combined_scoring([sr], now=NOW)
-        recency_boost = 1.0 + _RECENCY_ALPHA * (1.0 - 0.5)
-        temporal_boost = 1.0 + _TEMPORAL_ALPHA * (1.0 - 0.5)
-        expected = 0.5 * recency_boost * temporal_boost
-        assert abs(sr.weight - expected) < 1e-6
-
-    def test_boost_is_proportional_to_ce(self):
-        """The absolute boost from recency scales with the CE score."""
-        sr_high = _make_result(ce_norm=0.9, occurred_start=NOW)
-        sr_low = _make_result(ce_norm=0.3, occurred_start=NOW)
-        apply_combined_scoring([sr_high, sr_low], now=NOW)
-
-        # Both get the same recency boost factor — absolute gain is proportional to CE
-        boost_factor = 1.0 + _RECENCY_ALPHA * 0.5
-        assert abs(sr_high.weight - 0.9 * boost_factor) < 1e-6
-        assert abs(sr_low.weight - 0.3 * boost_factor) < 1e-6
-
-    def test_boost_capped(self):
-        """Max boost: recency=1.0 + temporal=1.0 gives ≤21% uplift on CE."""
-        sr = _make_result(ce_norm=1.0, occurred_start=NOW, temporal_proximity=1.0)
-        apply_combined_scoring([sr], now=NOW)
-        assert sr.weight <= 1.0 * (1 + _RECENCY_ALPHA / 2) * (1 + _TEMPORAL_ALPHA / 2) + 1e-9
-
-    def test_rrf_normalized_always_zero(self):
-        """RRF is excluded from scoring; rrf_normalized is set to 0.0 for trace clarity."""
-        sr = _make_result(ce_norm=0.5)
-        apply_combined_scoring([sr], now=NOW)
-        assert sr.rrf_normalized == 0.0
-
-    def test_combined_score_equals_weight(self):
-        """combined_score and weight must stay in sync."""
-        sr = _make_result(ce_norm=0.7, occurred_start=NOW, temporal_proximity=0.8)
-        apply_combined_scoring([sr], now=NOW)
-        assert sr.combined_score == sr.weight
-
-    def test_model_calibration_independence(self):
-        """
-        A low-calibration model (low CE scores) and a high-calibration model
-        (high CE scores) should produce the same ranking for identical content.
-
-        With additive scoring the recency term would dominate for low-CE models;
-        with multiplicative boosting the relative ranking is stable.
-        """
-        recent = NOW - timedelta(days=10)
-        old = NOW - timedelta(days=300)
-
-        # High-calibration model: clear winner is #1 (more relevant, slightly older)
-        h_relevant = _make_result(ce_norm=0.85, occurred_start=old)
-        h_recent = _make_result(ce_norm=0.60, occurred_start=recent)
-        apply_combined_scoring([h_relevant, h_recent], now=NOW)
-        assert h_relevant.weight > h_recent.weight, "High-CE model: relevance should win"
-
-        # Low-calibration model: same relative difference, just compressed scores
-        l_relevant = _make_result(ce_norm=0.34, occurred_start=old)
-        l_recent = _make_result(ce_norm=0.24, occurred_start=recent)
-        apply_combined_scoring([l_relevant, l_recent], now=NOW)
-        assert l_relevant.weight > l_recent.weight, "Low-CE model: relevance should still win"
-
-    def test_no_occurred_start_defaults_recency_neutral(self):
-        """Missing occurred_start → recency=0.5 → no boost/penalty."""
-        sr = _make_result(ce_norm=0.5, occurred_start=None)
-        apply_combined_scoring([sr], now=NOW)
-        assert sr.recency == 0.5
-        assert abs(sr.weight - 0.5) < 1e-9
-
-    def test_timezone_naive_occurred_start_handled(self):
-        """Naive datetimes in occurred_start should not raise."""
-        naive_date = datetime(2024, 1, 1)  # no tzinfo
-        sr = _make_result(ce_norm=0.5, occurred_start=naive_date)
-        apply_combined_scoring([sr], now=NOW)  # must not raise
-        assert 0.0 < sr.weight < 1.0
-
-    def test_custom_alpha_values(self):
-        """Custom alpha parameters are respected."""
-        sr = _make_result(ce_norm=0.5, occurred_start=NOW)
-        apply_combined_scoring([sr], now=NOW, recency_alpha=0.4, temporal_alpha=0.0)
-        expected = 0.5 * (1.0 + 0.4 * 0.5) * 1.0
-        assert abs(sr.weight - expected) < 1e-6
-
-    def test_future_event_recency_capped_at_one(self):
-        """Events in the future must not produce recency > 1.0, keeping boost within bounds."""
-        future = NOW + timedelta(days=180)
-        sr = _make_result(ce_norm=0.5, occurred_start=future)
-        apply_combined_scoring([sr], now=NOW)
-        assert sr.recency == 1.0
-        expected_max_boost = 1.0 + _RECENCY_ALPHA * 0.5
-        assert sr.weight <= 0.5 * expected_max_boost + 1e-9
-
-    def test_empty_list_is_noop(self):
-        apply_combined_scoring([], now=NOW)  # must not raise

hindsight-api-slim/tests/test_consolidation_failure_recovery.py

@@ -1,476 +0,0 @@
-"""Tests for consolidation failure handling: adaptive batch splitting, consolidation_failed_at,
-and the recovery API.
-
-These tests use a mock LLM to simulate LLM failures deterministically, without making real
-API calls. All tests insert memories directly into the database to bypass retain's LLM calls
-and focus exclusively on the consolidation code paths.
-"""
-
-import uuid
-from unittest.mock import MagicMock
-
-import pytest
-import pytest_asyncio
-
-from hindsight_api.engine.consolidation.consolidator import run_consolidation_job
-from hindsight_api.engine.memory_engine import MemoryEngine
-from hindsight_api.engine.providers.mock_llm import MockLLM
-from hindsight_api.engine.task_backend import SyncTaskBackend
-
-
-@pytest_asyncio.fixture(scope="function")
-async def memory_no_llm_verify(pg0_db_url, embeddings, cross_encoder, query_analyzer):
-    """MemoryEngine with mock LLM.
-
-    Migrations are already applied by the session-scoped pg0_db_url fixture, so
-    run_migrations=False avoids advisory-lock serialization overhead per test.
-    """
-    mem = MemoryEngine(
-        db_url=pg0_db_url,
-        memory_llm_provider="mock",
-        memory_llm_api_key="",
-        memory_llm_model="mock",
-        embeddings=embeddings,
-        cross_encoder=cross_encoder,
-        query_analyzer=query_analyzer,
-        pool_min_size=1,
-        pool_max_size=5,
-        run_migrations=False,
-        task_backend=SyncTaskBackend(),
-        skip_llm_verification=True,
-    )
-    await mem.initialize()
-    yield mem
-    try:
-        if mem._pool and not mem._pool._closing:
-            await mem.close()
-    except Exception:
-        pass
-
-
-@pytest.fixture(autouse=True)
-def enable_observations():
-    """Enable observations for all tests in this module."""
-    from hindsight_api.config import _get_raw_config
-
-    config = _get_raw_config()
-    original = config.enable_observations
-    config.enable_observations = True
-    yield
-    config.enable_observations = original
-
-
-def _make_failing_mock_llm(*, fail_first_n: int = 999) -> MockLLM:
-    """Return a MockLLM that raises ValueError for the first `fail_first_n` consolidation calls."""
-    mock_llm = MockLLM(provider="mock", api_key="", base_url="", model="mock-model")
-    call_count = 0
-
-    def callback(messages, scope):
-        nonlocal call_count
-        if scope == "consolidation":
-            call_count += 1
-            if call_count <= fail_first_n:
-                raise ValueError(f"Simulated LLM failure (call {call_count})")
-        # Return empty response — no creates/updates/deletes
-        from hindsight_api.engine.consolidation.consolidator import _ConsolidationBatchResponse
-
-        return _ConsolidationBatchResponse()
-
-    mock_llm.set_response_callback(callback)
-    return mock_llm
-
-
-def _make_always_success_mock_llm() -> MockLLM:
-    """Return a MockLLM that always succeeds with an empty consolidation response."""
-    mock_llm = MockLLM(provider="mock", api_key="", base_url="", model="mock-model")
-
-    def callback(messages, scope):
-        from hindsight_api.engine.consolidation.consolidator import _ConsolidationBatchResponse
-
-        return _ConsolidationBatchResponse()
-
-    mock_llm.set_response_callback(callback)
-    return mock_llm
-
-
-def _inject_mock_llm(memory: MemoryEngine, mock_llm: MockLLM) -> None:
-    """Replace memory._consolidation_llm_config with a wrapper that returns mock_llm from with_config."""
-    wrapper = MagicMock()
-    wrapper.with_config.return_value = mock_llm
-    memory._consolidation_llm_config = wrapper
-
-
-async def _insert_memories(conn, bank_id: str, texts: list[str]) -> list[uuid.UUID]:
-    """Insert experience memories directly, bypassing LLM-based retain."""
-    ids = []
-    for text in texts:
-        mem_id = uuid.uuid4()
-        await conn.execute(
-            """
-            INSERT INTO memory_units (id, bank_id, text, fact_type, created_at)
-            VALUES ($1, $2, $3, 'experience', now())
-            """,
-            mem_id,
-            bank_id,
-            text,
-        )
-        ids.append(mem_id)
-    return ids
-
-
-class TestAdaptiveBatchSplitting:
-    """Verify that a failing batch is halved and retried until batch_size=1 succeeds."""
-
-    @pytest.mark.asyncio
-    async def test_splitting_recovers_all_memories(self, memory_no_llm_verify: MemoryEngine, request_context):
-        """When a batch of 2 fails, both are retried individually and succeed."""
-        bank_id = f"test-split-recovery-{uuid.uuid4().hex[:8]}"
-        await memory_no_llm_verify.get_bank_profile(bank_id=bank_id, request_context=request_context)
-
-        async with memory_no_llm_verify._pool.acquire() as conn:
-            mem_ids = await _insert_memories(
-                conn,
-                bank_id,
-                [
-                    "Alice runs marathons every spring.",
-                    "Alice trained for six months for her last race.",
-                ],
-            )
-
-        # Exhaust all 3 retries for batch=2 (calls 1-3 fail), then each batch=1 succeeds (calls 4-5)
-        mock_llm = _make_failing_mock_llm(fail_first_n=3)
-        _inject_mock_llm(memory_no_llm_verify, mock_llm)
-
-        result = await run_consolidation_job(
-            memory_engine=memory_no_llm_verify,
-            bank_id=bank_id,
-            request_context=request_context,
-        )
-
-        assert result["status"] == "completed"
-        assert result["memories_processed"] == 2
-        assert result["memories_failed"] == 0
-
-        # Both memories must have consolidated_at set and consolidation_failed_at NULL
-        async with memory_no_llm_verify._pool.acquire() as conn:
-            rows = await conn.fetch(
-                """
-                SELECT id, consolidated_at, consolidation_failed_at
-                FROM memory_units
-                WHERE bank_id = $1 AND fact_type = 'experience'
-                """,
-                bank_id,
-            )
-        assert len(rows) == 2
-        for row in rows:
-            assert row["consolidated_at"] is not None, f"Memory {row['id']} should have consolidated_at set"
-            assert row["consolidation_failed_at"] is None, (
-                f"Memory {row['id']} should NOT have consolidation_failed_at set"
-            )
-
-        # LLM called 5 times: 3 retries failed (batch=2) + 1 succeeded (batch=1) + 1 succeeded (batch=1)
-        consolidation_calls = [c for c in mock_llm.get_mock_calls() if c["scope"] == "consolidation"]
-        assert len(consolidation_calls) == 5
-
-        await memory_no_llm_verify.delete_bank(bank_id, request_context=request_context)
-
-    @pytest.mark.asyncio
-    async def test_splitting_with_larger_batch(self, memory_no_llm_verify: MemoryEngine, request_context):
-        """A batch of 4 that always fails at size>1 resolves to 4 individual calls."""
-        bank_id = f"test-split-large-{uuid.uuid4().hex[:8]}"
-        await memory_no_llm_verify.get_bank_profile(bank_id=bank_id, request_context=request_context)
-
-        async with memory_no_llm_verify._pool.acquire() as conn:
-            await _insert_memories(
-                conn,
-                bank_id,
-                [
-                    "Bob plays chess competitively.",
-                    "Bob won a regional chess tournament.",
-                    "Bob practices tactics every morning.",
-                    "Bob coaches youth chess on weekends.",
-                ],
-            )
-
-        # Exhaust all 3 retries for batch=4 (calls 1-3 fail), then both batch=2 halves succeed
-        # (calls 4-5). This verifies that halving once is sufficient when batch=2 works.
-        mock_llm = _make_failing_mock_llm(fail_first_n=3)
-        _inject_mock_llm(memory_no_llm_verify, mock_llm)
-
-        result = await run_consolidation_job(
-            memory_engine=memory_no_llm_verify,
-            bank_id=bank_id,
-            request_context=request_context,
-        )
-
-        assert result["memories_processed"] == 4
-        assert result["memories_failed"] == 0
-
-        async with memory_no_llm_verify._pool.acquire() as conn:
-            rows = await conn.fetch(
-                "SELECT consolidated_at, consolidation_failed_at FROM memory_units "
-                "WHERE bank_id = $1 AND fact_type = 'experience'",
-                bank_id,
-            )
-        assert all(r["consolidated_at"] is not None for r in rows)
-        assert all(r["consolidation_failed_at"] is None for r in rows)
-
-        await memory_no_llm_verify.delete_bank(bank_id, request_context=request_context)
-
-
-class TestConsolidationFailedAt:
-    """Verify that consolidation_failed_at is set — and consolidated_at is NOT — when all retries fail."""
-
-    @pytest.mark.asyncio
-    async def test_single_memory_permanent_failure(self, memory_no_llm_verify: MemoryEngine, request_context):
-        """A single memory that exhausts all LLM retries gets consolidation_failed_at, not consolidated_at."""
-        bank_id = f"test-perm-fail-{uuid.uuid4().hex[:8]}"
-        await memory_no_llm_verify.get_bank_profile(bank_id=bank_id, request_context=request_context)
-
-        async with memory_no_llm_verify._pool.acquire() as conn:
-            (mem_id,) = await _insert_memories(conn, bank_id, ["Carol enjoys painting watercolors."])
-
-        # Always fail
-        mock_llm = _make_failing_mock_llm(fail_first_n=999)
-        _inject_mock_llm(memory_no_llm_verify, mock_llm)
-
-        result = await run_consolidation_job(
-            memory_engine=memory_no_llm_verify,
-            bank_id=bank_id,
-            request_context=request_context,
-        )
-
-        assert result["memories_failed"] == 1
-        assert result["memories_processed"] == 1
-
-        async with memory_no_llm_verify._pool.acquire() as conn:
-            row = await conn.fetchrow(
-                "SELECT consolidated_at, consolidation_failed_at FROM memory_units WHERE id = $1",
-                mem_id,
-            )
-
-        assert row["consolidated_at"] is None, "consolidated_at must NOT be set for a permanently failed memory"
-        assert row["consolidation_failed_at"] is not None, "consolidation_failed_at must be set"
-
-        await memory_no_llm_verify.delete_bank(bank_id, request_context=request_context)
-
-    @pytest.mark.asyncio
-    async def test_failed_memory_excluded_from_next_run(self, memory_no_llm_verify: MemoryEngine, request_context):
-        """A memory marked consolidation_failed_at is not re-processed on the next consolidation run."""
-        bank_id = f"test-excluded-{uuid.uuid4().hex[:8]}"
-        await memory_no_llm_verify.get_bank_profile(bank_id=bank_id, request_context=request_context)
-
-        async with memory_no_llm_verify._pool.acquire() as conn:
-            (mem_id,) = await _insert_memories(conn, bank_id, ["Dave collects vinyl records."])
-            # Manually stamp consolidation_failed_at to simulate a prior failed run
-            await conn.execute(
-                "UPDATE memory_units SET consolidation_failed_at = NOW() WHERE id = $1",
-                mem_id,
-            )
-
-        # Even with a healthy LLM, the memory should be skipped
-        mock_llm = _make_always_success_mock_llm()
-        _inject_mock_llm(memory_no_llm_verify, mock_llm)
-
-        result = await run_consolidation_job(
-            memory_engine=memory_no_llm_verify,
-            bank_id=bank_id,
-            request_context=request_context,
-        )
-
-        # No unconsolidated memories to pick up (consolidation_failed_at ≠ NULL, consolidated_at = NULL
-        # but the SELECT filters on consolidated_at IS NULL AND fact_type IN ('experience','world'))
-        assert result["status"] in ("no_new_memories", "completed")
-        if result["status"] == "completed":
-            assert result["memories_processed"] == 0
-
-        # Memory still has consolidation_failed_at set and consolidated_at NULL
-        async with memory_no_llm_verify._pool.acquire() as conn:
-            row = await conn.fetchrow(
-                "SELECT consolidated_at, consolidation_failed_at FROM memory_units WHERE id = $1",
-                mem_id,
-            )
-        assert row["consolidated_at"] is None
-        assert row["consolidation_failed_at"] is not None
-
-        await memory_no_llm_verify.delete_bank(bank_id, request_context=request_context)
-
-    @pytest.mark.asyncio
-    async def test_partial_batch_failure(self, memory_no_llm_verify: MemoryEngine, request_context):
-        """In a batch of 2, if only the first individual retry fails, the second still succeeds."""
-        bank_id = f"test-partial-fail-{uuid.uuid4().hex[:8]}"
-        await memory_no_llm_verify.get_bank_profile(bank_id=bank_id, request_context=request_context)
-
-        async with memory_no_llm_verify._pool.acquire() as conn:
-            mem_ids = await _insert_memories(
-                conn,
-                bank_id,
-                [
-                    "Eve speaks three languages fluently.",
-                    "Eve learned Japanese in two years.",
-                ],
-            )
-
-        # Exhaust 3 retries for batch=2 (calls 1-3), exhaust 3 retries for first batch=1 (calls 4-6),
-        # second batch=1 succeeds (call 7)
-        mock_llm = _make_failing_mock_llm(fail_first_n=6)
-        _inject_mock_llm(memory_no_llm_verify, mock_llm)
-
-        result = await run_consolidation_job(
-            memory_engine=memory_no_llm_verify,
-            bank_id=bank_id,
-            request_context=request_context,
-        )
-
-        assert result["memories_processed"] == 2
-        assert result["memories_failed"] == 1
-
-        async with memory_no_llm_verify._pool.acquire() as conn:
-            rows = {
-                str(r["id"]): r
-                for r in await conn.fetch(
-                    "SELECT id, consolidated_at, consolidation_failed_at FROM memory_units "
-                    "WHERE bank_id = $1 AND fact_type = 'experience'",
-                    bank_id,
-                )
-            }
-
-        # One should have failed, one should have succeeded
-        failed = [r for r in rows.values() if r["consolidation_failed_at"] is not None]
-        succeeded = [r for r in rows.values() if r["consolidated_at"] is not None]
-        assert len(failed) == 1
-        assert len(succeeded) == 1
-        # They must be different memories
-        assert str(failed[0]["id"]) != str(succeeded[0]["id"])
-
-        await memory_no_llm_verify.delete_bank(bank_id, request_context=request_context)
-
-
-class TestRecoverConsolidation:
-    """Verify the retry_failed_consolidation() method and the /consolidation/recover endpoint."""
-
-    @pytest.mark.asyncio
-    async def test_recover_resets_failed_memories(self, memory_no_llm_verify: MemoryEngine, request_context):
-        """retry_failed_consolidation resets consolidation_failed_at and consolidated_at."""
-        bank_id = f"test-recover-reset-{uuid.uuid4().hex[:8]}"
-        await memory_no_llm_verify.get_bank_profile(bank_id=bank_id, request_context=request_context)
-
-        async with memory_no_llm_verify._pool.acquire() as conn:
-            ids = await _insert_memories(
-                conn,
-                bank_id,
-                [
-                    "Frank is a competitive cyclist.",
-                    "Frank completed the Tour de France route.",
-                ],
-            )
-            # Mark both as failed
-            for mem_id in ids:
-                await conn.execute(
-                    "UPDATE memory_units SET consolidation_failed_at = NOW() WHERE id = $1",
-                    mem_id,
-                )
-
-        result = await memory_no_llm_verify.retry_failed_consolidation(
-            bank_id, request_context=request_context
-        )
-
-        assert result["retried_count"] == 2
-
-        async with memory_no_llm_verify._pool.acquire() as conn:
-            rows = await conn.fetch(
-                "SELECT consolidated_at, consolidation_failed_at FROM memory_units "
-                "WHERE bank_id = $1 AND fact_type = 'experience'",
-                bank_id,
-            )
-        assert all(r["consolidation_failed_at"] is None for r in rows), "consolidation_failed_at must be cleared"
-        assert all(r["consolidated_at"] is None for r in rows), "consolidated_at must also be cleared"
-
-        await memory_no_llm_verify.delete_bank(bank_id, request_context=request_context)
-
-    @pytest.mark.asyncio
-    async def test_recover_returns_zero_when_none_failed(self, memory_no_llm_verify: MemoryEngine, request_context):
-        """retry_failed_consolidation returns 0 when no memories have failed."""
-        bank_id = f"test-recover-zero-{uuid.uuid4().hex[:8]}"
-        await memory_no_llm_verify.get_bank_profile(bank_id=bank_id, request_context=request_context)
-
-        result = await memory_no_llm_verify.retry_failed_consolidation(
-            bank_id, request_context=request_context
-        )
-
-        assert result["retried_count"] == 0
-
-        await memory_no_llm_verify.delete_bank(bank_id, request_context=request_context)
-
-    @pytest.mark.asyncio
-    async def test_recover_then_consolidate_succeeds(self, memory_no_llm_verify: MemoryEngine, request_context):
-        """After recovery, the memory is picked up by the next consolidation run."""
-        bank_id = f"test-recover-consolidate-{uuid.uuid4().hex[:8]}"
-        await memory_no_llm_verify.get_bank_profile(bank_id=bank_id, request_context=request_context)
-
-        async with memory_no_llm_verify._pool.acquire() as conn:
-            (mem_id,) = await _insert_memories(conn, bank_id, ["Grace is an expert rock climber."])
-            await conn.execute(
-                "UPDATE memory_units SET consolidation_failed_at = NOW() WHERE id = $1", mem_id
-            )
-
-        # Recover
-        recover_result = await memory_no_llm_verify.retry_failed_consolidation(
-            bank_id, request_context=request_context
-        )
-        assert recover_result["retried_count"] == 1
-
-        # Now consolidate with a healthy LLM
-        mock_llm = _make_always_success_mock_llm()
-        _inject_mock_llm(memory_no_llm_verify, mock_llm)
-
-        run_result = await run_consolidation_job(
-            memory_engine=memory_no_llm_verify,
-            bank_id=bank_id,
-            request_context=request_context,
-        )
-
-        assert run_result["memories_processed"] == 1
-        assert run_result["memories_failed"] == 0
-
-        async with memory_no_llm_verify._pool.acquire() as conn:
-            row = await conn.fetchrow(
-                "SELECT consolidated_at, consolidation_failed_at FROM memory_units WHERE id = $1",
-                mem_id,
-            )
-        assert row["consolidated_at"] is not None, "Memory should be consolidated after recovery"
-        assert row["consolidation_failed_at"] is None
-
-        await memory_no_llm_verify.delete_bank(bank_id, request_context=request_context)
-
-    @pytest.mark.asyncio
-    async def test_recover_endpoint_via_http(self, memory_no_llm_verify: MemoryEngine, request_context):
-        """The POST /consolidation/recover endpoint returns the correct retried_count."""
-        import httpx
-
-        from hindsight_api.api.http import create_app
-
-        bank_id = f"test-recover-http-{uuid.uuid4().hex[:8]}"
-        await memory_no_llm_verify.get_bank_profile(bank_id=bank_id, request_context=request_context)
-
-        async with memory_no_llm_verify._pool.acquire() as conn:
-            ids = await _insert_memories(
-                conn,
-                bank_id,
-                ["Henry is a professional chef.", "Henry trained at Le Cordon Bleu."],
-            )
-            for mem_id in ids:
-                await conn.execute(
-                    "UPDATE memory_units SET consolidation_failed_at = NOW() WHERE id = $1", mem_id
-                )
-
-        app = create_app(memory_no_llm_verify, initialize_memory=False)
-        transport = httpx.ASGITransport(app=app)
-        async with httpx.AsyncClient(transport=transport, base_url="http://test") as client:
-            response = await client.post(f"/v1/default/banks/{bank_id}/consolidation/recover")
-
-        assert response.status_code == 200
-        body = response.json()
-        assert body["retried_count"] == 2
-
-        await memory_no_llm_verify.delete_bank(bank_id, request_context=request_context)

