fix: update tests for duplicate document_id validation

- test_async_retain_tags: verify operation structure instead of exact UUID
- test_delete_bank: use unique document_ids (team-doc-1, team-doc-2)

.env.example

@@ -2,14 +2,72 @@
 # Copy this file to .env and fill in your values
 
 # LLM Configuration (Required)
+# 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
+# HINDSIGHT_API_LLM_PROVIDER=anthropic
+# HINDSIGHT_API_LLM_API_KEY=your-anthropic-api-key
+# HINDSIGHT_API_LLM_MODEL=claude-sonnet-4-20250514
+
+# Example: Google Vertex AI configuration
+# HINDSIGHT_API_LLM_PROVIDER=vertexai
+# HINDSIGHT_API_LLM_MODEL=google/gemini-2.0-flash-001
+# HINDSIGHT_API_LLM_VERTEXAI_PROJECT_ID=your-gcp-project-id
+# 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: LM Studio local configuration (Qwen 2.5 32B recommended)
+# HINDSIGHT_API_LLM_PROVIDER=lmstudio
+# HINDSIGHT_API_LLM_API_KEY=lmstudio
+# HINDSIGHT_API_LLM_BASE_URL=http://localhost:1234/v1
+# HINDSIGHT_API_LLM_MODEL=qwen2.5-32b-instruct
+
 # API Configuration (Optional)
 HINDSIGHT_API_HOST=0.0.0.0
 HINDSIGHT_API_PORT=8888
 HINDSIGHT_API_LOG_LEVEL=info
 
+# Base Path / Reverse Proxy Support (Optional)
+# Set these when deploying behind a reverse proxy with path-based routing
+# Example: To deploy at example.com/hindsight/, set both to "/hindsight"
+# HINDSIGHT_API_BASE_PATH=/hindsight
+# NEXT_PUBLIC_BASE_PATH=/hindsight
+
 # Database (Optional - uses embedded pg0 by default)
 # HINDSIGHT_API_DATABASE_URL=postgresql://user:pass@host:5432/db
+# HINDSIGHT_API_DATABASE_SCHEMA=public  # PostgreSQL schema name (default: public)
+
+# Embeddings Configuration (Optional - uses local by default)
+# Provider: "local" (default) or "tei" (HuggingFace Text Embeddings Inference)
+# HINDSIGHT_API_EMBEDDINGS_PROVIDER=local
+# For local provider:
+# HINDSIGHT_API_EMBEDDINGS_LOCAL_MODEL=BAAI/bge-small-en-v1.5
+# For TEI provider:
+# HINDSIGHT_API_EMBEDDINGS_TEI_URL=http://localhost:8080
+
+# Reranker Configuration (Optional - uses local by default)
+# Provider: "local" (default) or "tei" (HuggingFace Text Embeddings Inference)
+# HINDSIGHT_API_RERANKER_PROVIDER=local
+# For local provider:
+# HINDSIGHT_API_RERANKER_LOCAL_MODEL=cross-encoder/ms-marco-MiniLM-L-6-v2
+# For TEI provider:
+# HINDSIGHT_API_RERANKER_TEI_URL=http://localhost:8081
+
+# Observability & Tracing (Optional - disabled by default)
+# Enable OpenTelemetry tracing for LLM calls (GenAI semantic conventions)
+# HINDSIGHT_API_OTEL_TRACES_ENABLED=true
+#
+# Local development with Grafana LGTM stack (recommended - see scripts/dev/grafana/README.md)
+# HINDSIGHT_API_OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:4318
+#
+# Cloud backends (Grafana Cloud, Langfuse, DataDog, etc.)
+# HINDSIGHT_API_OTEL_EXPORTER_OTLP_ENDPOINT=https://your-backend-url
+# HINDSIGHT_API_OTEL_EXPORTER_OTLP_HEADERS="Authorization=Bearer your-token"
+#
+# Custom service name and environment (optional, defaults: hindsight-api, development)
+# HINDSIGHT_API_OTEL_SERVICE_NAME=hindsight-production
+# HINDSIGHT_API_OTEL_DEPLOYMENT_ENVIRONMENT=production

.githooks/pre-commit

@@ -0,0 +1,27 @@
+#!/bin/bash
+# Pre-commit hook - runs all scripts in scripts/hooks/
+
+set -e
+
+REPO_ROOT="$(git rev-parse --show-toplevel)"
+HOOKS_DIR="$REPO_ROOT/scripts/hooks"
+
+if [ ! -d "$HOOKS_DIR" ]; then
+    exit 0
+fi
+
+echo ""
+echo "=== Running pre-commit hooks ==="
+echo ""
+
+# Run all executable scripts in hooks directory
+for hook in "$HOOKS_DIR"/*.sh; do
+    if [ -x "$hook" ]; then
+        echo "[hook] $(basename "$hook")"
+        (cd "$REPO_ROOT" && "$hook")
+    fi
+done
+
+echo ""
+echo "=== Pre-commit hooks completed ==="
+echo ""

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -0,0 +1,71 @@
+name: Bug Report
+description: Report a bug or unexpected behavior
+labels: ["bug", "triage"]
+body:
+  - type: markdown
+    attributes:
+      value: |
+        Thanks for taking the time to report a bug! Please fill out the sections below.
+
+  - type: textarea
+    id: description
+    attributes:
+      label: Bug Description
+      description: A clear and concise description of the bug
+      placeholder: What happened?
+    validations:
+      required: true
+
+  - type: textarea
+    id: reproduction
+    attributes:
+      label: Steps to Reproduce
+      description: Steps to reproduce the behavior
+      placeholder: |
+        1. Configure '...'
+        2. Call '...'
+        3. See error
+    validations:
+      required: true