hindsight-api-slim/tests/test_document_tracking.py

@@ -1,366 +0,0 @@
-"""
-Tests for document tracking and upsert functionality.
-"""
-import logging
-from datetime import datetime, timezone
-from unittest.mock import patch
-
-import pytest
-
-from hindsight_api import RequestContext
-from hindsight_api.engine.response_models import TokenUsage
-
-
-@pytest.mark.asyncio
-async def test_document_creation_and_retrieval(memory, request_context):
-    """Test that documents are created and can be retrieved."""
-    bank_id = f"test_doc_{datetime.now(timezone.utc).timestamp()}"
-
-    try:
-        document_id = "meeting-001"
-
-        # Store memory with document tracking
-        await memory.retain_async(
-            bank_id=bank_id,
-            content="Alice works at Google. Bob works at Microsoft.",
-            context="Team meeting",
-            document_id=document_id,
-            request_context=request_context,
-        )
-
-        # Retrieve document
-        doc = await memory.get_document(document_id, bank_id, request_context=request_context)
-
-        assert doc is not None
-        assert doc["id"] == document_id
-        assert doc["bank_id"] == bank_id
-        assert "Alice works at Google" in doc["original_text"]
-        assert doc["memory_unit_count"] > 0
-
-    finally:
-        await memory.delete_bank(bank_id, request_context=request_context)
-
-
-@pytest.mark.asyncio
-async def test_document_upsert(memory, request_context):
-    """Test that providing the same document_id automatically upserts (deletes old units and creates new ones)."""
-    bank_id = f"test_upsert_{datetime.now(timezone.utc).timestamp()}"
-
-    try:
-        document_id = "meeting-002"
-
-        # First version
-        units_v1 = await memory.retain_async(
-            bank_id=bank_id,
-            content="Alice works at Google.",
-            context="Initial",
-            document_id=document_id,
-            request_context=request_context,
-        )
-
-        # Get document stats
-        doc_v1 = await memory.get_document(document_id, bank_id, request_context=request_context)
-        count_v1 = doc_v1["memory_unit_count"]
-
-        # Update with different content (automatic upsert when same document_id is provided)
-        units_v2 = await memory.retain_async(
-            bank_id=bank_id,
-            content="Alice works at Microsoft. Bob works at Apple.",
-            context="Updated",
-            document_id=document_id,
-            request_context=request_context,
-        )
-
-        # Get updated document stats
-        doc_v2 = await memory.get_document(document_id, bank_id, request_context=request_context)
-        count_v2 = doc_v2["memory_unit_count"]
-
-        # Verify old units were replaced
-        assert "Microsoft" in doc_v2["original_text"]
-        assert doc_v2["updated_at"] > doc_v1["created_at"]
-
-        # Different unit IDs (old ones deleted, new ones created)
-        assert set(units_v1).isdisjoint(set(units_v2))
-
-    finally:
-        await memory.delete_bank(bank_id, request_context=request_context)
-
-
-@pytest.mark.asyncio
-async def test_document_deletion(memory, request_context):
-    """Test that deleting a document cascades to memory units."""
-    bank_id = f"test_delete_{datetime.now(timezone.utc).timestamp()}"
-
-    try:
-        document_id = "meeting-003"
-
-        # Create document
-        await memory.retain_async(
-            bank_id=bank_id,
-            content="Alice works at Google.",
-            context="Test",
-            document_id=document_id,
-            request_context=request_context,
-        )
-
-        # Verify it exists
-        doc = await memory.get_document(document_id, bank_id, request_context=request_context)
-        assert doc is not None
-        assert doc["memory_unit_count"] > 0
-
-        # Delete document
-        result = await memory.delete_document(document_id, bank_id, request_context=request_context)
-        assert result["document_deleted"] == 1
-        assert result["memory_units_deleted"] > 0
-
-        # Verify it's gone
-        doc_after = await memory.get_document(document_id, bank_id, request_context=request_context)
-        assert doc_after is None
-
-    finally:
-        await memory.delete_bank(bank_id, request_context=request_context)
-
-
-@pytest.mark.asyncio
-async def test_memory_without_document(memory, request_context):
-    """Test that memories can still be created without document tracking."""
-    bank_id = f"test_no_doc_{datetime.now(timezone.utc).timestamp()}"
-
-    try:
-        # Create memory without document_id (backward compatibility)
-        units = await memory.retain_async(
-            bank_id=bank_id,
-            content="Alice works at Google.",
-            context="Test",
-            request_context=request_context,
-        )
-
-        assert len(units) > 0
-
-    finally:
-        await memory.delete_bank(bank_id, request_context=request_context)
-
-
-@pytest.mark.asyncio
-async def test_document_persisted_with_zero_facts(memory, request_context):
-    """
-    Test that documents are persisted even when zero facts are extracted.
-
-    This is a regression test for issue #324 where documents with no extractable
-    facts were reported as disappearing from the system.
-    """
-    bank_id = f"test_zero_facts_{datetime.now(timezone.utc).timestamp()}"
-
-    try:
-        document_id = "doc-zero-facts"
-
-        # Retain content that produces zero facts (gibberish/random characters)
-        units = await memory.retain_async(
-            bank_id=bank_id,
-            content="xyzabc123 !!!### @@@ $$$",  # Random characters unlikely to produce facts
-            context="Test zero facts",
-            document_id=document_id,
-            request_context=request_context,
-        )
-
-        # Should return empty unit list (no facts extracted)
-        assert len(units) == 0, "Should extract zero facts from gibberish content"
-
-        # But document should still be persisted and retrievable
-        doc = await memory.get_document(document_id, bank_id, request_context=request_context)
-        assert doc is not None, "Document should be persisted even with zero facts"
-        assert doc["id"] == document_id
-        assert doc["bank_id"] == bank_id
-        assert doc["memory_unit_count"] == 0, "Should have zero memory units"
-        assert len(doc["original_text"]) > 0, "Should have non-zero text length"
-        assert "xyzabc123" in doc["original_text"], "Should contain original content"
-
-        # Document should also appear in list
-        docs_list = await memory.list_documents(
-            bank_id=bank_id,
-            search_query=None,
-            limit=100,
-            offset=0,
-            request_context=request_context,
-        )
-        assert docs_list["total"] == 1, "Document should appear in list"
-        assert any(d["id"] == document_id for d in docs_list["items"]), "Document should be in items"
-
-        listed_doc = next(d for d in docs_list["items"] if d["id"] == document_id)
-        assert listed_doc["memory_unit_count"] == 0, "Listed document should show zero memory units"
-
-    finally:
-        await memory.delete_bank(bank_id, request_context=request_context)
-
-
-@pytest.mark.asyncio
-async def test_document_persisted_with_zero_facts_batch(memory, request_context):
-    """
-    Test that documents are persisted with zero facts in batch retain operations.
-
-    This tests the async batch code path to ensure it also handles zero facts correctly.
-    """
-    bank_id = f"test_zero_facts_batch_{datetime.now(timezone.utc).timestamp()}"
-
-    try:
-        # Mix of content: some produces facts, some produces zero facts
-        contents = [
-            {
-                "content": "Alice works at Google",
-                "document_id": "doc-with-facts",
-            },
-            {
-                "content": "!@# $$$ %%% ^^^ &&& ***",  # Gibberish - zero facts expected
-                "document_id": "doc-zero-facts",
-            },
-        ]
-
-        unit_ids = await memory.retain_batch_async(
-            bank_id=bank_id,
-            contents=contents,
-            request_context=request_context,
-        )
-
-        # First content should produce facts, second should not
-        assert len(unit_ids[0]) > 0, "First content should produce facts"
-        assert len(unit_ids[1]) == 0, "Second content should produce zero facts"
-
-        # Both documents should be persisted
-        doc_with_facts = await memory.get_document("doc-with-facts", bank_id, request_context=request_context)
-        assert doc_with_facts is not None
-        assert doc_with_facts["memory_unit_count"] > 0
-
-        doc_zero_facts = await memory.get_document("doc-zero-facts", bank_id, request_context=request_context)
-        assert doc_zero_facts is not None, "Document with zero facts should be persisted"
-        assert doc_zero_facts["memory_unit_count"] == 0, "Should have zero memory units"
-        assert "!@#" in doc_zero_facts["original_text"]
-
-        # Both should appear in list
-        docs_list = await memory.list_documents(
-            bank_id=bank_id,
-            search_query=None,
-            limit=100,
-            offset=0,
-            request_context=request_context,
-        )
-        assert docs_list["total"] == 2, "Both documents should appear in list"
-
-    finally:
-        await memory.delete_bank(bank_id, request_context=request_context)
-
-
-@pytest.mark.asyncio
-async def test_document_persisted_with_zero_facts_async_submit(memory, request_context):
-    """
-    Test that documents are persisted with zero facts in fire-and-forget async retain.
-
-    This tests the submit_async_retain (background task) code path to ensure it also
-    handles zero facts correctly.
-    """
-    import asyncio
-
-    bank_id = f"test_zero_facts_async_{datetime.now(timezone.utc).timestamp()}"
-
-    try:
-        # Submit async retain with gibberish content
-        result = await memory.submit_async_retain(
-            bank_id=bank_id,
-            contents=[
-                {
-                    "content": "!@# $$$ %%% ^^^ &&& ***",  # Gibberish - zero facts expected
-                    "document_id": "doc-async-zero-facts",
-                }
-            ],
-            request_context=request_context,
-        )
-
-        operation_id = result["operation_id"]
-        assert operation_id is not None, "Should return operation_id"
-
-        # Wait for background task to complete
-        max_wait = 60  # 60 seconds max
-        wait_interval = 0.5
-        elapsed = 0
-
-        while elapsed < max_wait:
-            await asyncio.sleep(wait_interval)
-            elapsed += wait_interval
-
-            # Check if document exists
-            doc = await memory.get_document(
-                "doc-async-zero-facts", bank_id, request_context=request_context
-            )
-            if doc is not None:
-                break
-
-        # Document should be persisted even with zero facts
-        assert doc is not None, "Document should be persisted after async task completes"
-        assert doc["id"] == "doc-async-zero-facts"
-        assert doc["memory_unit_count"] == 0, "Should have zero memory units"
-        assert "!@#" in doc["original_text"]
-
-        # Document should appear in list
-        docs_list = await memory.list_documents(
-            bank_id=bank_id,
-            search_query=None,
-            limit=100,
-            offset=0,
-            request_context=request_context,
-        )
-        assert docs_list["total"] == 1, "Document should appear in list"
-        assert any(d["id"] == "doc-async-zero-facts" for d in docs_list["items"])
-
-        listed_doc = next(d for d in docs_list["items"] if d["id"] == "doc-async-zero-facts")
-        assert listed_doc["memory_unit_count"] == 0, "Listed document should show zero memory units"
-
-    finally:
-        await memory.delete_bank(bank_id, request_context=request_context)
-
-
-@pytest.mark.asyncio
-async def test_document_stored_without_chunks_when_zero_facts(memory_no_llm_verify, request_context):
-    """
-    Regression test: when 0 facts are extracted from chunked content, the document row
-    must be stored but no chunk rows should be written.
-    """
-    bank_id = f"test_zero_facts_no_chunks_{datetime.now(timezone.utc).timestamp()}"
-    document_id = "doc-zero-facts-chunked"
-
-    # Content large enough to exceed default retain_chunk_size (3000 chars) so chunking is triggered
-    content = "Alice works at Google. " * 200  # ~4600 chars
-
-    async def mock_llm_zero_facts(*args, **kwargs):
-        response = {"facts": []}
-        if kwargs.get("return_usage", False):
-            return response, TokenUsage(input_tokens=10, output_tokens=2)
-        return response
-
-    try:
-        with patch("hindsight_api.engine.llm_wrapper.LLMProvider.call", new=mock_llm_zero_facts):
-            units = await memory_no_llm_verify.retain_async(
-                bank_id=bank_id,
-                content=content,
-                document_id=document_id,
-                request_context=request_context,
-            )
-
-        assert units == [], "Should return no memory units when LLM extracts zero facts"
-
-        # Document row must exist
-        doc = await memory_no_llm_verify.get_document(document_id, bank_id, request_context=request_context)
-        assert doc is not None, "Document row must be stored even when zero facts are extracted"
-        assert doc["id"] == document_id
-        assert doc["memory_unit_count"] == 0
-
-        # No chunk rows should be stored
-        pool = await memory_no_llm_verify._get_pool()
-        async with pool.acquire() as conn:
-            chunk_count = await conn.fetchval(
-                "SELECT COUNT(*) FROM chunks WHERE document_id = $1 AND bank_id = $2",
-                document_id,
-                bank_id,
-            )
-        assert chunk_count == 0, "No chunk rows should be stored when zero facts are extracted"
-
-    finally:
-        await memory_no_llm_verify.delete_bank(bank_id, request_context=request_context)

hindsight-api-slim/tests/test_entity_resolver.py

@@ -1,114 +0,0 @@
-"""
-Tests for EntityResolver edge cases.
-"""
-
-import uuid
-from datetime import datetime, timezone
-
-import asyncpg
-import pytest
-
-from hindsight_api.engine.entity_resolver import EntityResolver
-from hindsight_api.pg0 import resolve_database_url
-
-# ---------------------------------------------------------------------------
-# Unit tests for discard_pending_stats() — no database required
-# ---------------------------------------------------------------------------
-
-
-@pytest.mark.asyncio
-async def test_discard_pending_stats_clears_both_dicts():
-    """discard_pending_stats() must remove entries for the current task from
-    both _pending_stats and _pending_cooccurrences."""
-    resolver = EntityResolver(pool=None)  # type: ignore[arg-type]
-    key = resolver._task_key()
-
-    resolver._pending_stats[key] = [object()]  # type: ignore[list-item]
-    resolver._pending_cooccurrences[key] = [object()]  # type: ignore[list-item]
-
-    resolver.discard_pending_stats()
-
-    assert key not in resolver._pending_stats
-    assert key not in resolver._pending_cooccurrences
-
-
-@pytest.mark.asyncio
-async def test_discard_pending_stats_is_idempotent():
-    """Calling discard_pending_stats() when nothing is pending must not raise."""
-    resolver = EntityResolver(pool=None)  # type: ignore[arg-type]
-    resolver.discard_pending_stats()
-    resolver.discard_pending_stats()  # second call — still safe
-
-
-@pytest.mark.asyncio
-async def test_discard_pending_stats_does_not_affect_other_task_keys():
-    """discard_pending_stats() must only remove the current task's entries,
-    leaving entries keyed under other task IDs untouched."""
-    resolver = EntityResolver(pool=None)  # type: ignore[arg-type]
-    other_key = -1  # A fake key that can never be a real task id
-
-    resolver._pending_stats[other_key] = [object()]  # type: ignore[list-item]
-    resolver._pending_cooccurrences[other_key] = [object()]  # type: ignore[list-item]
-
-    resolver.discard_pending_stats()  # discards current task's key only
-
-    assert other_key in resolver._pending_stats, "other task's stats must be preserved"
-    assert other_key in resolver._pending_cooccurrences, "other task's cooccurrences must be preserved"
-
-
-@pytest.mark.asyncio
-async def test_resolve_entities_batch_handles_unicode_lower_conflicts(pg0_db_url):
-    """
-    Existing entities with PostgreSQL/Python lowercase mismatches should resolve
-    to the conflicted row instead of leaving a missing entity_id.
-    """
-    resolved_url = await resolve_database_url(pg0_db_url)
-    pool = await asyncpg.create_pool(resolved_url, min_size=1, max_size=2, command_timeout=30)
-    bank_id = f"test-entity-resolver-{uuid.uuid4().hex[:8]}"
-    event_date = datetime(2024, 1, 15, tzinfo=timezone.utc)
-    resolver = EntityResolver(pool=pool, entity_lookup="full")
-
-    try:
-        async with pool.acquire() as conn:
-            existing_entity_id = await conn.fetchval(
-                """
-                INSERT INTO entities (bank_id, canonical_name, first_seen, last_seen, mention_count)
-                VALUES ($1, $2, $3, $3, 1)
-                RETURNING id
-                """,
-                bank_id,
-                "İstanbul",
-                event_date,
-            )
-
-            resolved_ids = await resolver.resolve_entities_batch(
-                bank_id=bank_id,
-                entities_data=[
-                    {
-                        "text": "istanbul",
-                        "nearby_entities": [],
-                        "event_date": event_date,
-                    }
-                ],
-                context="unicode case mismatch",
-                unit_event_date=event_date,
-                conn=conn,
-            )
-
-            entity_rows = await conn.fetch(
-                """
-                SELECT id, canonical_name
-                FROM entities
-                WHERE bank_id = $1
-                ORDER BY canonical_name
-                """,
-                bank_id,
-            )
-
-        assert resolved_ids == [existing_entity_id]
-        assert len(entity_rows) == 1
-        assert entity_rows[0]["id"] == existing_entity_id
-        assert entity_rows[0]["canonical_name"] == "İstanbul"
-    finally:
-        await pool.execute("DELETE FROM entities WHERE bank_id = $1", bank_id)
-        await pool.close()

hindsight-api-slim/tests/test_entity_resolver_pg_trgm.py

@@ -1,163 +0,0 @@
-"""
-Unit tests for EntityResolver pg_trgm auto-detection (PR #626/#649).
-
-These tests verify:
-1. When entity_lookup="trigram" and pg_trgm IS available, the trigram path is used.
-2. When entity_lookup="trigram" and pg_trgm is NOT available, the resolver falls back
-   to entity_lookup="full" and uses the full-scan path.
-3. The pg_trgm check is only performed once (_pg_trgm_checked flag prevents re-checking).
-4. When entity_lookup="full" from the start, the trgm check is never performed.
-"""
-
-from unittest.mock import AsyncMock, MagicMock, patch
-
-import pytest
-
-from hindsight_api.engine.entity_resolver import EntityResolver
-
-
-def _make_conn(pg_trgm_available: bool) -> MagicMock:
-    """Create a minimal mock asyncpg connection for the pg_trgm availability check."""
-    conn = MagicMock()
-    conn.fetchval = AsyncMock(return_value=pg_trgm_available)
-    conn.fetch = AsyncMock(return_value=[])
-    conn.executemany = AsyncMock()
-    conn.fetchrow = AsyncMock(return_value=None)
-    return conn
-
-
-def _make_resolver(entity_lookup: str = "trigram") -> EntityResolver:
-    """Return an EntityResolver with a None pool (not needed for unit tests)."""
-    return EntityResolver(pool=None, entity_lookup=entity_lookup)  # type: ignore[arg-type]
-
-
-class TestPgTrgmAutoDetection:
-    """Unit tests for pg_trgm detection logic inside _resolve_entities_batch_impl."""
-
-    @pytest.mark.asyncio
-    async def test_falls_back_to_full_when_pg_trgm_unavailable(self):
-        """When pg_trgm is absent the resolver switches to 'full' and calls the full-scan path."""
-        resolver = _make_resolver(entity_lookup="trigram")
-        conn = _make_conn(pg_trgm_available=False)
-
-        with (
-            patch.object(resolver, "_resolve_entities_batch_full", new=AsyncMock(return_value=[])) as mock_full,
-            patch.object(resolver, "_resolve_entities_batch_trigram", new=AsyncMock(return_value=[])) as mock_trgm,
-        ):
-            await resolver._resolve_entities_batch_impl(
-                conn=conn,
-                bank_id="test-bank",
-                entities_data=[],
-                context="",
-                unit_event_date=None,
-            )
-
-        # Trigram path must NOT be called
-        mock_trgm.assert_not_called()
-        # Full-scan path must be called as the fallback
-        mock_full.assert_called_once()
-        # Strategy is permanently downgraded
-        assert resolver.entity_lookup == "full"
-        assert resolver._pg_trgm_checked is True
-
-    @pytest.mark.asyncio
-    async def test_uses_trigram_when_pg_trgm_available(self):
-        """When pg_trgm is present the trigram path is used."""
-        resolver = _make_resolver(entity_lookup="trigram")
-        conn = _make_conn(pg_trgm_available=True)
-
-        with (
-            patch.object(resolver, "_resolve_entities_batch_full", new=AsyncMock(return_value=[])) as mock_full,
-            patch.object(resolver, "_resolve_entities_batch_trigram", new=AsyncMock(return_value=[])) as mock_trgm,
-        ):
-            await resolver._resolve_entities_batch_impl(
-                conn=conn,
-                bank_id="test-bank",
-                entities_data=[],
-                context="",
-                unit_event_date=None,
-            )
-
-        mock_trgm.assert_called_once()
-        mock_full.assert_not_called()
-        assert resolver.entity_lookup == "trigram"
-        assert resolver._pg_trgm_checked is True
-
-    @pytest.mark.asyncio
-    async def test_pg_trgm_check_performed_only_once(self):
-        """The fetchval check is only issued on the first call; subsequent calls skip it."""
-        resolver = _make_resolver(entity_lookup="trigram")
-        conn = _make_conn(pg_trgm_available=True)
-
-        with patch.object(resolver, "_resolve_entities_batch_trigram", new=AsyncMock(return_value=[])):
-            # First call — check is issued
-            await resolver._resolve_entities_batch_impl(
-                conn=conn,
-                bank_id="test-bank",
-                entities_data=[],
-                context="",
-                unit_event_date=None,
-            )
-            # Second call — check must NOT be issued again
-            await resolver._resolve_entities_batch_impl(
-                conn=conn,
-                bank_id="test-bank",
-                entities_data=[],
-                context="",
-                unit_event_date=None,
-            )
-
-        # fetchval (the pg_trgm availability query) should be called exactly once
-        assert conn.fetchval.call_count == 1
-
-    @pytest.mark.asyncio
-    async def test_full_strategy_skips_pg_trgm_check(self):
-        """When entity_lookup='full' from the start, no pg_trgm check is ever issued."""
-        resolver = _make_resolver(entity_lookup="full")
-        conn = _make_conn(pg_trgm_available=False)
-
-        with patch.object(resolver, "_resolve_entities_batch_full", new=AsyncMock(return_value=[])):
-            await resolver._resolve_entities_batch_impl(
-                conn=conn,
-                bank_id="test-bank",
-                entities_data=[],
-                context="",
-                unit_event_date=None,
-            )
-
-        # fetchval should never be called when entity_lookup is already "full"
-        conn.fetchval.assert_not_called()
-
-    @pytest.mark.asyncio
-    async def test_fallback_is_sticky_across_calls(self):
-        """After falling back to 'full', subsequent calls also use the full path."""
-        resolver = _make_resolver(entity_lookup="trigram")
-        conn = _make_conn(pg_trgm_available=False)
-
-        with (
-            patch.object(resolver, "_resolve_entities_batch_full", new=AsyncMock(return_value=[])) as mock_full,
-            patch.object(resolver, "_resolve_entities_batch_trigram", new=AsyncMock(return_value=[])) as mock_trgm,
-        ):
-            # First call triggers the fallback
-            await resolver._resolve_entities_batch_impl(
-                conn=conn,
-                bank_id="b",
-                entities_data=[],
-                context="",
-                unit_event_date=None,
-            )
-            # Second call — _pg_trgm_checked is True so no re-check; entity_lookup=="full"
-            await resolver._resolve_entities_batch_impl(
-                conn=conn,
-                bank_id="b",
-                entities_data=[],
-                context="",
-                unit_event_date=None,
-            )
-
-        # Trigram path is never called
-        mock_trgm.assert_not_called()
-        # Full-scan path is called both times
-        assert mock_full.call_count == 2
-        # pg_trgm check was issued exactly once
-        assert conn.fetchval.call_count == 1

hindsight-api-slim/tests/test_fact_extraction_metadata.py

@@ -1,61 +0,0 @@
-"""
-Unit tests for metadata inclusion in fact extraction LLM prompt.
-"""
-from datetime import datetime
-
-from hindsight_api.engine.retain.fact_extraction import _build_user_message
-
-
-def test_build_user_message_includes_metadata():
-    """Metadata key-value pairs should appear in the user message."""
-    event_date = datetime(2024, 6, 15, 12, 0, 0)
-    metadata = {"title": "Q2 Planning Doc", "source": "confluence", "author": "Alice"}
-
-    msg = _build_user_message(
-        chunk="Some content.",
-        chunk_index=0,
-        total_chunks=1,
-        event_date=event_date,
-        context="planning meeting",
-        metadata=metadata,
-    )
-
-    assert "title" in msg
-    assert "Q2 Planning Doc" in msg
-    assert "source" in msg
-    assert "confluence" in msg
-    assert "author" in msg
-    assert "Alice" in msg
-
-
-def test_build_user_message_no_metadata():
-    """When metadata is empty, the message should still be valid and not include a metadata section."""
-    event_date = datetime(2024, 6, 15, 12, 0, 0)
-
-    msg = _build_user_message(
-        chunk="Some content.",
-        chunk_index=0,
-        total_chunks=1,
-        event_date=event_date,
-        context="planning meeting",
-        metadata={},
-    )
-
-    assert "Some content." in msg
-    assert "Metadata:" not in msg
-
-
-def test_build_user_message_without_metadata_arg():
-    """Calling without metadata (default) should behave the same as empty metadata."""
-    event_date = datetime(2024, 6, 15, 12, 0, 0)
-
-    msg = _build_user_message(
-        chunk="Some content.",
-        chunk_index=0,
-        total_chunks=1,
-        event_date=event_date,
-        context="none",
-    )
-
-    assert "Some content." in msg
-    assert "Metadata:" not in msg

hindsight-api-slim/tests/test_fact_extraction_retry.py

@@ -1,141 +0,0 @@
-"""
-Unit tests for fact extraction retry logic.
-
-Tests the fix for the TypeError when LLM returns invalid JSON across all retries.
-Previously, `raise last_error` would raise None (TypeError) because last_error was
-only set in the BadRequestError handler, not when the LLM returned non-dict JSON.
-"""
-
-from datetime import datetime, timezone
-from unittest.mock import AsyncMock, MagicMock, patch
-
-import pytest
-
-
-def _make_config(llm_max_retries: int = 3, retain_llm_max_retries: int | None = None):
-    """Build a minimal HindsightConfig for fact extraction tests."""
-    from hindsight_api.config import HindsightConfig
-
-    cfg = MagicMock(spec=HindsightConfig)
-    cfg.retain_llm_max_retries = retain_llm_max_retries
-    cfg.llm_max_retries = llm_max_retries
-    cfg.retain_llm_initial_backoff = None
-    cfg.llm_initial_backoff = 0.0
-    cfg.retain_llm_max_backoff = None
-    cfg.llm_max_backoff = 0.0
-    cfg.retain_max_completion_tokens = 8192
-    cfg.retain_extraction_mode = "concise"
-    cfg.retain_extract_causal_links = False
-    cfg.retain_mission = None
-    return cfg
-
-
-def _make_llm_config(mock_response):
-    """Build a mock LLMProvider that returns the given response."""
-    from hindsight_api.engine.llm_wrapper import LLMProvider
-
-    llm = MagicMock(spec=LLMProvider)
-    llm.provider = "mock"
-    token_usage = MagicMock()
-    token_usage.__add__ = lambda self, other: self
-    llm.call = AsyncMock(return_value=(mock_response, token_usage))
-    return llm
-
-
-@pytest.mark.asyncio
-async def test_non_dict_json_all_retries_returns_empty():
-    """
-    When LLM returns non-dict JSON on every attempt, extraction should return []
-    without raising TypeError ('exceptions must derive from BaseException').
-
-    This was the bug: the loop ran range(2) times (hardcoded), but comparisons
-    used config.llm_max_retries (default 10). On the last loop iteration (attempt=1),
-    `attempt < 10 - 1` was True, so the code called `continue`, the loop
-    exhausted, and `raise last_error` raised None → TypeError.
-    """
-    from hindsight_api.engine.retain.fact_extraction import _extract_facts_from_chunk
-
-    # llm_max_retries=3 ensures the bug triggers with the old code (3 != 2 hardcoded)
-    config = _make_config(llm_max_retries=3, retain_llm_max_retries=None)
-
-    # Mock: always returns a list (non-dict), which is invalid
-    llm_config = _make_llm_config(mock_response=[{"invalid": "response"}])
-
-    with patch(
-        "hindsight_api.engine.retain.fact_extraction._build_extraction_prompt_and_schema",
-        return_value=("system prompt", MagicMock()),
-    ):
-        facts, usage = await _extract_facts_from_chunk(
-            chunk="Alice visited Paris in 2023.",
-            chunk_index=0,
-            total_chunks=1,
-            event_date=datetime(2023, 1, 1, tzinfo=timezone.utc),
-            context="travel notes",
-            llm_config=llm_config,
-            config=config,
-            agent_name="test-agent",
-        )
-
-    assert facts == []
-
-
-@pytest.mark.asyncio
-async def test_non_dict_json_with_default_max_retries_returns_empty():
-    """
-    Same scenario with the default llm_max_retries=10 (matching real default config).
-    The old code ran range(2) but checked against 10, always continuing until
-    the loop exhausted, then raised None → TypeError.
-    """
-    from hindsight_api.engine.retain.fact_extraction import _extract_facts_from_chunk
-
-    config = _make_config(llm_max_retries=10, retain_llm_max_retries=None)
-    llm_config = _make_llm_config(mock_response="not a dict at all")
-
-    with patch(
-        "hindsight_api.engine.retain.fact_extraction._build_extraction_prompt_and_schema",
-        return_value=("system prompt", MagicMock()),
-    ):
-        facts, usage = await _extract_facts_from_chunk(
-            chunk="Some text.",
-            chunk_index=0,
-            total_chunks=1,
-            event_date=datetime(2023, 6, 1, tzinfo=timezone.utc),
-            context="",
-            llm_config=llm_config,
-            config=config,
-            agent_name="agent",
-        )
-
-    assert facts == []
-
-
-@pytest.mark.asyncio
-async def test_retain_llm_max_retries_overrides_global():
-    """
-    When retain_llm_max_retries is set, it should be used for the loop range
-    and all comparisons (no shadowing bug).
-    """
-    from hindsight_api.engine.retain.fact_extraction import _extract_facts_from_chunk
-
-    # retain_llm_max_retries=5 should override llm_max_retries=10
-    config = _make_config(llm_max_retries=10, retain_llm_max_retries=5)
-    llm_config = _make_llm_config(mock_response=42)  # non-dict: integer
-
-    with patch(
-        "hindsight_api.engine.retain.fact_extraction._build_extraction_prompt_and_schema",
-        return_value=("system prompt", MagicMock()),
-    ):
-        facts, usage = await _extract_facts_from_chunk(
-            chunk="Bob likes Python.",
-            chunk_index=0,
-            total_chunks=1,
-            event_date=datetime(2024, 1, 1, tzinfo=timezone.utc),
-            context="",
-            llm_config=llm_config,
-            config=config,
-            agent_name="agent",
-        )
-
-    assert facts == []
-    # Verify it retried exactly retain_llm_max_retries times
-    assert llm_config.call.call_count == 5

hindsight-api-slim/tests/test_file_retain.py

@@ -1,839 +0,0 @@
-"""
-End-to-end tests for file retain (upload, convert, retain) functionality.
-"""
-
-import asyncio
-import io
-import json
-from datetime import datetime, timezone
-
-import pytest
-from httpx import ASGITransport, AsyncClient
-
-from hindsight_api.extensions import FileConvertResult, OperationValidatorExtension, ValidationResult
-from hindsight_api.extensions.operation_validator import (
-    RecallContext,
-    RecallResult,
-    ReflectContext,
-    RetainContext,
-    RetainResult,
-)
-
-
-@pytest.fixture
-def sample_pdf_content():
-    """Create a simple PDF-like content for testing."""
-    # This is a minimal PDF that markitdown can parse
-    return b"""%PDF-1.4
-1 0 obj
-<<
-/Type /Catalog
-/Pages 2 0 R
->>
-endobj
-2 0 obj
-<<
-/Type /Pages
-/Kids [3 0 R]
-/Count 1
->>
-endobj
-3 0 obj
-<<
-/Type /Page
-/Parent 2 0 R
-/MediaBox [0 0 612 792]
-/Contents 4 0 R
-/Resources <<
-/Font <<
-/F1 <<
-/Type /Font
-/Subtype /Type1
-/BaseFont /Helvetica
->>
->>
->>
->>
-endobj
-4 0 obj
-<<
-/Length 44
->>
-stream
-BT
-/F1 12 Tf
-100 700 Td
-(Test Document) Tj
-ET
-endstream
-endobj
-xref
-0 5
-0000000000 65535 f
-0000000009 00000 n
-0000000058 00000 n
-0000000115 00000 n
-0000000317 00000 n
-trailer
-<<
-/Size 5
-/Root 1 0 R
->>
-startxref
-410
-%%EOF
-"""
-
-
-@pytest.fixture
-def sample_txt_content():
-    """Create simple text content."""
-    return b"This is a test document.\nIt contains some important information.\nAlice works at Google."
-
-
-@pytest.mark.asyncio
-async def test_file_retain_basic(memory_no_llm_verify, sample_txt_content):
-    """Test basic file upload and conversion."""
-    from hindsight_api.api.http import create_app
-
-    app = create_app(memory_no_llm_verify, initialize_memory=False)
-
-    async with AsyncClient(transport=ASGITransport(app=app), base_url="http://test") as client:
-        # Create a bank first
-        bank_response = await client.put("/v1/default/banks/test-file-bank", json={"name": "Test File Bank"})
-        assert bank_response.status_code in (200, 201)
-
-        # Upload file
-        request_data = {
-            "document_tags": ["test"],
-            "async": True,
-        }
-
-        files = {"files": ("test.txt", sample_txt_content, "text/plain")}
-        data = {"request": json.dumps(request_data)}
-
-        response = await client.post(
-            "/v1/default/banks/test-file-bank/files/retain",
-            files=files,
-            data=data,
-        )
-
-        assert response.status_code == 200
-        result = response.json()
-        assert "operation_ids" in result
-        assert len(result["operation_ids"]) == 1
-
-
-@pytest.mark.asyncio
-async def test_file_retain_with_metadata(memory_no_llm_verify, sample_txt_content):
-    """Test file upload with per-file metadata."""
-    from hindsight_api.api.http import create_app
-
-    app = create_app(memory_no_llm_verify, initialize_memory=False)
-
-    async with AsyncClient(transport=ASGITransport(app=app), base_url="http://test") as client:
-        # Create bank
-        bank_response = await client.put("/v1/default/banks/test-file-meta-bank", json={"name": "Test Meta Bank"})
-        assert bank_response.status_code in (200, 201)
-
-        # Upload file with metadata
-        request_data = {
-            "document_tags": ["work", "reports"],
-            "async": True,
-            "files_metadata": [
-                {
-                    "document_id": "test_doc_123",
-                    "context": "quarterly report",
-                    "metadata": {"author": "Alice", "year": "2024"},
-                    "tags": ["Q1"],
-                }
-            ],
-        }
-
-        files = {"files": ("report.txt", sample_txt_content, "text/plain")}
-        data = {"request": json.dumps(request_data)}
-
-        response = await client.post(
-            "/v1/default/banks/test-file-meta-bank/files/retain",
-            files=files,
-            data=data,
-        )
-
-        assert response.status_code == 200
-        result = response.json()
-        assert "operation_ids" in result
-        assert len(result["operation_ids"]) == 1
-
-
-@pytest.mark.asyncio
-async def test_file_retain_multiple_files(memory_no_llm_verify, sample_txt_content):
-    """Test uploading multiple files at once."""
-    from hindsight_api.api.http import create_app
-
-    app = create_app(memory_no_llm_verify, initialize_memory=False)
-
-    async with AsyncClient(transport=ASGITransport(app=app), base_url="http://test") as client:
-        # Create bank
-        bank_response = await client.put("/v1/default/banks/test-multi-file-bank", json={"name": "Test Multi Bank"})
-        assert bank_response.status_code in (200, 201)
-
-        # Upload multiple files
-        request_data = {
-            "async": True,
-            "files_metadata": [
-                {"document_id": "doc1", "tags": ["file1"]},
-                {"document_id": "doc2", "tags": ["file2"]},
-            ],
-        }
-
-        content1 = b"First document content"
-        content2 = b"Second document content"
-
-        files = [
-            ("files", ("file1.txt", content1, "text/plain")),
-            ("files", ("file2.txt", content2, "text/plain")),
-        ]
-        data = {"request": json.dumps(request_data)}
-
-        response = await client.post(
-            "/v1/default/banks/test-multi-file-bank/files/retain",
-            files=files,
-            data=data,
-        )
-
-        assert response.status_code == 200
-        result = response.json()
-        assert "operation_ids" in result
-        assert len(result["operation_ids"]) == 2
-
-
-@pytest.mark.asyncio
-async def test_file_retain_validation_errors(memory_no_llm_verify):
-    """Test validation errors."""
-    from hindsight_api.api.http import create_app
-
-    app = create_app(memory_no_llm_verify, initialize_memory=False)
-
-    async with AsyncClient(transport=ASGITransport(app=app), base_url="http://test") as client:
-        # Create bank
-        bank_response = await client.put("/v1/default/banks/test-validation-bank", json={"name": "Test Validation Bank"})
-        assert bank_response.status_code in (200, 201)
-
-        # Test: metadata count mismatch
-        request_data = {
-            "async": True,
-            "files_metadata": [
-                {"document_id": "doc1"},
-                {"document_id": "doc2"},  # 2 metadata entries
-            ],
-        }
-
-        files = {"files": ("file1.txt", b"content", "text/plain")}  # But only 1 file
-        data = {"request": json.dumps(request_data)}
-
-        response = await client.post(
-            "/v1/default/banks/test-validation-bank/files/retain",
-            files=files,
-            data=data,
-        )
-
-        assert response.status_code == 400
-        assert "files_metadata count" in response.json()["detail"]
-
-
-@pytest.mark.asyncio
-async def test_file_retain_no_files(memory_no_llm_verify):
-    """Test error when no files provided."""
-    from hindsight_api.api.http import create_app
-
-    app = create_app(memory_no_llm_verify, initialize_memory=False)
-
-    async with AsyncClient(transport=ASGITransport(app=app), base_url="http://test") as client:
-        # Create bank
-        bank_response = await client.put("/v1/default/banks/test-no-files-bank", json={"name": "Test No Files Bank"})
-        assert bank_response.status_code in (200, 201)
-
-        request_data = {
-            "async": True,
-        }
-
-        # No files provided
-        data = {"request": json.dumps(request_data)}
-
-        response = await client.post(
-            "/v1/default/banks/test-no-files-bank/files/retain",
-            data=data,
-        )
-
-        # FastAPI will return 422 for missing required field
-        assert response.status_code == 422
-
-
-@pytest.mark.asyncio
-async def test_file_retain_sync_not_supported(memory_no_llm_verify, sample_txt_content):
-    """Test that file retain is always async (sync is not supported)."""
-    from hindsight_api.api.http import create_app
-
-    app = create_app(memory_no_llm_verify, initialize_memory=False)
-
-    async with AsyncClient(transport=ASGITransport(app=app), base_url="http://test") as client:
-        # Create bank
-        bank_response = await client.put("/v1/default/banks/test-sync-bank", json={"name": "Test Sync Bank"})
-        assert bank_response.status_code in (200, 201)
-
-        # File retain is always async - just verify it succeeds and returns operation_ids
-        files = {"files": ("test.txt", sample_txt_content, "text/plain")}
-        data = {"request": json.dumps({})}
-
-        response = await client.post(
-            "/v1/default/banks/test-sync-bank/files/retain",
-            files=files,
-            data=data,
-        )
-
-        assert response.status_code == 200
-        result = response.json()
-        assert "operation_ids" in result
-
-
-@pytest.mark.asyncio
-async def test_file_storage_postgresql(memory_no_llm_verify, sample_txt_content):
-    """Test file storage in PostgreSQL."""
-    # Test that files are stored and retrieved correctly
-    storage = memory_no_llm_verify._file_storage
-
-    # Store a file
-    key = "test/file1.txt"
-    stored_key = await storage.store(
-        file_data=sample_txt_content,
-        key=key,
-        metadata={"content_type": "text/plain"},
-    )
-
-    assert stored_key == key
-
-    # Retrieve the file
-    retrieved = await storage.retrieve(key)
-    assert retrieved == sample_txt_content
-
-    # Check if file exists
-    exists = await storage.exists(key)
-    assert exists is True
-
-    # Delete the file
-    await storage.delete(key)
-
-    # Check file no longer exists
-    exists_after = await storage.exists(key)
-    assert exists_after is False
-
-
-@pytest.mark.asyncio
-async def test_markitdown_converter():
-    """Test markitdown parser."""
-    from hindsight_api.engine.parsers import MarkitdownParser
-
-    parser = MarkitdownParser()
-
-    # Test simple text file
-    text_content = b"This is a test document.\nWith multiple lines."
-    result = await parser.convert(text_content, "test.txt")
-
-    assert isinstance(result, str)
-    assert len(result) > 0
-    assert "test document" in result.lower() or "multiple lines" in result.lower()
-
-
-@pytest.mark.asyncio
-async def test_converter_registry():
-    """Test file parser registry."""
-    from hindsight_api.engine.parsers import FileParserRegistry, MarkitdownParser
-
-    registry = FileParserRegistry()
-    parser = MarkitdownParser()
-    registry.register(parser)
-
-    # Test get by name
-    retrieved = registry.get_parser("markitdown", "test.txt")
-    assert retrieved is parser
-
-    # Test auto-detection
-    auto = registry.get_parser(None, "test.pdf")
-    assert auto is parser
-
-    # Test unsupported format
-    with pytest.raises(ValueError, match="No parser found"):
-        registry.get_parser(None, "test.xyz")
-
-
-@pytest.mark.asyncio
-async def test_file_conversion_creates_separate_retain_operation(memory_no_llm_verify, sample_txt_content):
-    """Test that file conversion and retain are two separate async operations.
-
-    The file_convert_retain task should:
-    1. Convert the file to markdown
-    2. In a single transaction: create a separate 'retain' operation AND mark itself as 'completed'
-    3. Free the worker slot immediately after conversion
-
-    The retain then runs as its own task. This prevents deadlocks where file conversion
-    tasks hold worker slots while waiting for inline retain to finish.
-    """
-    from hindsight_api.models import RequestContext
-
-    bank_id = "test_file_two_phase_bank"
-
-    context = RequestContext(internal=True)
-    await memory_no_llm_verify.get_bank_profile(bank_id, request_context=context)
-
-    class MockFile:
-        def __init__(self, content, filename, content_type):
-            self.content = content
-            self.filename = filename
-            self.content_type = content_type
-
-        async def read(self):
-            return self.content
-
-    mock_file = MockFile(sample_txt_content, "test.txt", "text/plain")
-
-    file_items = [
-        {
-            "file": mock_file,
-            "document_id": "test_doc_two_phase",
-            "context": "test context",
-            "metadata": {"source": "test"},
-            "tags": ["test_tag"],
-            "timestamp": None,
-            "parser": ["markitdown"],
-        }
-    ]
-
-    result = await memory_no_llm_verify.submit_async_file_retain(
-        bank_id=bank_id,
-        file_items=file_items,
-        document_tags=["two_phase_test"],
-        request_context=context,
-    )
-
-    assert "operation_ids" in result
-    assert len(result["operation_ids"]) == 1
-    convert_operation_id = result["operation_ids"][0]
-
-    import asyncio
-
-    await asyncio.sleep(0.1)
-
-    pool = await memory_no_llm_verify._get_pool()
-    from hindsight_api.engine.memory_engine import get_current_schema
-
-    schema = get_current_schema()
-
-    async with pool.acquire() as conn:
-        # 1. The file_convert_retain operation must be completed
-        convert_op = await conn.fetchrow(
-            f"SELECT status, operation_type FROM {schema}.async_operations WHERE operation_id = $1",
-            convert_operation_id,
-        )
-        assert convert_op is not None
-        assert convert_op["operation_type"] == "file_convert_retain"
-        assert convert_op["status"] == "completed", (
-            f"file_convert_retain should be 'completed' after conversion, got '{convert_op['status']}'"
-        )
-
-        # 2. A separate retain operation must have been created
-        retain_op = await conn.fetchrow(
-            f"""
-            SELECT status, operation_type
-            FROM {schema}.async_operations
-            WHERE bank_id = $1 AND operation_type = 'retain' AND operation_id != $2
-            """,
-            bank_id,
-            convert_operation_id,
-        )
-        assert retain_op is not None, "A separate 'retain' operation should have been created by file conversion"
-        # With SyncTaskBackend the retain runs immediately, so it should be completed
-        assert retain_op["status"] == "completed"
-
-        # 3. The document should exist with file metadata and retained content
-        doc = await conn.fetchrow(
-            f"""
-            SELECT id, original_text, file_original_name, file_content_type
-            FROM {schema}.documents
-            WHERE id = $1 AND bank_id = $2
-            """,
-            "test_doc_two_phase",
-            bank_id,
-        )
-
-    assert doc is not None
-    assert doc["file_original_name"] == "test.txt"
-    assert doc["file_content_type"] == "text/plain"
-    assert doc["original_text"] is not None
-    assert len(doc["original_text"]) > 0
-
-
-@pytest.mark.asyncio
-async def test_async_file_retain_serializes_datetime_timestamp(memory_no_llm_verify, sample_txt_content):
-    """Async file retain should accept Python datetimes in task payloads."""
-    from hindsight_api.engine.parsers.base import FileParser
-    from hindsight_api.models import RequestContext
-
-    bank_id = f"test_file_timestamp_bank_{datetime.now(timezone.utc).timestamp()}"
-    timestamp = datetime(2024, 1, 15, 10, 30, tzinfo=timezone.utc)
-
-    context = RequestContext(internal=True)
-    await memory_no_llm_verify.get_bank_profile(bank_id, request_context=context)
-
-    class MockFile:
-        def __init__(self, content, filename, content_type):
-            self.content = content
-            self.filename = filename
-            self.content_type = content_type
-
-        async def read(self):
-            return self.content
-
-    class TimestampParser(FileParser):
-        async def convert(self, file_data: bytes, filename: str) -> str:
-            return file_data.decode("utf-8")
-
-        def supports(self, filename: str, content_type: str | None = None) -> bool:
-            return filename.endswith(".txt")
-
-        def name(self) -> str:
-            return "timestamp_parser"
-
-    memory_no_llm_verify._parser_registry.register(TimestampParser())
-
-    mock_file = MockFile(sample_txt_content, "timestamped.txt", "text/plain")
-
-    result = await memory_no_llm_verify.submit_async_file_retain(
-        bank_id=bank_id,
-        file_items=[
-            {
-                "file": mock_file,
-                "document_id": "timestamped_doc",
-                "context": "timestamp test",
-                "metadata": {},
-                "tags": [],
-                "timestamp": timestamp,
-                "parser": ["timestamp_parser"],
-            }
-        ],
-        document_tags=None,
-        request_context=context,
-    )
-
-    operation_id = result["operation_ids"][0]
-    pool = await memory_no_llm_verify._get_pool()
-    from hindsight_api.engine.memory_engine import get_current_schema
-
-    async with pool.acquire() as conn:
-        row = await conn.fetchrow(
-            f"""
-            SELECT status, task_payload->>'timestamp' AS timestamp
-            FROM {get_current_schema()}.async_operations
-            WHERE operation_id = $1
-            """,
-            operation_id,
-        )
-
-    assert row is not None
-    assert row["status"] == "completed"
-    assert row["timestamp"] == "2024-01-15T10:30:00+00:00"
-
-
-@pytest.mark.asyncio
-async def test_file_conversion_failure_sets_status_to_failed(memory_no_llm_verify, sample_txt_content):
-    """Test that when file conversion fails, the operation status is set to 'failed' not 'completed'."""
-    from hindsight_api.engine.parsers.base import FileParser
-    from hindsight_api.models import RequestContext
-
-    bank_id = "test_file_failure_bank"
-
-    # Create a mock parser that always fails
-    class FailingParser(FileParser):
-        """Mock parser that raises an error."""
-
-        async def convert(self, file_data: bytes, filename: str) -> str:
-            # Simulate conversion failure
-            raise RuntimeError(f"Failed to convert '{filename}': Mock conversion error")
-
-        def supports(self, filename: str, content_type: str | None = None) -> bool:
-            return filename.endswith(".fail")
-
-        def name(self) -> str:
-            return "failing_converter"
-
-    # Register the failing parser
-    failing_converter = FailingParser()
-    memory_no_llm_verify._parser_registry.register(failing_converter)
-
-    # Create bank
-    context = RequestContext(internal=True)
-    await memory_no_llm_verify.get_bank_profile(bank_id, request_context=context)
-
-    # Create mock file
-    class MockFile:
-        def __init__(self, content, filename, content_type):
-            self.content = content
-            self.filename = filename
-            self.content_type = content_type
-
-        async def read(self):
-            return self.content
-
-    mock_file = MockFile(sample_txt_content, "test.fail", "application/octet-stream")
-
-    file_items = [
-        {
-            "file": mock_file,
-            "document_id": "test_doc_fail",
-            "context": None,
-            "metadata": {},
-            "tags": [],
-            "timestamp": None,
-            "parser": ["failing_converter"],
-        }
-    ]
-
-    # Submit async file retain with failing parser
-    result = await memory_no_llm_verify.submit_async_file_retain(
-        bank_id=bank_id,
-        file_items=file_items,
-        document_tags=None,
-        request_context=context,
-    )
-
-    assert "operation_ids" in result
-    assert len(result["operation_ids"]) == 1
-    operation_id = result["operation_ids"][0]
-
-    # Wait for async processing (with SyncTaskBackend, this is immediate)
-    import asyncio
-
-    await asyncio.sleep(0.2)
-
-    # Check operation status - should be 'failed' not 'completed'
-    pool = await memory_no_llm_verify._get_pool()
-    from hindsight_api.engine.memory_engine import get_current_schema
-
-    async with pool.acquire() as conn:
-        operation = await conn.fetchrow(
-            f"""
-            SELECT status, error_message
-            FROM {get_current_schema()}.async_operations
-            WHERE operation_id = $1
-            """,
-            operation_id,
-        )
-
-    assert operation is not None, f"Operation {operation_id} not found"
-    assert operation["status"] == "failed", f"Expected status 'failed' but got '{operation['status']}'"
-    assert operation["error_message"] is not None
-    assert "Mock conversion error" in operation["error_message"]
-    assert "test.fail" in operation["error_message"]
-
-
-class FileConvertTrackingValidator(OperationValidatorExtension):
-    """Validator that tracks on_file_convert_complete hook calls."""
-
-    def __init__(self):
-        super().__init__({})
-        self.convert_calls: list[FileConvertResult] = []
-
-    async def validate_retain(self, ctx: RetainContext) -> ValidationResult:
-        return ValidationResult.accept()
-
-    async def validate_recall(self, ctx: RecallContext) -> ValidationResult:
-        return ValidationResult.accept()
-
-    async def validate_reflect(self, ctx: ReflectContext) -> ValidationResult:
-        return ValidationResult.accept()
-
-    async def on_retain_complete(self, result: RetainResult) -> None:
-        pass
-
-    async def on_recall_complete(self, result: RecallResult) -> None:
-        pass
-
-    async def on_file_convert_complete(self, result: FileConvertResult) -> None:
-        self.convert_calls.append(result)
-
-
-@pytest.mark.asyncio
-async def test_on_file_convert_complete_hook_called(memory_no_llm_verify, sample_txt_content):
-    """Test that on_file_convert_complete hook is called after file conversion with correct parameters."""
-    from hindsight_api.models import RequestContext
-
-    bank_id = "test_file_convert_hook_bank"
-    validator = FileConvertTrackingValidator()
-    memory_no_llm_verify._operation_validator = validator
-
-    context = RequestContext(internal=True, api_key_id="test-key-id", tenant_id="test-tenant")
-    await memory_no_llm_verify.get_bank_profile(bank_id, request_context=context)
-
-    class MockFile:
-        def __init__(self, content, filename, content_type):
-            self.content = content
-            self.filename = filename
-            self.content_type = content_type
-
-        async def read(self):
-            return self.content
-
-    mock_file = MockFile(sample_txt_content, "report.txt", "text/plain")
-
-    file_items = [
-        {
-            "file": mock_file,
-            "document_id": "hook_test_doc",
-            "context": "test context",
-            "metadata": {},
-            "tags": [],
-            "timestamp": None,
-            "parser": ["markitdown"],
-        }
-    ]
-
-    await memory_no_llm_verify.submit_async_file_retain(
-        bank_id=bank_id,
-        file_items=file_items,
-        document_tags=None,
-        request_context=context,
-    )
-
-    await asyncio.sleep(0.1)
-
-    assert len(validator.convert_calls) == 1
-    result = validator.convert_calls[0]
-    assert result.bank_id == bank_id
-    assert result.filename == "report.txt"
-    assert result.parser_name == "markitdown"
-    assert result.output_chars > 0
-    assert result.output_text is not None
-    assert len(result.output_text) == result.output_chars
-    assert result.success is True
-    assert result.error is None
-    assert result.request_context is not None
-    assert result.request_context.api_key_id == "test-key-id"
-    assert result.request_context.tenant_id == "test-tenant"
-
-
-@pytest.mark.asyncio
-async def test_on_file_convert_complete_hook_called_for_each_file(memory_no_llm_verify, sample_txt_content):
-    """Test that on_file_convert_complete is called once per file when uploading multiple files."""
-    from hindsight_api.models import RequestContext
-
-    bank_id = "test_file_convert_hook_multi_bank"
-    validator = FileConvertTrackingValidator()
-    memory_no_llm_verify._operation_validator = validator
-
-    context = RequestContext(internal=True)
-    await memory_no_llm_verify.get_bank_profile(bank_id, request_context=context)
-
-    class MockFile:
-        def __init__(self, content, filename, content_type):
-            self.content = content
-            self.filename = filename
-            self.content_type = content_type
-
-        async def read(self):
-            return self.content
-
-    file_items = [
-        {
-            "file": MockFile(b"First document content", "first.txt", "text/plain"),
-            "document_id": "doc_1",
-            "context": None,
-            "metadata": {},
-            "tags": [],
-            "timestamp": None,
-            "parser": ["markitdown"],
-        },
-        {
-            "file": MockFile(b"Second document content", "second.txt", "text/plain"),
-            "document_id": "doc_2",
-            "context": None,
-            "metadata": {},
-            "tags": [],
-            "timestamp": None,
-            "parser": ["markitdown"],
-        },
-    ]
-
-    await memory_no_llm_verify.submit_async_file_retain(
-        bank_id=bank_id,
-        file_items=file_items,
-        document_tags=None,
-        request_context=context,
-    )
-
-    await asyncio.sleep(0.2)
-
-    assert len(validator.convert_calls) == 2
-    filenames = {r.filename for r in validator.convert_calls}
-    assert filenames == {"first.txt", "second.txt"}
-    for result in validator.convert_calls:
-        assert result.bank_id == bank_id
-        assert result.parser_name == "markitdown"
-        assert result.output_chars > 0
-        assert result.success is True
-
-
-@pytest.mark.asyncio
-async def test_on_file_convert_complete_hook_not_called_on_conversion_failure(memory_no_llm_verify, sample_txt_content):
-    """Test that on_file_convert_complete is NOT called when file conversion fails."""
-    from hindsight_api.engine.parsers.base import FileParser
-    from hindsight_api.models import RequestContext
-
-    bank_id = "test_file_convert_hook_fail_bank"
-    validator = FileConvertTrackingValidator()
-    memory_no_llm_verify._operation_validator = validator
-
-    class FailingParser(FileParser):
-        async def convert(self, file_data: bytes, filename: str) -> str:
-            raise RuntimeError("Mock conversion failure")
-
-        def supports(self, filename: str, content_type: str | None = None) -> bool:
-            return filename.endswith(".hookfail")
-
-        def name(self) -> str:
-            return "hookfail_parser"
-
-    memory_no_llm_verify._parser_registry.register(FailingParser())
-
-    context = RequestContext(internal=True)
-    await memory_no_llm_verify.get_bank_profile(bank_id, request_context=context)
-
-    class MockFile:
-        def __init__(self, content, filename, content_type):
-            self.content = content
-            self.filename = filename
-            self.content_type = content_type
-
-        async def read(self):
-            return self.content
-
-    file_items = [
-        {
-            "file": MockFile(sample_txt_content, "bad.hookfail", "application/octet-stream"),
-            "document_id": "fail_hook_doc",
-            "context": None,
-            "metadata": {},
-            "tags": [],
-            "timestamp": None,
-            "parser": ["hookfail_parser"],
-        }
-    ]
-
-    await memory_no_llm_verify.submit_async_file_retain(
-        bank_id=bank_id,
-        file_items=file_items,
-        document_tags=None,
-        request_context=context,
-    )
-
-    await asyncio.sleep(0.2)
-
-    assert len(validator.convert_calls) == 0

hindsight-api-slim/tests/test_file_storage_s3.py

@@ -1,257 +0,0 @@
-"""
-Integration tests for S3FileStorage against a SeaweedFS Docker container.
-
-SeaweedFS (Apache 2.0) provides an S3-compatible API via `weed server -s3`.
-Requires Docker to be running. Tests are skipped automatically if Docker is unavailable.
-"""
-
-import json
-import logging
-import os
-import subprocess
-import tempfile
-import time
-import uuid
-
-import httpx
-import pytest
-from httpx import ASGITransport, AsyncClient
-
-logger = logging.getLogger(__name__)
-
-try:
-    from testcontainers.core.container import DockerContainer
-
-    _has_testcontainers = True
-except ImportError:
-    _has_testcontainers = False
-
-_in_ci = os.getenv("CI") == "true"
-
-pytestmark = [
-    pytest.mark.skipif(not _has_testcontainers, reason="testcontainers not installed"),
-    pytest.mark.skipif(_in_ci, reason="SeaweedFS Docker image pull too slow in CI"),
-    pytest.mark.timeout(300),
-]
-
-SEAWEEDFS_S3_PORT = 8333
-TEST_BUCKET = "hindsight-test"
-ACCESS_KEY = "test_access_key"
-SECRET_KEY = "test_secret_key"
-
-# SeaweedFS S3 IAM config granting full access to our test credentials
-_S3_CONFIG = {
-    "identities": [
-        {
-            "name": "test-user",
-            "credentials": [{"accessKey": ACCESS_KEY, "secretKey": SECRET_KEY}],
-            "actions": ["Admin", "Read", "Write", "List"],
-        }
-    ]
-}
-
-
-def _docker_available() -> bool:
-    """Check if Docker daemon is running."""
-    try:
-        result = subprocess.run(
-            ["docker", "info"],
-            capture_output=True,
-            timeout=5,
-        )
-        return result.returncode == 0
-    except (FileNotFoundError, subprocess.TimeoutExpired):
-        return False
-
-
-def _wait_for_seaweedfs(endpoint: str, timeout: int = 30) -> None:
-    """Poll SeaweedFS S3 endpoint until ready."""
-    deadline = time.time() + timeout
-    while time.time() < deadline:
-        try:
-            resp = httpx.get(endpoint, timeout=2)
-            # 200 = no auth, 403 = auth enabled but gateway is up — either means ready
-            if resp.status_code in (200, 403):
-                logger.info("SeaweedFS S3 is ready at %s", endpoint)
-                return
-        except httpx.HTTPError:
-            pass
-        time.sleep(0.5)
-    raise TimeoutError(f"SeaweedFS did not become ready at {endpoint} within {timeout}s")
-
-
-@pytest.fixture(scope="module")
-def seaweedfs_container():
-    """Start a SeaweedFS container for the test module, shared across all tests.
-
-    Mounts an s3.json config file to set up S3 credentials for the test user.
-    """
-    if not _docker_available():
-        pytest.skip("Docker is not available")
-
-    # Write S3 IAM config to a temp file that persists for the module scope
-    s3_config_file = tempfile.NamedTemporaryFile(mode="w", suffix=".json", delete=False)
-    json.dump(_S3_CONFIG, s3_config_file)
-    s3_config_file.flush()
-
-    container = (
-        DockerContainer(image="chrislusf/seaweedfs:latest")
-        .with_exposed_ports(SEAWEEDFS_S3_PORT)
-        .with_volume_mapping(s3_config_file.name, "/etc/seaweedfs/s3.json", "ro")
-        .with_command(
-            f"server -s3 -s3.port={SEAWEEDFS_S3_PORT} -s3.config=/etc/seaweedfs/s3.json -ip.bind=0.0.0.0"
-        )
-    )
-
-    container.start()
-
-    try:
-        host = container.get_container_host_ip()
-        port = container.get_exposed_port(SEAWEEDFS_S3_PORT)
-        endpoint = f"http://{host}:{port}"
-
-        _wait_for_seaweedfs(endpoint, timeout=240)
-
-        # Create test bucket using obstore (proper SigV4 signing)
-        import obstore as obs
-        from obstore.store import S3Store
-
-        admin_store = S3Store(
-            TEST_BUCKET,
-            endpoint=endpoint,
-            region="us-east-1",
-            access_key_id=ACCESS_KEY,
-            secret_access_key=SECRET_KEY,
-            allow_http=True,
-        )
-        # SeaweedFS auto-creates buckets on first write
-        obs.put(admin_store, ".bucket-init", b"")
-        obs.delete(admin_store, ".bucket-init")
-        logger.info("Test bucket '%s' is ready", TEST_BUCKET)
-
-        yield {
-            "endpoint": endpoint,
-            "access_key": ACCESS_KEY,
-            "secret_key": SECRET_KEY,
-            "bucket": TEST_BUCKET,
-        }
-    finally:
-        container.stop()
-        import os
-
-        os.unlink(s3_config_file.name)
-
-
-@pytest.fixture
-def s3_storage(seaweedfs_container):
-    """Create an S3FileStorage instance pointing at the SeaweedFS container."""
-    from hindsight_api.engine.storage.s3 import S3FileStorage
-
-    return S3FileStorage(
-        bucket=seaweedfs_container["bucket"],
-        region="us-east-1",
-        endpoint=seaweedfs_container["endpoint"],
-        access_key_id=seaweedfs_container["access_key"],
-        secret_access_key=seaweedfs_container["secret_key"],
-    )
-
-
-@pytest.mark.asyncio
-async def test_s3_storage_store_and_retrieve(s3_storage):
-    """Store a file, retrieve it, verify bytes match."""
-    content = b"Hello, SeaweedFS! This is a test file."
-    key = f"test/{uuid.uuid4()}.txt"
-
-    stored_key = await s3_storage.store(
-        file_data=content,
-        key=key,
-        metadata={"content_type": "text/plain"},
-    )
-    assert stored_key == key
-
-    retrieved = await s3_storage.retrieve(key)
-    assert retrieved == content
-
-
-@pytest.mark.asyncio
-async def test_s3_storage_exists_and_delete(s3_storage):
-    """Store, check exists=True, delete, check exists=False."""
-    content = b"File to be deleted."
-    key = f"test/{uuid.uuid4()}.txt"
-
-    await s3_storage.store(file_data=content, key=key)
-
-    assert await s3_storage.exists(key) is True
-
-    await s3_storage.delete(key)
-
-    assert await s3_storage.exists(key) is False
-
-
-@pytest.mark.asyncio
-async def test_s3_storage_file_not_found(s3_storage):
-    """Retrieve a non-existent key, expect FileNotFoundError."""
-    with pytest.raises(FileNotFoundError):
-        await s3_storage.retrieve(f"nonexistent/{uuid.uuid4()}.txt")
-
-
-@pytest.mark.asyncio
-async def test_s3_storage_get_download_url(s3_storage):
-    """Store a file, get a presigned URL, verify it's a valid URL string."""
-    content = b"Presigned URL test content."
-    key = f"test/{uuid.uuid4()}.txt"
-
-    await s3_storage.store(file_data=content, key=key)
-
-    url = await s3_storage.get_download_url(key, expires_in=300)
-    assert isinstance(url, str)
-    assert url.startswith("http")
-    assert key in url
-
-
-@pytest.mark.asyncio
-async def test_s3_file_retain_api_end_to_end(seaweedfs_container, memory_no_llm_verify):
-    """Full HTTP API flow: upload file via /files/retain with S3 storage backend."""
-    from hindsight_api.api.http import create_app
-    from hindsight_api.engine.storage.s3 import S3FileStorage
-
-    # Swap the engine's file storage to use the SeaweedFS-backed S3 storage
-    original_storage = memory_no_llm_verify._file_storage
-    s3_storage = S3FileStorage(
-        bucket=seaweedfs_container["bucket"],
-        region="us-east-1",
-        endpoint=seaweedfs_container["endpoint"],
-        access_key_id=seaweedfs_container["access_key"],
-        secret_access_key=seaweedfs_container["secret_key"],
-    )
-    memory_no_llm_verify._file_storage = s3_storage
-
-    try:
-        app = create_app(memory_no_llm_verify, initialize_memory=False)
-
-        async with AsyncClient(transport=ASGITransport(app=app), base_url="http://test") as client:
-            bank_id = f"test-s3-bank-{uuid.uuid4().hex[:8]}"
-            bank_response = await client.put(f"/v1/default/banks/{bank_id}", json={"name": "S3 Test Bank"})
-            assert bank_response.status_code in (200, 201)
-
-            txt_content = b"Alice works at Acme Corp. She joined in 2024."
-            request_data = {
-                "document_tags": ["s3-test"],
-                "async": True,
-            }
-
-            files = {"files": ("notes.txt", txt_content, "text/plain")}
-            data = {"request": json.dumps(request_data)}
-
-            response = await client.post(
-                f"/v1/default/banks/{bank_id}/files/retain",
-                files=files,
-                data=data,
-            )
-
-            assert response.status_code == 200
-            result = response.json()
-            assert "operation_ids" in result
-            assert len(result["operation_ids"]) == 1
-    finally:
-        memory_no_llm_verify._file_storage = original_storage

hindsight-api-slim/tests/test_gemini_safety_settings.py

@@ -1,340 +0,0 @@
-"""
-Tests for Gemini safety settings feature.
-
-Verifies that:
-- Safety settings are read from env var and stored on GeminiLLM instances
-- Settings are applied to GenerateContentConfig in call() and call_with_tools()
-- The context variable override allows per-bank settings at request time
-- None (unset) means Gemini's default safety settings are used (no override)
-"""
-
-import os
-from unittest.mock import AsyncMock, MagicMock, patch
-
-import pytest
-
-pytest.importorskip("google.genai")
-
-
-SAMPLE_SAFETY_SETTINGS = [
-    {"category": "HARM_CATEGORY_HARASSMENT", "threshold": "BLOCK_NONE"},
-    {"category": "HARM_CATEGORY_HATE_SPEECH", "threshold": "BLOCK_NONE"},
-    {"category": "HARM_CATEGORY_SEXUALLY_EXPLICIT", "threshold": "BLOCK_NONE"},
-    {"category": "HARM_CATEGORY_DANGEROUS_CONTENT", "threshold": "BLOCK_NONE"},
-]
-
-
-# ─── Config / env var parsing ─────────────────────────────────────────────────
-
-
-def test_gemini_safety_settings_parsed_from_env():
-    """Safety settings JSON from env var is parsed into HindsightConfig."""
-    import json
-
-    from hindsight_api.config import ENV_LLM_GEMINI_SAFETY_SETTINGS, HindsightConfig, clear_config_cache
-
-    settings_json = json.dumps(SAMPLE_SAFETY_SETTINGS)
-    with patch.dict(os.environ, {ENV_LLM_GEMINI_SAFETY_SETTINGS: settings_json}, clear=False):
-        clear_config_cache()
-        config = HindsightConfig.from_env()
-        assert config.llm_gemini_safety_settings == SAMPLE_SAFETY_SETTINGS
-        clear_config_cache()
-
-
-def test_gemini_safety_settings_default_is_none():
-    """When env var is not set, llm_gemini_safety_settings defaults to None."""
-    from hindsight_api.config import ENV_LLM_GEMINI_SAFETY_SETTINGS, HindsightConfig, clear_config_cache
-
-    env = {k: v for k, v in os.environ.items() if k != ENV_LLM_GEMINI_SAFETY_SETTINGS}
-    with patch.dict(os.environ, env, clear=True):
-        clear_config_cache()
-        config = HindsightConfig.from_env()
-        assert config.llm_gemini_safety_settings is None
-        clear_config_cache()
-
-
-def test_gemini_safety_settings_is_configurable_field():
-    """llm_gemini_safety_settings appears in configurable (per-bank) fields."""
-    from hindsight_api.config import HindsightConfig
-
-    assert "llm_gemini_safety_settings" in HindsightConfig.get_configurable_fields()
-
-
-def test_gemini_safety_settings_not_in_credential_fields():
-    """llm_gemini_safety_settings is NOT a credential — it is safe to expose via API."""
-    from hindsight_api.config import HindsightConfig
-
-    assert "llm_gemini_safety_settings" not in HindsightConfig.get_credential_fields()
-
-
-# ─── GeminiLLM instance ───────────────────────────────────────────────────────
-
-
-def _make_gemini_provider(safety_settings=None):
-    """Return a GeminiLLM instance with a mocked genai.Client."""
-    with patch("google.genai.Client") as mock_client_cls:
-        mock_client_cls.return_value = MagicMock()
-        from hindsight_api.engine.providers.gemini_llm import GeminiLLM
-
-        provider = GeminiLLM(
-            provider="gemini",
-            api_key="fake-api-key",
-            base_url="",
-            model="gemini-2.5-flash",
-            gemini_safety_settings=safety_settings,
-        )
-        # Replace client with a fresh mock so we can inspect calls
-        provider._client = MagicMock()
-        return provider
-
-
-def test_gemini_llm_stores_safety_settings():
-    """GeminiLLM stores safety settings passed at construction."""
-    provider = _make_gemini_provider(safety_settings=SAMPLE_SAFETY_SETTINGS)
-    assert provider._safety_settings == SAMPLE_SAFETY_SETTINGS
-
-
-def test_gemini_llm_no_safety_settings_is_none():
-    """GeminiLLM._safety_settings is None when not provided."""
-    provider = _make_gemini_provider(safety_settings=None)
-    assert provider._safety_settings is None
-
-
-# ─── call() applies safety settings ──────────────────────────────────────────
-
-
-@pytest.mark.asyncio
-async def test_call_applies_safety_settings():
-    """call() includes safety_settings in GenerateContentConfig when configured."""
-    from google.genai import types as genai_types
-
-    provider = _make_gemini_provider(safety_settings=SAMPLE_SAFETY_SETTINGS)
-
-    # Build a fake successful response
-    fake_response = MagicMock()
-    fake_response.text = "hello"
-    fake_response.candidates = [MagicMock(finish_reason="STOP")]
-    fake_response.usage_metadata = MagicMock(prompt_token_count=5, candidates_token_count=2)
-
-    provider._client.aio.models.generate_content = AsyncMock(return_value=fake_response)
-
-    await provider.call(
-        messages=[{"role": "user", "content": "hi"}],
-        scope="test",
-    )
-
-    # Inspect the config passed to generate_content
-    call_args = provider._client.aio.models.generate_content.call_args
-    config_arg = call_args.kwargs.get("config") or call_args.args[0] if call_args.args else None
-    # config may be in kwargs or positional; grab from kwargs
-    config_arg = call_args.kwargs.get("config")
-
-    assert config_arg is not None, "GenerateContentConfig should have been passed"
-    assert hasattr(config_arg, "safety_settings"), "Config should have safety_settings"
-    assert config_arg.safety_settings is not None
-
-    categories = [s.category.value if hasattr(s.category, "value") else str(s.category) for s in config_arg.safety_settings]
-    assert "HARM_CATEGORY_HARASSMENT" in categories
-    assert "HARM_CATEGORY_HATE_SPEECH" in categories
-    assert "HARM_CATEGORY_SEXUALLY_EXPLICIT" in categories
-    assert "HARM_CATEGORY_DANGEROUS_CONTENT" in categories
-
-    thresholds = [s.threshold.value if hasattr(s.threshold, "value") else str(s.threshold) for s in config_arg.safety_settings]
-    assert all(t == "BLOCK_NONE" for t in thresholds)
-
-
-@pytest.mark.asyncio
-async def test_call_no_safety_settings_omits_key():
-    """call() does NOT add safety_settings to GenerateContentConfig when none configured."""
-    provider = _make_gemini_provider(safety_settings=None)
-
-    fake_response = MagicMock()
-    fake_response.text = "hello"
-    fake_response.candidates = [MagicMock(finish_reason="STOP")]
-    fake_response.usage_metadata = MagicMock(prompt_token_count=5, candidates_token_count=2)
-
-    provider._client.aio.models.generate_content = AsyncMock(return_value=fake_response)
-
-    await provider.call(
-        messages=[{"role": "user", "content": "hi"}],
-        scope="test",
-    )
-
-    call_args = provider._client.aio.models.generate_content.call_args
-    config_arg = call_args.kwargs.get("config")
-
-    # When no safety settings, config is either None or lacks safety_settings
-    if config_arg is not None:
-        assert not hasattr(config_arg, "safety_settings") or config_arg.safety_settings is None
-
-
-# ─── call_with_tools() applies safety settings ────────────────────────────────
-
-
-@pytest.mark.asyncio
-async def test_call_with_tools_applies_safety_settings():
-    """call_with_tools() includes safety_settings in GenerateContentConfig."""
-    provider = _make_gemini_provider(safety_settings=SAMPLE_SAFETY_SETTINGS)
-
-    # Build a fake tool-use response (no tool calls, just text)
-    fake_part = MagicMock()
-    fake_part.text = "answer"
-    fake_part.function_call = None
-
-    fake_candidate = MagicMock()
-    fake_candidate.content = MagicMock(parts=[fake_part])
-
-    fake_response = MagicMock()
-    fake_response.candidates = [fake_candidate]
-    fake_response.usage_metadata = MagicMock(prompt_token_count=5, candidates_token_count=3)
-
-    provider._client.aio.models.generate_content = AsyncMock(return_value=fake_response)
-
-    tools = [
-        {
-            "type": "function",
-            "function": {
-                "name": "test_tool",
-                "description": "A test tool",
-                "parameters": {"type": "object", "properties": {}, "required": []},
-            },
-        }
-    ]
-
-    await provider.call_with_tools(
-        messages=[{"role": "user", "content": "hi"}],
-        tools=tools,
-        scope="test",
-    )
-
-    call_args = provider._client.aio.models.generate_content.call_args
-    config_arg = call_args.kwargs.get("config")
-
-    assert config_arg is not None
-    assert config_arg.safety_settings is not None
-    categories = [s.category.value if hasattr(s.category, "value") else str(s.category) for s in config_arg.safety_settings]
-    assert "HARM_CATEGORY_HARASSMENT" in categories
-
-
-# ─── with_config() override ───────────────────────────────────────────────────
-
-
-def _make_llm_provider(safety_settings=None):
-    """Return an LLMProvider (wrapping GeminiLLM) with a mocked genai.Client."""
-    with patch("google.genai.Client") as mock_client_cls:
-        mock_client_cls.return_value = MagicMock()
-        from hindsight_api.engine.llm_wrapper import LLMProvider
-
-        provider = LLMProvider(
-            provider="gemini",
-            api_key="fake-api-key",
-            base_url="",
-            model="gemini-2.5-flash",
-            gemini_safety_settings=safety_settings,
-        )
-        # Replace the underlying Gemini client with a fresh mock
-        provider._provider_impl._client = MagicMock()
-        return provider
-
-
-def _fake_response():
-    r = MagicMock()
-    r.text = "hello"
-    r.candidates = [MagicMock(finish_reason="STOP")]
-    r.usage_metadata = MagicMock(prompt_token_count=5, candidates_token_count=2)
-    return r
-
-
-def _make_config(safety_settings):
-    """Return a minimal config-like object with llm_gemini_safety_settings."""
-    cfg = MagicMock()
-    cfg.llm_gemini_safety_settings = safety_settings
-    return cfg
-
-
-@pytest.mark.asyncio
-async def test_with_config_overrides_instance_settings():
-    """with_config() settings take precedence over the provider instance defaults."""
-    instance_settings = [{"category": "HARM_CATEGORY_HARASSMENT", "threshold": "BLOCK_ONLY_HIGH"}]
-    override_settings = [{"category": "HARM_CATEGORY_HATE_SPEECH", "threshold": "BLOCK_NONE"}]
-
-    provider = _make_llm_provider(safety_settings=instance_settings)
-    provider._provider_impl._client.aio.models.generate_content = AsyncMock(return_value=_fake_response())
-
-    configured = provider.with_config(_make_config(override_settings))
-    await configured.call(messages=[{"role": "user", "content": "hi"}], scope="test")
-
-    config_arg = provider._provider_impl._client.aio.models.generate_content.call_args.kwargs.get("config")
-    assert config_arg is not None
-    categories = [s.category.value if hasattr(s.category, "value") else str(s.category) for s in config_arg.safety_settings]
-    # Should use override_settings (HATE_SPEECH), not instance_settings (HARASSMENT)
-    assert "HARM_CATEGORY_HATE_SPEECH" in categories
-    assert "HARM_CATEGORY_HARASSMENT" not in categories
-
-
-@pytest.mark.asyncio
-async def test_with_config_none_falls_back_to_instance():
-    """When with_config() supplies None, the instance default is used."""
-    instance_settings = [{"category": "HARM_CATEGORY_HARASSMENT", "threshold": "BLOCK_NONE"}]
-
-    provider = _make_llm_provider(safety_settings=instance_settings)
-    provider._provider_impl._client.aio.models.generate_content = AsyncMock(return_value=_fake_response())
-
-    configured = provider.with_config(_make_config(None))
-    await configured.call(messages=[{"role": "user", "content": "hi"}], scope="test")
-
-    config_arg = provider._provider_impl._client.aio.models.generate_content.call_args.kwargs.get("config")
-    assert config_arg is not None
-    categories = [s.category.value if hasattr(s.category, "value") else str(s.category) for s in config_arg.safety_settings]
-    assert "HARM_CATEGORY_HARASSMENT" in categories
-
-
-@pytest.mark.asyncio
-async def test_with_config_resets_after_call():
-    """The ContextVar is properly reset after a with_config() call (no leakage)."""
-    from hindsight_api.engine.providers.gemini_llm import _safety_settings_ctx
-
-    settings = [{"category": "HARM_CATEGORY_HARASSMENT", "threshold": "BLOCK_NONE"}]
-    provider = _make_llm_provider(safety_settings=None)
-    provider._provider_impl._client.aio.models.generate_content = AsyncMock(return_value=_fake_response())
-
-    before = _safety_settings_ctx.get()
-    configured = provider.with_config(_make_config(settings))
-    await configured.call(messages=[{"role": "user", "content": "hi"}], scope="test")
-    after = _safety_settings_ctx.get()
-
-    assert after == before  # ContextVar restored to its original value
-
-
-# ─── LLMProvider reads safety settings from config ────────────────────────────
-
-
-def test_llm_provider_reads_safety_settings_from_config():
-    """LLMProvider reads llm_gemini_safety_settings from global config for Gemini provider."""
-    import json
-
-    from hindsight_api.config import ENV_LLM_GEMINI_SAFETY_SETTINGS, clear_config_cache
-
-    settings_json = json.dumps(SAMPLE_SAFETY_SETTINGS)
-    env_overrides = {
-        "HINDSIGHT_API_LLM_PROVIDER": "gemini",
-        "HINDSIGHT_API_LLM_API_KEY": "fake-key",
-        ENV_LLM_GEMINI_SAFETY_SETTINGS: settings_json,
-    }
-
-    with patch.dict(os.environ, env_overrides, clear=False):
-        clear_config_cache()
-        with patch("google.genai.Client") as mock_client_cls:
-            mock_client_cls.return_value = MagicMock()
-            from hindsight_api.engine.llm_wrapper import LLMProvider
-
-            provider = LLMProvider(
-                provider="gemini",
-                api_key="fake-key",
-                base_url="",
-                model="gemini-2.5-flash",
-            )
-
-            assert provider.gemini_safety_settings == SAMPLE_SAFETY_SETTINGS
-
-        clear_config_cache()

hindsight-api-slim/tests/test_graph_filtering.py

@@ -1,171 +0,0 @@
-"""
-Tests for server-side filtering in the graph API endpoint.
-
-Verifies that q (text search) and tags filters work correctly
-when passed as query parameters to GET /v1/default/banks/{bank_id}/graph.
-"""
-from datetime import datetime
-
-import httpx
-import pytest
-import pytest_asyncio
-
-from hindsight_api.api import create_app
-
-
-@pytest_asyncio.fixture
-async def api_client(memory):
-    """Create an async test client for the FastAPI app."""
-    app = create_app(memory, initialize_memory=False)
-    transport = httpx.ASGITransport(app=app)
-    async with httpx.AsyncClient(transport=transport, base_url="http://test") as client:
-        yield client
-
-
-@pytest.fixture
-def test_bank_id():
-    """Provide a unique bank ID for this test run."""
-    return f"graph_filter_test_{datetime.now().timestamp()}"
-
-
-@pytest.mark.asyncio
-async def test_graph_no_filter_returns_all(api_client, test_bank_id):
-    """Without filters the graph endpoint returns all memories."""
-    response = await api_client.post(
-        f"/v1/default/banks/{test_bank_id}/memories",
-        json={
-            "items": [
-                {"content": "Alice loves hiking in the mountains.", "tags": ["user_alice"]},
-                {"content": "Bob enjoys swimming at the beach.", "tags": ["user_bob"]},
-            ]
-        },
-    )
-    assert response.status_code == 200
-
-    response = await api_client.get(f"/v1/default/banks/{test_bank_id}/graph")
-    assert response.status_code == 200
-    data = response.json()
-    assert "table_rows" in data
-    texts = [row["text"] for row in data["table_rows"]]
-    assert any("Alice" in t for t in texts)
-    assert any("Bob" in t for t in texts)
-
-
-@pytest.mark.asyncio
-async def test_graph_q_filter_returns_matching(api_client, test_bank_id):
-    """The q parameter filters memories by text content."""
-    response = await api_client.post(
-        f"/v1/default/banks/{test_bank_id}/memories",
-        json={
-            "items": [
-                {"content": "Alice loves hiking in the mountains."},
-                {"content": "Bob enjoys swimming at the beach."},
-            ]
-        },
-    )
-    assert response.status_code == 200
-
-    response = await api_client.get(f"/v1/default/banks/{test_bank_id}/graph", params={"q": "Alice"})
-    assert response.status_code == 200
-    data = response.json()
-    texts = [row["text"] for row in data["table_rows"]]
-    assert all("Alice" in t or "alice" in t.lower() for t in texts), (
-        f"Expected only Alice memories, got: {texts}"
-    )
-    assert not any("Bob" in t for t in texts)
-
-
-@pytest.mark.asyncio
-async def test_graph_q_filter_case_insensitive(api_client, test_bank_id):
-    """The q filter is case-insensitive."""
-    response = await api_client.post(
-        f"/v1/default/banks/{test_bank_id}/memories",
-        json={
-            "items": [
-                {"content": "Alice loves hiking in the mountains."},
-                {"content": "Bob enjoys swimming at the beach."},
-            ]
-        },
-    )
-    assert response.status_code == 200
-
-    response = await api_client.get(f"/v1/default/banks/{test_bank_id}/graph", params={"q": "alice"})
-    assert response.status_code == 200
-    data = response.json()
-    texts = [row["text"] for row in data["table_rows"]]
-    assert any("Alice" in t for t in texts)
-    assert not any("Bob" in t for t in texts)
-
-
-@pytest.mark.asyncio
-async def test_graph_tags_filter_returns_matching(api_client, test_bank_id):
-    """The tags parameter filters memories to only those with matching tags."""
-    response = await api_client.post(
-        f"/v1/default/banks/{test_bank_id}/memories",
-        json={
-            "items": [
-                {"content": "Alice loves hiking.", "tags": ["user_alice"]},
-                {"content": "Bob enjoys swimming.", "tags": ["user_bob"]},
-            ]
-        },
-    )
-    assert response.status_code == 200
-
-    response = await api_client.get(
-        f"/v1/default/banks/{test_bank_id}/graph",
-        params={"tags": "user_alice", "tags_match": "all_strict"},
-    )
-    assert response.status_code == 200
-    data = response.json()
-    texts = [row["text"] for row in data["table_rows"]]
-    assert any("Alice" in t for t in texts)
-    assert not any("Bob" in t for t in texts)
-
-
-@pytest.mark.asyncio
-async def test_graph_q_and_tags_filter_combined(api_client, test_bank_id):
-    """Combining q and tags filters applies both server-side."""
-    response = await api_client.post(
-        f"/v1/default/banks/{test_bank_id}/memories",
-        json={
-            "items": [
-                {"content": "Alice loves hiking.", "tags": ["user_alice"]},
-                {"content": "Alice also loves coding.", "tags": ["user_alice"]},
-                {"content": "Bob enjoys swimming.", "tags": ["user_bob"]},
-            ]
-        },
-    )
-    assert response.status_code == 200
-
-    response = await api_client.get(
-        f"/v1/default/banks/{test_bank_id}/graph",
-        params={"q": "hiking", "tags": "user_alice", "tags_match": "all_strict"},
-    )
-    assert response.status_code == 200
-    data = response.json()
-    texts = [row["text"] for row in data["table_rows"]]
-    assert any("hiking" in t.lower() for t in texts)
-    assert not any("coding" in t.lower() for t in texts)
-    assert not any("Bob" in t for t in texts)
-
-
-@pytest.mark.asyncio
-async def test_graph_q_filter_empty_results(api_client, test_bank_id):
-    """The q filter returns empty results when no memory matches."""
-    response = await api_client.post(
-        f"/v1/default/banks/{test_bank_id}/memories",
-        json={
-            "items": [
-                {"content": "Alice loves hiking."},
-            ]
-        },
-    )
-    assert response.status_code == 200
-
-    response = await api_client.get(
-        f"/v1/default/banks/{test_bank_id}/graph",
-        params={"q": "zzznomatchzzz"},
-    )
-    assert response.status_code == 200
-    data = response.json()
-    assert data["table_rows"] == []

hindsight-api-slim/tests/test_hnsw_indexes.py

@@ -1,193 +0,0 @@
-"""
-Tests for per-bank HNSW index lifecycle and UNION ALL retrieval.
-
-Covers:
-- _hnsw_index_name deterministic naming
-- Per-bank HNSW indexes created on bank creation (retain_async / ensure_bank_exists)
-- Per-bank HNSW indexes dropped on bank deletion
-- retrieve_semantic_bm25_combined groups results correctly by fact_type and source
-"""
-import uuid
-from datetime import datetime, timezone
-
-import pytest
-
-from hindsight_api.engine.retain.bank_utils import _HNSW_FACT_TYPES, _hnsw_index_name
-
-
-# ---------------------------------------------------------------------------
-# Unit tests — no DB required
-# ---------------------------------------------------------------------------
-
-
-class TestHnswIndexName:
-    def test_deterministic(self):
-        uid = "550e8400-e29b-41d4-a716-446655440000"
-        assert _hnsw_index_name("world", uid) == _hnsw_index_name("world", uid)
-
-    def test_strips_dashes(self):
-        uid = "550e8400-e29b-41d4-a716-446655440000"
-        name = _hnsw_index_name("world", uid)
-        # uid16 should be hex chars only
-        assert "-" not in name
-
-    def test_uses_first_16_hex_chars(self):
-        uid = "550e8400-e29b-41d4-a716-446655440000"
-        uid16 = uid.replace("-", "")[:16]  # "550e8400e29b41d4"
-        assert name_ends_with(name=_hnsw_index_name("world", uid), suffix=uid16)
-
-    def test_suffix_per_fact_type(self):
-        uid = "550e8400-e29b-41d4-a716-446655440000"
-        names = {ft: _hnsw_index_name(ft, uid) for ft in _HNSW_FACT_TYPES}
-        # All three names must be distinct
-        assert len(set(names.values())) == 3
-
-    def test_all_fact_types_covered(self):
-        assert set(_HNSW_FACT_TYPES) == {"world", "experience", "observation"}
-
-    def test_fits_pg_identifier_limit(self):
-        # PostgreSQL max identifier length is 63 chars
-        uid = "f" * 32  # simulated UUID without dashes
-        for ft in _HNSW_FACT_TYPES:
-            assert len(_hnsw_index_name(ft, uid)) <= 63
-
-
-def name_ends_with(name: str, suffix: str) -> bool:
-    return name.endswith(suffix)
-
-
-# ---------------------------------------------------------------------------
-# Integration tests — require DB (memory fixture)
-# ---------------------------------------------------------------------------
-
-
-async def _get_bank_hnsw_indexes(pool, bank_id: str) -> list[str]:
-    """Return index names for memory_units that match the per-bank pattern."""
-    async with pool.acquire() as conn:
-        rows = await conn.fetch(
-            """
-            SELECT indexname
-            FROM pg_indexes
-            WHERE tablename = 'memory_units'
-              AND indexname LIKE 'idx_mu_emb_%'
-              AND indexdef LIKE $1
-            ORDER BY indexname
-            """,
-            f"%bank_id = '{bank_id}'%",
-        )
-    return [row["indexname"] for row in rows]
-
-
-@pytest.mark.asyncio
-async def test_retain_creates_per_bank_hnsw_indexes(memory, request_context):
-    """retain_async on a new bank must create 3 per-(bank, fact_type) HNSW indexes."""
-    bank_id = f"test_hnsw_create_{uuid.uuid4().hex[:8]}"
-    try:
-        await memory.retain_async(
-            bank_id=bank_id,
-            content="Alice is a software engineer.",
-            request_context=request_context,
-        )
-        indexes = await _get_bank_hnsw_indexes(memory._pool, bank_id)
-        assert len(indexes) == 3, f"Expected 3 per-bank HNSW indexes, got: {indexes}"
-        for ft_short in _HNSW_FACT_TYPES.values():
-            assert any(ft_short in idx for idx in indexes), (
-                f"Missing index for fact_type short '{ft_short}' in {indexes}"
-            )
-    finally:
-        await memory.delete_bank(bank_id, request_context=request_context)
-
-
-@pytest.mark.asyncio
-async def test_delete_bank_drops_hnsw_indexes(memory, request_context):
-    """delete_bank must drop all per-bank HNSW indexes."""
-    bank_id = f"test_hnsw_drop_{uuid.uuid4().hex[:8]}"
-
-    await memory.retain_async(
-        bank_id=bank_id,
-        content="Bob is a data scientist.",
-        request_context=request_context,
-    )
-    # Verify indexes exist before deletion
-    indexes_before = await _get_bank_hnsw_indexes(memory._pool, bank_id)
-    assert len(indexes_before) == 3
-
-    await memory.delete_bank(bank_id, request_context=request_context)
-
-    indexes_after = await _get_bank_hnsw_indexes(memory._pool, bank_id)
-    assert indexes_after == [], f"Indexes should be dropped after bank deletion, got: {indexes_after}"
-
-
-@pytest.mark.asyncio
-async def test_retain_idempotent_bank_creation(memory, request_context):
-    """Retaining into the same bank twice must not error and still have exactly 3 indexes."""
-    bank_id = f"test_hnsw_idem_{uuid.uuid4().hex[:8]}"
-    try:
-        await memory.retain_async(
-            bank_id=bank_id,
-            content="Carol is a product manager.",
-            request_context=request_context,
-        )
-        await memory.retain_async(
-            bank_id=bank_id,
-            content="Carol joined the company in 2022.",
-            request_context=request_context,
-        )
-        indexes = await _get_bank_hnsw_indexes(memory._pool, bank_id)
-        assert len(indexes) == 3
-    finally:
-        await memory.delete_bank(bank_id, request_context=request_context)
-
-
-@pytest.mark.asyncio
-async def test_retrieve_semantic_bm25_grouped_by_fact_type(memory, request_context):
-    """
-    retrieve_semantic_bm25_combined must return a dict keyed by fact_type with
-    (semantic_list, bm25_list) tuples.  All returned facts must belong to their
-    declared fact_type.
-    """
-    from hindsight_api.engine.search.retrieval import retrieve_semantic_bm25_combined
-
-    bank_id = f"test_retrieval_{uuid.uuid4().hex[:8]}"
-    try:
-        await memory.retain_async(
-            bank_id=bank_id,
-            content=(
-                "Alice is a software engineer at TechCorp. "
-                "She visited Paris in 2023 for a conference."
-            ),
-            context="background",
-            event_date=datetime(2023, 6, 1, tzinfo=timezone.utc),
-            request_context=request_context,
-        )
-
-        query_emb = memory.embeddings.encode(["software engineer Alice"])
-        query_emb_str = str(query_emb[0])
-
-        fact_types = ["world", "experience"]
-        async with memory._pool.acquire() as conn:
-            results = await retrieve_semantic_bm25_combined(
-                conn=conn,
-                query_emb_str=query_emb_str,
-                query_text="software engineer Alice",
-                bank_id=bank_id,
-                fact_types=fact_types,
-                limit=5,
-            )
-
-        # Must return an entry for every requested fact_type
-        assert set(results.keys()) == set(fact_types)
-
-        for ft, (sem, bm25) in results.items():
-            # Semantic and BM25 lists must be lists
-            assert isinstance(sem, list)
-            assert isinstance(bm25, list)
-            # All semantic results must declare the correct fact_type
-            for r in sem:
-                assert r.fact_type == ft, f"Semantic result has wrong fact_type: {r.fact_type}"
-            # All BM25 results must declare the correct fact_type
-            for r in bm25:
-                assert r.fact_type == ft, f"BM25 result has wrong fact_type: {r.fact_type}"
-
-    finally:
-        await memory.delete_bank(bank_id, request_context=request_context)

hindsight-api-slim/tests/test_iris_parser.py

@@ -1,72 +0,0 @@
-"""
-Integration tests for the Iris file parser.
-
-Tests are skipped automatically if HINDSIGHT_API_FILE_PARSER_IRIS_TOKEN
-and HINDSIGHT_API_FILE_PARSER_IRIS_ORG_ID are not set in the environment.
-"""
-
-import os
-
-import pytest
-
-from hindsight_api.config import ENV_FILE_PARSER_IRIS_ORG_ID, ENV_FILE_PARSER_IRIS_TOKEN
-from hindsight_api.engine.parsers.iris import IrisParser
-
-_token = os.getenv(ENV_FILE_PARSER_IRIS_TOKEN)
-_org_id = os.getenv(ENV_FILE_PARSER_IRIS_ORG_ID)
-
-pytestmark = pytest.mark.skipif(
-    not (_token and _org_id),
-    reason="HINDSIGHT_API_FILE_PARSER_IRIS_TOKEN and HINDSIGHT_API_FILE_PARSER_IRIS_ORG_ID not set",
-)
-
-# Minimal valid PDF with the text "Hello from Hindsight"
-_SAMPLE_PDF = b"""%PDF-1.4
-1 0 obj
-<< /Type /Catalog /Pages 2 0 R >>
-endobj
-2 0 obj
-<< /Type /Pages /Kids [3 0 R] /Count 1 >>
-endobj
-3 0 obj
-<< /Type /Page /Parent 2 0 R /MediaBox [0 0 612 792]
-   /Contents 4 0 R /Resources << /Font << /F1 << /Type /Font /Subtype /Type1 /BaseFont /Helvetica >> >> >> >>
-endobj
-4 0 obj
-<< /Length 44 >>
-stream
-BT /F1 12 Tf 100 700 Td (Hello from Hindsight) Tj ET
-endstream
-endobj
-xref
-0 5
-0000000000 65535 f
-0000000009 00000 n
-0000000058 00000 n
-0000000115 00000 n
-0000000274 00000 n
-trailer << /Size 5 /Root 1 0 R >>
-startxref
-369
-%%EOF"""
-
-
-@pytest.fixture
-def iris_parser() -> IrisParser:
-    return IrisParser(token=_token, org_id=_org_id)
-
-
-@pytest.mark.asyncio
-async def test_iris_parser_converts_pdf(iris_parser: IrisParser):
-    """IrisParser should extract text from a valid PDF."""
-    result = await iris_parser.convert(_SAMPLE_PDF, "sample.pdf")
-    assert isinstance(result, str)
-    assert len(result) > 0
-
-
-@pytest.mark.asyncio
-async def test_iris_parser_name(iris_parser: IrisParser):
-    """IrisParser.name() should return 'iris'."""
-    assert iris_parser.name() == "iris"
-
-

hindsight-api-slim/tests/test_list_documents.py

@@ -1,174 +0,0 @@
-"""
-Tests for list_documents pagination and tags filtering.
-"""
-from datetime import datetime, timezone
-
-import pytest
-
-
-async def _retain_doc(memory, bank_id, document_id, tags, request_context):
-    """Helper to retain a document with given tags. Uses gibberish content to avoid LLM
-    fact extraction (documents are persisted even with zero facts)."""
-    await memory.retain_batch_async(
-        bank_id=bank_id,
-        contents=[{"content": f"xyzabc123 !@# $$$ {document_id}"}],
-        document_id=document_id,
-        document_tags=tags or None,
-        request_context=request_context,
-    )
-
-
-@pytest.mark.asyncio
-async def test_list_documents_offset_pagination(memory, request_context):
-    """offset parameter returns the correct slice of documents."""
-    bank_id = f"test_list_docs_offset_{datetime.now(timezone.utc).timestamp()}"
-
-    try:
-        for i in range(4):
-            await _retain_doc(memory, bank_id, f"doc-{i:02d}", [], request_context)
-
-        # All documents, ordered by created_at DESC → doc-03, doc-02, doc-01, doc-00
-        all_docs = await memory.list_documents(
-            bank_id=bank_id, limit=10, offset=0, request_context=request_context
-        )
-        assert all_docs["total"] == 4
-        assert len(all_docs["items"]) == 4
-        all_ids = [d["id"] for d in all_docs["items"]]
-
-        # offset=2 should skip the first two and return the remaining two
-        page2 = await memory.list_documents(
-            bank_id=bank_id, limit=10, offset=2, request_context=request_context
-        )
-        assert page2["total"] == 4  # total is always the full count
-        assert len(page2["items"]) == 2
-        assert [d["id"] for d in page2["items"]] == all_ids[2:]
-
-        # offset beyond total returns empty items but correct total
-        beyond = await memory.list_documents(
-            bank_id=bank_id, limit=10, offset=10, request_context=request_context
-        )
-        assert beyond["total"] == 4
-        assert beyond["items"] == []
-
-    finally:
-        await memory.delete_bank(bank_id, request_context=request_context)
-
-
-@pytest.mark.asyncio
-async def test_list_documents_tags_filter_any_strict(memory, request_context):
-    """tags filter with any_strict returns only tagged documents that match."""
-    bank_id = f"test_list_docs_tags_{datetime.now(timezone.utc).timestamp()}"
-
-    try:
-        await _retain_doc(memory, bank_id, "doc-alpha", ["team-a"], request_context)
-        await _retain_doc(memory, bank_id, "doc-beta", ["team-b"], request_context)
-        await _retain_doc(memory, bank_id, "doc-both", ["team-a", "team-b"], request_context)
-        await _retain_doc(memory, bank_id, "doc-untagged", [], request_context)
-
-        # any_strict: only docs with at least one of the given tags, untagged excluded
-        result = await memory.list_documents(
-            bank_id=bank_id,
-            tags=["team-a"],
-            tags_match="any_strict",
-            request_context=request_context,
-        )
-        ids = {d["id"] for d in result["items"]}
-        assert ids == {"doc-alpha", "doc-both"}
-        assert result["total"] == 2
-
-    finally:
-        await memory.delete_bank(bank_id, request_context=request_context)
-
-
-@pytest.mark.asyncio
-async def test_list_documents_tags_filter_any_includes_untagged(memory, request_context):
-    """tags filter with 'any' mode includes untagged documents."""
-    bank_id = f"test_list_docs_tags_any_{datetime.now(timezone.utc).timestamp()}"
-
-    try:
-        await _retain_doc(memory, bank_id, "doc-tagged", ["team-a"], request_context)
-        await _retain_doc(memory, bank_id, "doc-other", ["team-b"], request_context)
-        await _retain_doc(memory, bank_id, "doc-untagged", [], request_context)
-
-        result = await memory.list_documents(
-            bank_id=bank_id,
-            tags=["team-a"],
-            tags_match="any",
-            request_context=request_context,
-        )
-        ids = {d["id"] for d in result["items"]}
-        # "any" includes untagged + matching tagged
-        assert "doc-tagged" in ids
-        assert "doc-untagged" in ids
-        assert "doc-other" not in ids
-
-    finally:
-        await memory.delete_bank(bank_id, request_context=request_context)
-
-
-@pytest.mark.asyncio
-async def test_list_documents_tags_filter_all_strict(memory, request_context):
-    """tags filter with all_strict returns only docs that have ALL the specified tags."""
-    bank_id = f"test_list_docs_tags_all_{datetime.now(timezone.utc).timestamp()}"
-
-    try:
-        await _retain_doc(memory, bank_id, "doc-a-only", ["team-a"], request_context)
-        await _retain_doc(memory, bank_id, "doc-a-and-b", ["team-a", "team-b"], request_context)
-        await _retain_doc(memory, bank_id, "doc-untagged", [], request_context)
-
-        result = await memory.list_documents(
-            bank_id=bank_id,
-            tags=["team-a", "team-b"],
-            tags_match="all_strict",
-            request_context=request_context,
-        )
-        ids = {d["id"] for d in result["items"]}
-        assert ids == {"doc-a-and-b"}
-
-    finally:
-        await memory.delete_bank(bank_id, request_context=request_context)
-
-
-@pytest.mark.asyncio
-async def test_list_documents_no_tags_filter_returns_all(memory, request_context):
-    """When no tags filter is specified, all documents are returned."""
-    bank_id = f"test_list_docs_no_tags_{datetime.now(timezone.utc).timestamp()}"
-
-    try:
-        await _retain_doc(memory, bank_id, "doc-tagged", ["team-a"], request_context)
-        await _retain_doc(memory, bank_id, "doc-untagged", [], request_context)
-
-        result = await memory.list_documents(
-            bank_id=bank_id,
-            tags=None,
-            request_context=request_context,
-        )
-        ids = {d["id"] for d in result["items"]}
-        assert ids == {"doc-tagged", "doc-untagged"}
-
-    finally:
-        await memory.delete_bank(bank_id, request_context=request_context)
-
-
-@pytest.mark.asyncio
-async def test_list_documents_tags_and_search_query_combined(memory, request_context):
-    """tags filter and q (search_query) can be combined."""
-    bank_id = f"test_list_docs_tags_q_{datetime.now(timezone.utc).timestamp()}"
-
-    try:
-        await _retain_doc(memory, bank_id, "report-2024", ["team-a"], request_context)
-        await _retain_doc(memory, bank_id, "report-2025", ["team-b"], request_context)
-        await _retain_doc(memory, bank_id, "summary-2024", ["team-a"], request_context)
-
-        result = await memory.list_documents(
-            bank_id=bank_id,
-            search_query="report",
-            tags=["team-a"],
-            tags_match="any_strict",
-            request_context=request_context,
-        )
-        ids = {d["id"] for d in result["items"]}
-        assert ids == {"report-2024"}
-
-    finally:
-        await memory.delete_bank(bank_id, request_context=request_context)

hindsight-api-slim/tests/test_litellm_sdk_cross_encoder.py

@@ -1,392 +0,0 @@
-"""
-Tests for LiteLLMSDKCrossEncoder.
-
-Tests the LiteLLM SDK-based cross-encoder implementation for reranking.
-"""
-
-import os
-from unittest.mock import AsyncMock, MagicMock, patch
-
-import pytest
-
-from hindsight_api.engine.cross_encoder import LiteLLMSDKCrossEncoder, create_cross_encoder_from_env
-
-
-class TestLiteLLMSDKCrossEncoder:
-    """Test suite for LiteLLMSDKCrossEncoder class."""
-
-    @pytest.mark.asyncio
-    async def test_initialization_success(self):
-        """Test successful initialization with valid config."""
-        encoder = LiteLLMSDKCrossEncoder(
-            api_key="test_key",
-            model="deepinfra/Qwen3-reranker-8B",
-        )
-
-        assert encoder.provider_name == "litellm-sdk"
-        assert encoder.api_key == "test_key"
-        assert encoder.model == "deepinfra/Qwen3-reranker-8B"
-        assert encoder._initialized is False
-
-        # Mock the litellm import
-        mock_litellm = MagicMock()
-        with patch.dict("sys.modules", {"litellm": mock_litellm}):
-            await encoder.initialize()
-            assert encoder._initialized is True
-
-    @pytest.mark.asyncio
-    async def test_initialization_missing_package(self):
-        """Test initialization fails when litellm package is missing."""
-        encoder = LiteLLMSDKCrossEncoder(
-            api_key="test_key",
-            model="cohere/rerank-english-v3.0",
-        )
-
-        with patch.dict("sys.modules", {"litellm": None}):
-            with pytest.raises(ImportError, match="litellm is required"):
-                await encoder.initialize()
-
-    @pytest.mark.asyncio
-    async def test_initialization_idempotent(self):
-        """Test that calling initialize() multiple times is safe."""
-        encoder = LiteLLMSDKCrossEncoder(
-            api_key="test_key",
-            model="cohere/rerank-english-v3.0",
-        )
-
-        mock_litellm = MagicMock()
-        with patch.dict("sys.modules", {"litellm": mock_litellm}):
-            await encoder.initialize()
-            assert encoder._initialized is True
-
-            # Second call should be no-op
-            await encoder.initialize()
-            assert encoder._initialized is True
-
-    @pytest.mark.asyncio
-    async def test_predict_single_query(self):
-        """Test prediction with a single query and multiple documents."""
-        encoder = LiteLLMSDKCrossEncoder(
-            api_key="test_key",
-            model="deepinfra/Qwen3-reranker-8B",
-        )
-
-        # Create mock response with results as TypedDicts
-        mock_response = MagicMock()
-        mock_response.results = [
-            {"index": 0, "relevance_score": 0.9},
-            {"index": 1, "relevance_score": 0.7},
-            {"index": 2, "relevance_score": 0.5},
-        ]
-
-        mock_litellm = MagicMock()
-        mock_litellm.arerank = AsyncMock(return_value=mock_response)
-
-        with patch.dict("sys.modules", {"litellm": mock_litellm}):
-            await encoder.initialize()
-
-            pairs = [
-                ("What is Python?", "Python is a programming language"),
-                ("What is Python?", "Python is a snake"),
-                ("What is Python?", "Python is a British comedy group"),
-            ]
-
-            scores = await encoder.predict(pairs)
-
-            assert len(scores) == 3
-            assert scores == [0.9, 0.7, 0.5]
-
-            # Verify arerank was called correctly
-            mock_litellm.arerank.assert_called_once()
-            call_args = mock_litellm.arerank.call_args
-            assert call_args.kwargs["model"] == "deepinfra/Qwen3-reranker-8B"
-            assert call_args.kwargs["query"] == "What is Python?"
-            assert len(call_args.kwargs["documents"]) == 3
-            assert call_args.kwargs["api_key"] == "test_key"
-
-    @pytest.mark.asyncio
-    async def test_predict_multiple_queries(self):
-        """Test prediction with multiple different queries (grouped efficiently)."""
-        encoder = LiteLLMSDKCrossEncoder(
-            api_key="test_key",
-            model="cohere/rerank-english-v3.0",
-        )
-
-        # First query response
-        mock_response1 = MagicMock()
-        mock_response1.results = [
-            {"index": 0, "relevance_score": 0.9},
-            {"index": 1, "relevance_score": 0.7},
-        ]
-
-        # Second query response
-        mock_response2 = MagicMock()
-        mock_response2.results = [
-            {"index": 0, "relevance_score": 0.8},
-        ]
-
-        mock_litellm = MagicMock()
-        mock_litellm.arerank = AsyncMock(side_effect=[mock_response1, mock_response2])
-
-        with patch.dict("sys.modules", {"litellm": mock_litellm}):
-            await encoder.initialize()
-
-            pairs = [
-                ("What is Python?", "Python is a programming language"),
-                ("What is Python?", "Python is a snake"),
-                ("What is Java?", "Java is a programming language"),
-            ]
-
-            scores = await encoder.predict(pairs)
-
-            assert len(scores) == 3
-            assert scores[0] == 0.9  # First query, first doc
-            assert scores[1] == 0.7  # First query, second doc
-            assert scores[2] == 0.8  # Second query, first doc
-
-            # Verify arerank was called twice (once per unique query)
-            assert mock_litellm.arerank.call_count == 2
-
-    @pytest.mark.asyncio
-    async def test_predict_empty_pairs(self):
-        """Test prediction with empty input."""
-        encoder = LiteLLMSDKCrossEncoder(
-            api_key="test_key",
-            model="cohere/rerank-english-v3.0",
-        )
-
-        mock_litellm = MagicMock()
-        with patch.dict("sys.modules", {"litellm": mock_litellm}):
-            await encoder.initialize()
-            scores = await encoder.predict([])
-            assert scores == []
-
-    @pytest.mark.asyncio
-    async def test_predict_not_initialized(self):
-        """Test that predict fails if encoder not initialized."""
-        encoder = LiteLLMSDKCrossEncoder(
-            api_key="test_key",
-            model="cohere/rerank-english-v3.0",
-        )
-
-        pairs = [("query", "document")]
-
-        with pytest.raises(RuntimeError, match="not initialized"):
-            await encoder.predict(pairs)
-
-    @pytest.mark.asyncio
-    async def test_predict_error_handling(self):
-        """Test that errors during prediction are raised."""
-        encoder = LiteLLMSDKCrossEncoder(
-            api_key="test_key",
-            model="cohere/rerank-english-v3.0",
-        )
-
-        # Mock litellm to raise an error
-        mock_litellm = MagicMock()
-        mock_litellm.arerank = AsyncMock(side_effect=Exception("API Error"))
-
-        with patch.dict("sys.modules", {"litellm": mock_litellm}):
-            await encoder.initialize()
-
-            pairs = [
-                ("What is Python?", "Python is a programming language"),
-            ]
-
-            # Should raise the exception
-            with pytest.raises(Exception, match="API Error"):
-                await encoder.predict(pairs)
-
-    @pytest.mark.asyncio
-    async def test_custom_api_base(self):
-        """Test that custom API base URL is passed to rerank calls."""
-        encoder = LiteLLMSDKCrossEncoder(
-            api_key="test_key",
-            model="cohere/rerank-english-v3.0",
-            api_base="https://custom.api.example.com",
-        )
-
-        mock_response = MagicMock()
-        mock_response.results = [
-            {"index": 0, "relevance_score": 0.9},
-        ]
-
-        mock_litellm = MagicMock()
-        mock_litellm.arerank = AsyncMock(return_value=mock_response)
-
-        with patch.dict("sys.modules", {"litellm": mock_litellm}):
-            await encoder.initialize()
-
-            # Test that api_base is passed to arerank
-            pairs = [("query", "document")]
-            scores = await encoder.predict(pairs)
-
-            assert scores == [0.9]
-            mock_litellm.arerank.assert_called_once()
-            call_args = mock_litellm.arerank.call_args
-            assert call_args.kwargs["api_base"] == "https://custom.api.example.com"
-
-    @pytest.mark.asyncio
-    async def test_response_with_direct_score_list(self):
-        """Test handling of response format with direct score list."""
-        encoder = LiteLLMSDKCrossEncoder(
-            api_key="test_key",
-            model="some-provider/model",
-        )
-
-        # Mock litellm to return direct list of scores
-        mock_litellm = MagicMock()
-        mock_litellm.arerank = AsyncMock(return_value=[0.9, 0.7, 0.5])
-
-        with patch.dict("sys.modules", {"litellm": mock_litellm}):
-            await encoder.initialize()
-
-            pairs = [
-                ("query", "doc1"),
-                ("query", "doc2"),
-                ("query", "doc3"),
-            ]
-
-            scores = await encoder.predict(pairs)
-
-            assert scores == [0.9, 0.7, 0.5]
-
-
-class TestFactoryFunction:
-    """Test suite for create_cross_encoder_from_env factory function."""
-
-    @pytest.mark.asyncio
-    async def test_create_litellm_sdk_from_env(self):
-        """Test creating LiteLLM SDK cross-encoder from environment variables."""
-        env_vars = {
-            "HINDSIGHT_API_RERANKER_PROVIDER": "litellm-sdk",
-            "HINDSIGHT_API_RERANKER_LITELLM_SDK_API_KEY": "test_key",
-            "HINDSIGHT_API_RERANKER_LITELLM_SDK_MODEL": "deepinfra/Qwen3-reranker-8B",
-        }
-
-        with patch.dict(os.environ, env_vars, clear=False):
-            # Need to reload config to pick up env vars
-            from hindsight_api.config import HindsightConfig
-
-            config = HindsightConfig.from_env()
-
-            with patch("hindsight_api.config.get_config", return_value=config):
-                encoder = create_cross_encoder_from_env()
-
-                assert isinstance(encoder, LiteLLMSDKCrossEncoder)
-                assert encoder.api_key == "test_key"
-                assert encoder.model == "deepinfra/Qwen3-reranker-8B"
-
-    @pytest.mark.asyncio
-    async def test_create_litellm_sdk_missing_api_key(self):
-        """Test that factory raises error when API key is missing."""
-        env_vars = {
-            "HINDSIGHT_API_RERANKER_PROVIDER": "litellm-sdk",
-            "HINDSIGHT_API_RERANKER_LITELLM_SDK_MODEL": "deepinfra/Qwen3-reranker-8B",
-        }
-
-        with patch.dict(os.environ, env_vars, clear=False):
-            # Remove API key if set
-            if "HINDSIGHT_API_RERANKER_LITELLM_SDK_API_KEY" in os.environ:
-                del os.environ["HINDSIGHT_API_RERANKER_LITELLM_SDK_API_KEY"]
-
-            from hindsight_api.config import HindsightConfig
-
-            config = HindsightConfig.from_env()
-
-            with patch("hindsight_api.config.get_config", return_value=config):
-                with pytest.raises(ValueError, match="HINDSIGHT_API_RERANKER_LITELLM_SDK_API_KEY is required"):
-                    create_cross_encoder_from_env()
-
-    @pytest.mark.asyncio
-    async def test_create_litellm_sdk_with_custom_api_base(self):
-        """Test creating LiteLLM SDK cross-encoder with custom API base."""
-        env_vars = {
-            "HINDSIGHT_API_RERANKER_PROVIDER": "litellm-sdk",
-            "HINDSIGHT_API_RERANKER_LITELLM_SDK_API_KEY": "test_key",
-            "HINDSIGHT_API_RERANKER_LITELLM_SDK_MODEL": "cohere/rerank-english-v3.0",
-            "HINDSIGHT_API_RERANKER_LITELLM_SDK_API_BASE": "https://custom.api.example.com",
-        }
-
-        with patch.dict(os.environ, env_vars, clear=False):
-            from hindsight_api.config import HindsightConfig
-
-            config = HindsightConfig.from_env()
-
-            with patch("hindsight_api.config.get_config", return_value=config):
-                encoder = create_cross_encoder_from_env()
-
-                assert isinstance(encoder, LiteLLMSDKCrossEncoder)
-                assert encoder.api_base == "https://custom.api.example.com"
-
-
-class TestLiteLLMSDKCohereCrossEncoder:
-    """Tests for LiteLLM SDK calling Cohere (runs in CI with COHERE_API_KEY)."""
-
-    @pytest.fixture
-    async def litellm_cohere_cross_encoder(self):
-        """Create LiteLLM SDK cross-encoder instance for Cohere."""
-        if not os.environ.get("COHERE_API_KEY"):
-            pytest.skip("Cohere API key not available (set COHERE_API_KEY)")
-
-        encoder = LiteLLMSDKCrossEncoder(
-            api_key=os.environ["COHERE_API_KEY"],
-            model="cohere/rerank-english-v3.0",
-        )
-        await encoder.initialize()
-        return encoder
-
-    @pytest.mark.asyncio
-    async def test_litellm_sdk_cohere_initialization(self, litellm_cohere_cross_encoder):
-        """Test that LiteLLM SDK Cohere cross-encoder initializes correctly."""
-        assert litellm_cohere_cross_encoder.provider_name == "litellm-sdk"
-        assert litellm_cohere_cross_encoder.model == "cohere/rerank-english-v3.0"
-
-    @pytest.mark.asyncio
-    async def test_litellm_sdk_cohere_predict(self, litellm_cohere_cross_encoder):
-        """Test that LiteLLM SDK can call Cohere rerank API."""
-        pairs = [
-            ("What is the capital of France?", "Paris is the capital of France."),
-            ("What is the capital of France?", "The Eiffel Tower is in Paris."),
-            ("What is the capital of France?", "Python is a programming language."),
-        ]
-        scores = await litellm_cohere_cross_encoder.predict(pairs)
-
-        assert len(scores) == 3
-        assert all(isinstance(s, float) for s in scores)
-        # The first result should be most relevant
-        assert scores[0] > scores[2], "Direct answer should score higher than unrelated text"
-        # All scores should be in valid range
-        assert all(0.0 <= score <= 1.0 for score in scores)
-
-
-class TestIntegration:
-    """Integration tests with real API (optional - requires API keys)."""
-
-    @pytest.mark.skipif(
-        not os.environ.get("DEEPINFRA_API_KEY"),
-        reason="DEEPINFRA_API_KEY not set - skipping integration test",
-    )
-    @pytest.mark.asyncio
-    async def test_real_deepinfra_api(self):
-        """Test with real DeepInfra API (requires DEEPINFRA_API_KEY env var)."""
-        encoder = LiteLLMSDKCrossEncoder(
-            api_key=os.environ["DEEPINFRA_API_KEY"],
-            model="deepinfra/Qwen3-reranker-8B",
-        )
-
-        await encoder.initialize()
-
-        pairs = [
-            ("What is Python?", "Python is a high-level programming language"),
-            ("What is Python?", "Python is a species of snake"),
-            ("What is Python?", "Python is unrelated text about cars"),
-        ]
-
-        scores = await encoder.predict(pairs)
-
-        # First doc should have highest score (most relevant)
-        assert len(scores) == 3
-        assert scores[0] > scores[1]
-        assert scores[1] > scores[2]
-        assert all(0.0 <= score <= 1.0 for score in scores)

hindsight-api-slim/tests/test_litellm_sdk_embeddings.py

@@ -1,389 +0,0 @@
-"""
-Tests for LiteLLM SDK embeddings implementation.
-
-These tests cover:
-1. Initialization (success, missing package, missing API key, idempotent)
-2. Encode (single text, multiple texts, batching, error handling)
-3. Provider-specific configuration (Cohere, OpenAI, etc.)
-4. Factory function (create from env, validation errors)
-5. Dimension detection
-"""
-
-import os
-from unittest.mock import AsyncMock, MagicMock, patch
-
-import pytest
-
-from hindsight_api.config import (
-    ENV_EMBEDDINGS_LITELLM_SDK_API_KEY,
-    ENV_EMBEDDINGS_LITELLM_SDK_MODEL,
-    ENV_EMBEDDINGS_PROVIDER,
-    HindsightConfig,
-)
-from hindsight_api.engine.embeddings import LiteLLMSDKEmbeddings, create_embeddings_from_env
-
-
-class TestLiteLLMSDKEmbeddings:
-    """Unit tests for LiteLLMSDKEmbeddings with mocked litellm responses."""
-
-    @pytest.fixture
-    def mock_litellm(self):
-        """Mock litellm module."""
-        mock = MagicMock()
-
-        # Mock aembedding (async) for initialization
-        mock_response = MagicMock()
-        mock_response.data = [{"embedding": [0.1] * 768, "index": 0}]
-        mock.aembedding = AsyncMock(return_value=mock_response)
-
-        # Mock embedding (sync) for encode
-        mock_sync_response = MagicMock()
-        mock_sync_response.data = [
-            {"embedding": [0.1] * 768, "index": 0},
-            {"embedding": [0.2] * 768, "index": 1},
-        ]
-        mock.embedding = MagicMock(return_value=mock_sync_response)
-
-        return mock
-
-    @pytest.fixture
-    async def embeddings(self, mock_litellm):
-        """Create initialized LiteLLMSDKEmbeddings instance."""
-        emb = LiteLLMSDKEmbeddings(
-            api_key="test_key",
-            model="cohere/embed-english-v3.0",
-            api_base=None,
-            batch_size=100,
-            timeout=60.0,
-        )
-        # Manually set the mock (simulating successful initialization)
-        emb._litellm = mock_litellm
-        emb._dimension = 768
-        return emb
-
-    async def test_initialization_success(self, mock_litellm):
-        """Test successful initialization."""
-        with patch("builtins.__import__", side_effect=lambda name, *args: mock_litellm if name == "litellm" else __import__(name, *args)):
-            emb = LiteLLMSDKEmbeddings(
-                api_key="test_key",
-                model="cohere/embed-english-v3.0",
-                api_base=None,
-                batch_size=100,
-                timeout=60.0,
-            )
-
-            assert emb._litellm is None
-            assert emb._dimension is None
-
-            await emb.initialize()
-
-            assert emb._litellm is not None
-            assert emb._dimension == 768
-
-            # Verify test embedding was called
-            mock_litellm.aembedding.assert_called_once_with(
-                model="cohere/embed-english-v3.0",
-                input=["test"],
-                api_key="test_key",
-                encoding_format="float",
-            )
-
-    async def test_initialization_missing_package(self):
-        """Test initialization fails gracefully when litellm is not installed."""
-        def mock_import(name, *args):
-            if name == "litellm":
-                raise ImportError("No module named 'litellm'")
-            return __import__(name, *args)
-
-        with patch("builtins.__import__", side_effect=mock_import):
-            emb = LiteLLMSDKEmbeddings(
-                api_key="test_key",
-                model="cohere/embed-english-v3.0",
-                api_base=None,
-                batch_size=100,
-                timeout=60.0,
-            )
-
-            with pytest.raises(ImportError, match="litellm is required"):
-                await emb.initialize()
-
-    async def test_initialization_idempotent(self, embeddings, mock_litellm):
-        """Test that calling initialize() multiple times is safe."""
-        # embeddings._litellm is already set in fixture
-        assert embeddings._litellm is not None
-
-        # Call again
-        await embeddings.initialize()
-
-        # Should still have same litellm instance
-        assert embeddings._litellm is not None
-
-    async def test_encode_single_text(self, embeddings, mock_litellm):
-        """Test encoding a single text."""
-        # Set up mock response
-        mock_litellm.embedding.return_value.data = [
-            {"embedding": [0.5] * 768, "index": 0},
-        ]
-
-        result = embeddings.encode(["Hello world"])
-
-        assert isinstance(result, list)
-        assert len(result) == 1
-        assert len(result[0]) == 768
-        assert all(isinstance(x, float) for x in result[0])
-        assert all(abs(x - 0.5) < 0.001 for x in result[0])
-
-        # Verify call
-        mock_litellm.embedding.assert_called_once_with(
-            model="cohere/embed-english-v3.0",
-            input=["Hello world"],
-            api_key="test_key",
-            encoding_format="float",
-        )
-
-    async def test_encode_multiple_texts(self, embeddings, mock_litellm):
-        """Test encoding multiple texts."""
-        # Set up mock response
-        mock_litellm.embedding.return_value.data = [
-            {"embedding": [0.1] * 768, "index": 0},
-            {"embedding": [0.2] * 768, "index": 1},
-            {"embedding": [0.3] * 768, "index": 2},
-        ]
-
-        texts = ["First text", "Second text", "Third text"]
-        result = embeddings.encode(texts)
-
-        assert isinstance(result, list)
-        assert len(result) == 3
-        assert len(result[0]) == 768
-        assert len(result[1]) == 768
-        assert len(result[2]) == 768
-        assert all(abs(x - 0.1) < 0.001 for x in result[0])
-        assert all(abs(x - 0.2) < 0.001 for x in result[1])
-        assert all(abs(x - 0.3) < 0.001 for x in result[2])
-
-    async def test_encode_batching(self, embeddings, mock_litellm):
-        """Test that large inputs are batched correctly."""
-        # Create embeddings with small batch size
-        emb = LiteLLMSDKEmbeddings(
-            api_key="test_key",
-            model="cohere/embed-english-v3.0",
-            api_base=None,
-            batch_size=2,  # Small batch for testing
-            timeout=60.0,
-        )
-        emb._litellm = mock_litellm
-        emb._initialized = True
-        emb._dimension = 768
-
-        # Mock responses for each batch
-        def mock_embedding_side_effect(model, input, **kwargs):
-            mock_response = MagicMock()
-            mock_response.data = [
-                {"embedding": [float(i)] * 768, "index": i} for i in range(len(input))
-            ]
-            return mock_response
-
-        mock_litellm.embedding.side_effect = mock_embedding_side_effect
-
-        # Encode 5 texts (should create 3 batches: 2, 2, 1)
-        texts = [f"Text {i}" for i in range(5)]
-        result = emb.encode(texts)
-
-        assert isinstance(result, list)
-        assert len(result) == 5
-        assert all(len(embedding) == 768 for embedding in result)
-
-        # Verify batching: should be called 3 times
-        assert mock_litellm.embedding.call_count == 3
-
-        # Verify batch sizes
-        calls = mock_litellm.embedding.call_args_list
-        assert len(calls[0][1]["input"]) == 2  # First batch
-        assert len(calls[1][1]["input"]) == 2  # Second batch
-        assert len(calls[2][1]["input"]) == 1  # Third batch
-
-    async def test_encode_empty_list(self, embeddings):
-        """Test encoding empty list returns empty list."""
-        result = embeddings.encode([])
-
-        assert isinstance(result, list)
-        assert len(result) == 0
-
-    async def test_encode_before_initialization(self, mock_litellm):
-        """Test that encode raises error if not initialized."""
-        emb = LiteLLMSDKEmbeddings(
-            api_key="test_key",
-            model="cohere/embed-english-v3.0",
-            api_base=None,
-            batch_size=100,
-            timeout=60.0,
-        )
-
-        with pytest.raises(RuntimeError, match="not initialized"):
-            emb.encode(["test"])
-
-    async def test_encode_error_handling(self, embeddings, mock_litellm):
-        """Test error handling during encoding."""
-        # Make embedding raise an error
-        mock_litellm.embedding.side_effect = Exception("API Error")
-
-        with pytest.raises(Exception, match="API Error"):
-            embeddings.encode(["test"])
-
-    async def test_dimension_property(self, embeddings):
-        """Test dimension property."""
-        assert embeddings.dimension == 768
-
-    async def test_dimension_before_initialization(self, mock_litellm):
-        """Test dimension raises error if not initialized."""
-        emb = LiteLLMSDKEmbeddings(
-            api_key="test_key",
-            model="cohere/embed-english-v3.0",
-            api_base=None,
-            batch_size=100,
-            timeout=60.0,
-        )
-
-        with pytest.raises(RuntimeError, match="not initialized"):
-            _ = emb.dimension
-
-    async def test_custom_api_base(self, mock_litellm):
-        """Test custom API base URL is passed to embedding calls."""
-        with patch("builtins.__import__", side_effect=lambda name, *args: mock_litellm if name == "litellm" else __import__(name, *args)):
-            emb = LiteLLMSDKEmbeddings(
-                api_key="test_key",
-                model="cohere/embed-english-v3.0",
-                api_base="https://custom.api.com",
-                batch_size=100,
-                timeout=60.0,
-            )
-
-            await emb.initialize()
-
-            # Verify api_base is set
-            assert emb.api_base == "https://custom.api.com"
-
-            # Verify api_base is passed to aembedding
-            mock_litellm.aembedding.assert_called_once()
-            call_args = mock_litellm.aembedding.call_args
-            assert call_args.kwargs["api_base"] == "https://custom.api.com"
-
-            # Test encode also passes api_base
-            mock_litellm.embedding.return_value.data = [{"embedding": [0.1] * 768, "index": 0}]
-            emb.encode(["test"])
-
-            mock_litellm.embedding.assert_called_once()
-            call_args = mock_litellm.embedding.call_args
-            assert call_args.kwargs["api_base"] == "https://custom.api.com"
-
-
-class TestLiteLLMSDKEmbeddingsFactory:
-    """Test the factory function for creating LiteLLM SDK embeddings."""
-
-    def test_create_from_env_success(self, monkeypatch):
-        """Test creating embeddings from environment variables."""
-        # Mock get_config() to return configured HindsightConfig
-        mock_config = MagicMock()
-        mock_config.embeddings_provider = "litellm-sdk"
-        mock_config.embeddings_litellm_sdk_api_key = "test_key"
-        mock_config.embeddings_litellm_sdk_model = "cohere/embed-english-v3.0"
-        mock_config.embeddings_litellm_sdk_api_base = None
-
-        with patch("hindsight_api.config.get_config", return_value=mock_config):
-            embeddings = create_embeddings_from_env()
-
-            assert isinstance(embeddings, LiteLLMSDKEmbeddings)
-            assert embeddings.api_key == "test_key"
-            assert embeddings.model == "cohere/embed-english-v3.0"
-
-    def test_create_from_env_missing_api_key(self, monkeypatch):
-        """Test that missing API key raises error."""
-        # Mock get_config() with missing API key
-        mock_config = MagicMock()
-        mock_config.embeddings_provider = "litellm-sdk"
-        mock_config.embeddings_litellm_sdk_api_key = None  # Missing key
-        mock_config.embeddings_litellm_sdk_model = "cohere/embed-english-v3.0"
-
-        with patch("hindsight_api.config.get_config", return_value=mock_config):
-            with pytest.raises(ValueError, match=ENV_EMBEDDINGS_LITELLM_SDK_API_KEY):
-                create_embeddings_from_env()
-
-    def test_create_from_env_with_api_base(self, monkeypatch):
-        """Test creating embeddings with custom API base."""
-        # Mock get_config() with custom API base
-        mock_config = MagicMock()
-        mock_config.embeddings_provider = "litellm-sdk"
-        mock_config.embeddings_litellm_sdk_api_key = "test_key"
-        mock_config.embeddings_litellm_sdk_model = "cohere/embed-english-v3.0"
-        mock_config.embeddings_litellm_sdk_api_base = "https://custom.api.com"
-
-        with patch("hindsight_api.config.get_config", return_value=mock_config):
-            embeddings = create_embeddings_from_env()
-
-            assert isinstance(embeddings, LiteLLMSDKEmbeddings)
-            assert embeddings.api_base == "https://custom.api.com"
-
-
-class TestLiteLLMSDKCohereEmbeddings:
-    """Integration tests calling real Cohere API (matches CI pattern)."""
-
-    @pytest.fixture
-    async def litellm_cohere_embeddings(self):
-        """Create embeddings instance with real Cohere API key."""
-        if not os.environ.get("COHERE_API_KEY"):
-            pytest.skip("Cohere API key not available")
-
-        emb = LiteLLMSDKEmbeddings(
-            api_key=os.environ["COHERE_API_KEY"],
-            model="cohere/embed-english-v3.0",
-            api_base=None,
-            batch_size=100,
-            timeout=60.0,
-        )
-        await emb.initialize()
-        return emb
-
-    @pytest.mark.asyncio
-    async def test_litellm_sdk_cohere_encode(self, litellm_cohere_embeddings):
-        """Test real Cohere API call for embeddings."""
-        texts = [
-            "The quick brown fox jumps over the lazy dog",
-            "Machine learning is a subset of artificial intelligence",
-            "Python is a popular programming language",
-        ]
-
-        result = litellm_cohere_embeddings.encode(texts)
-
-        # Verify result type and shape
-        assert isinstance(result, list)
-        assert len(result) == 3
-        assert all(len(embedding) > 0 for embedding in result)
-        assert all(isinstance(x, float) for x in result[0])
-
-        # Verify embeddings are not zeros (common API failure mode)
-        for i, embedding in enumerate(result):
-            assert not all(abs(x) < 0.0001 for x in embedding), f"Embedding {i} is all zeros"
-
-        # Verify embeddings are normalized (Cohere returns normalized vectors)
-        for i, embedding in enumerate(result):
-            norm = sum(x * x for x in embedding) ** 0.5
-            assert 0.9 < norm < 1.1, f"Embedding {i} norm {norm} is not close to 1.0"
-
-    @pytest.mark.asyncio
-    async def test_litellm_sdk_cohere_dimension(self, litellm_cohere_embeddings):
-        """Test dimension detection with real Cohere API."""
-        dimension = litellm_cohere_embeddings.dimension
-
-        # Cohere embed-english-v3.0 has 1024 dimensions
-        assert dimension == 1024
-
-    @pytest.mark.asyncio
-    async def test_litellm_sdk_cohere_single_text(self, litellm_cohere_embeddings):
-        """Test encoding single text with real Cohere API."""
-        result = litellm_cohere_embeddings.encode(["Hello world"])
-
-        assert isinstance(result, list)
-        assert len(result) == 1
-        assert len(result[0]) == 1024
-        assert not all(abs(x) < 0.0001 for x in result[0])

hindsight-api-slim/tests/test_llm_wrapper.py

@@ -1,37 +0,0 @@
-import pytest
-
-from hindsight_api.engine.llm_wrapper import sanitize_llm_output
-
-
-@pytest.mark.parametrize(
-    "input_text, expected",
-    [
-        # Null bytes stripped
-        ("hello\x00world", "helloworld"),
-        ("FIRST\u0000PAGE", "FIRSTPAGE"),
-        # Multiple null bytes
-        ("\x00\x00text\x00", "text"),
-        # Other control characters stripped (non-whitespace)
-        ("text\x01\x02\x03end", "textend"),
-        ("text\x08end", "textend"),  # backspace
-        ("text\x0cend", "textend"),  # form feed
-        ("text\x0bend", "textend"),  # vertical tab
-        ("text\x1fend", "textend"),  # unit separator
-        ("text\x7fend", "textend"),  # DEL
-        # Whitespace preserved
-        ("hello\tworld", "hello\tworld"),
-        ("hello\nworld", "hello\nworld"),
-        ("hello\r\nworld", "hello\r\nworld"),
-        # Unicode surrogates stripped
-        ("text\ud800end", "textend"),
-        ("text\udfffend", "textend"),
-        # Clean text unchanged
-        ("normal text", "normal text"),
-        ("unicode: café naïve", "unicode: café naïve"),
-        # Edge cases
-        ("", ""),
-        (None, None),
-    ],
-)
-def test_sanitize_llm_output(input_text, expected):
-    assert sanitize_llm_output(input_text) == expected

hindsight-api-slim/tests/test_lmstudio_tool_choice.py

@@ -1,321 +0,0 @@
-"""
-Reproduce issue #520: Reflect fails with LM Studio due to unsupported tool_choice format.
-
-The reflect agent forces tool selection via named tool_choice dicts on the first few iterations:
-  {"type": "function", "function": {"name": "search_mental_models"}}
-
-LM Studio (and Ollama) reject this format with HTTP 400:
-  "Tool choice of type 'function' is not supported. Use 'auto', 'none', or 'required'."
-
-The fix should convert named tool_choice to "required" and filter the tools list
-to only the requested tool for providers that don't support named tool_choice.
-"""
-
-import json
-from unittest.mock import AsyncMock, MagicMock, patch
-
-import pytest
-from openai import APIStatusError
-
-from hindsight_api.engine.providers.openai_compatible_llm import OpenAICompatibleLLM
-
-# Reflect agent tools (subset matching what agent.py uses)
-REFLECT_TOOLS = [
-    {
-        "type": "function",
-        "function": {
-            "name": "search_mental_models",
-            "description": "Search consolidated mental models",
-            "parameters": {
-                "type": "object",
-                "properties": {"query": {"type": "string"}},
-                "required": ["query"],
-            },
-        },
-    },
-    {
-        "type": "function",
-        "function": {
-            "name": "search_observations",
-            "description": "Search raw observations",
-            "parameters": {
-                "type": "object",
-                "properties": {"query": {"type": "string"}},
-                "required": ["query"],
-            },
-        },
-    },
-    {
-        "type": "function",
-        "function": {
-            "name": "recall",
-            "description": "Recall semantic memories",
-            "parameters": {
-                "type": "object",
-                "properties": {"query": {"type": "string"}},
-                "required": ["query"],
-            },
-        },
-    },
-    {
-        "type": "function",
-        "function": {
-            "name": "done",
-            "description": "Finish and return the answer",
-            "parameters": {
-                "type": "object",
-                "properties": {"answer": {"type": "string"}},
-                "required": ["answer"],
-            },
-        },
-    },
-]
-
-
-def _make_lmstudio_llm() -> OpenAICompatibleLLM:
-    return OpenAICompatibleLLM(
-        provider="lmstudio",
-        api_key="local",
-        base_url="http://localhost:1234/v1",
-        model="openai/gpt-oss-20b",
-    )
-
-
-def _lmstudio_400_error(msg: str = "Tool choice of type 'function' is not supported. Use 'auto', 'none', or 'required'.") -> APIStatusError:
-    """Simulate the HTTP 400 LM Studio returns for unsupported tool_choice format."""
-    mock_response = MagicMock()
-    mock_response.status_code = 400
-    mock_response.headers = {}
-    return APIStatusError(
-        message=msg,
-        response=mock_response,
-        body={"error": {"message": msg, "type": "invalid_request_error"}},
-    )
-
-
-def _make_tool_call_response(tool_name: str, arguments: dict) -> MagicMock:
-    """Build a mock successful tool call response from the LLM API."""
-    mock_tc = MagicMock()
-    mock_tc.id = "call_abc123"
-    mock_tc.function.name = tool_name
-    mock_tc.function.arguments = json.dumps(arguments)
-
-    mock_response = MagicMock()
-    mock_response.usage.prompt_tokens = 120
-    mock_response.usage.completion_tokens = 40
-    mock_response.usage.total_tokens = 160
-    mock_response.choices[0].finish_reason = "tool_calls"
-    mock_response.choices[0].message.content = None
-    mock_response.choices[0].message.tool_calls = [mock_tc]
-    return mock_response
-
-
-class TestLMStudioNamedToolChoiceBug:
-    """
-    Reproduces issue #520.
-
-    The reflect agent (agent.py lines 546-555) sets tool_choice to a named dict
-    on the first iterations to force sequential retrieval:
-
-        iteration=0, has_mental_models=True  → {"type": "function", "function": {"name": "search_mental_models"}}
-        iteration=0, has_mental_models=False → {"type": "function", "function": {"name": "search_observations"}}
-        iteration=1, has_mental_models=True  → {"type": "function", "function": {"name": "search_observations"}}
-        iteration=1 or (2 with models)       → {"type": "function", "function": {"name": "recall"}}
-
-    LM Studio rejects these dict formats with HTTP 400.
-    """
-
-    @pytest.mark.asyncio
-    async def test_lmstudio_named_tool_choice_no_longer_causes_400(self):
-        """
-        Regression test for issue #520: named tool_choice dict is converted to
-        "required" + filtered tools before the API call, so LM Studio never
-        sees the unsupported format and the 400 error no longer occurs.
-        """
-        llm = _make_lmstudio_llm()
-        named_tool_choice = {"type": "function", "function": {"name": "search_mental_models"}}
-        success_response = _make_tool_call_response("search_mental_models", {"query": "user name"})
-
-        with patch.object(llm._client.chat.completions, "create", new_callable=AsyncMock) as mock_create:
-            mock_create.return_value = success_response
-
-            # Should succeed — no 400 because the dict is converted before sending
-            result = await llm.call_with_tools(
-                messages=[{"role": "user", "content": "What is the user's name?"}],
-                tools=REFLECT_TOOLS,
-                tool_choice=named_tool_choice,
-                max_retries=0,
-            )
-
-        assert len(result.tool_calls) == 1
-        assert result.tool_calls[0].name == "search_mental_models"
-
-        sent_kwargs = mock_create.call_args.kwargs
-        assert sent_kwargs["tool_choice"] == "required"
-        assert len(sent_kwargs["tools"]) == 1
-        assert sent_kwargs["tools"][0]["function"]["name"] == "search_mental_models"
-
-    @pytest.mark.asyncio
-    @pytest.mark.parametrize(
-        "forced_tool_name",
-        ["search_mental_models", "search_observations", "recall"],
-    )
-    async def test_all_reflect_forced_tools_fail_on_lmstudio(self, forced_tool_name: str):
-        """
-        Each named tool_choice the reflect agent uses on iterations 0-2 triggers
-        the same 400 error on LM Studio.
-        """
-        llm = _make_lmstudio_llm()
-        named_tool_choice = {"type": "function", "function": {"name": forced_tool_name}}
-
-        with patch.object(llm._client.chat.completions, "create", new_callable=AsyncMock) as mock_create:
-            mock_create.side_effect = _lmstudio_400_error()
-
-            with pytest.raises(APIStatusError) as exc_info:
-                await llm.call_with_tools(
-                    messages=[{"role": "user", "content": "Test query"}],
-                    tools=REFLECT_TOOLS,
-                    tool_choice=named_tool_choice,
-                    max_retries=0,
-                )
-
-            assert exc_info.value.status_code == 400
-
-    @pytest.mark.asyncio
-    async def test_lmstudio_string_tool_choice_works_fine(self):
-        """
-        String tool_choice values ("auto", "none", "required") ARE supported by LM Studio.
-        Only the dict format {"type": "function", "function": {"name": "..."}} fails.
-        This test confirms the control case works.
-        """
-        llm = _make_lmstudio_llm()
-        success_response = _make_tool_call_response("search_mental_models", {"query": "user name"})
-
-        with patch.object(llm._client.chat.completions, "create", new_callable=AsyncMock) as mock_create:
-            mock_create.return_value = success_response
-
-            result = await llm.call_with_tools(
-                messages=[{"role": "user", "content": "What is the user's name?"}],
-                tools=REFLECT_TOOLS,
-                tool_choice="required",  # string form — LM Studio accepts this
-                max_retries=0,
-            )
-
-        assert len(result.tool_calls) == 1
-        assert result.tool_calls[0].name == "search_mental_models"
-
-        # Confirm "required" was sent, not a dict
-        sent_kwargs = mock_create.call_args.kwargs
-        assert sent_kwargs["tool_choice"] == "required"
-
-
-class TestExpectedFixBehavior:
-    """
-    Tests that document the EXPECTED behavior after the fix is applied.
-
-    For lmstudio (and ollama) providers, when tool_choice is a named dict:
-      {"type": "function", "function": {"name": "search_mental_models"}}
-
-    The fix should:
-      1. Convert tool_choice to "required"
-      2. Filter tools to only the requested tool
-
-    These tests currently FAIL (because the fix is not yet implemented).
-    After the fix is applied, they should PASS.
-    """
-
-    @pytest.mark.asyncio
-    async def test_fix_converts_named_tool_choice_to_required(self):
-        """
-        After fix: named tool_choice dict is converted to "required" for lmstudio.
-        The API receives tool_choice="required" instead of the unsupported dict.
-        """
-        llm = _make_lmstudio_llm()
-        named_tool_choice = {"type": "function", "function": {"name": "search_mental_models"}}
-        success_response = _make_tool_call_response("search_mental_models", {"query": "user name"})
-
-        with patch.object(llm._client.chat.completions, "create", new_callable=AsyncMock) as mock_create:
-            mock_create.return_value = success_response
-
-            result = await llm.call_with_tools(
-                messages=[{"role": "user", "content": "What is the user's name?"}],
-                tools=REFLECT_TOOLS,
-                tool_choice=named_tool_choice,
-                max_retries=0,
-            )
-
-        assert len(result.tool_calls) == 1
-        assert result.tool_calls[0].name == "search_mental_models"
-
-        sent_kwargs = mock_create.call_args.kwargs
-        # Fix: dict was converted to "required"
-        assert sent_kwargs["tool_choice"] == "required", (
-            f"Expected tool_choice='required', got {sent_kwargs['tool_choice']!r}"
-        )
-        # Fix: tools filtered to just the requested one
-        assert len(sent_kwargs["tools"]) == 1
-        assert sent_kwargs["tools"][0]["function"]["name"] == "search_mental_models"
-
-    @pytest.mark.asyncio
-    @pytest.mark.parametrize(
-        "forced_tool_name",
-        ["search_mental_models", "search_observations", "recall"],
-    )
-    async def test_fix_filters_tools_to_requested_tool(self, forced_tool_name: str):
-        """
-        After fix: tools list is filtered to only the forced tool so the model
-        can only call that one tool (equivalent to the named tool_choice behavior).
-        """
-        llm = _make_lmstudio_llm()
-        named_tool_choice = {"type": "function", "function": {"name": forced_tool_name}}
-        success_response = _make_tool_call_response(forced_tool_name, {"query": "test"})
-
-        with patch.object(llm._client.chat.completions, "create", new_callable=AsyncMock) as mock_create:
-            mock_create.return_value = success_response
-
-            await llm.call_with_tools(
-                messages=[{"role": "user", "content": "Test query"}],
-                tools=REFLECT_TOOLS,
-                tool_choice=named_tool_choice,
-                max_retries=0,
-            )
-
-        sent_kwargs = mock_create.call_args.kwargs
-        assert sent_kwargs["tool_choice"] == "required"
-        assert len(sent_kwargs["tools"]) == 1
-        assert sent_kwargs["tools"][0]["function"]["name"] == forced_tool_name
-
-    @pytest.mark.asyncio
-    async def test_fix_also_applies_to_openai_provider(self):
-        """
-        The fix is generalized: all providers convert named tool_choice to
-        "required" + filtered tools.  OpenAI natively supports the dict format
-        too, so the behaviour is semantically identical either way.
-        """
-        from hindsight_api.engine.providers.openai_compatible_llm import OpenAICompatibleLLM
-
-        openai_llm = OpenAICompatibleLLM(
-            provider="openai",
-            api_key="sk-test",
-            base_url="",
-            model="gpt-4o-mini",
-        )
-
-        named_tool_choice = {"type": "function", "function": {"name": "search_mental_models"}}
-        success_response = _make_tool_call_response("search_mental_models", {"query": "test"})
-
-        with patch.object(openai_llm._client.chat.completions, "create", new_callable=AsyncMock) as mock_create:
-            mock_create.return_value = success_response
-
-            await openai_llm.call_with_tools(
-                messages=[{"role": "user", "content": "Test"}],
-                tools=REFLECT_TOOLS,
-                tool_choice=named_tool_choice,
-                max_retries=0,
-            )
-
-        sent_kwargs = mock_create.call_args.kwargs
-        # Generalized fix applies to OpenAI too
-        assert sent_kwargs["tool_choice"] == "required"
-        assert len(sent_kwargs["tools"]) == 1
-        assert sent_kwargs["tools"][0]["function"]["name"] == "search_mental_models"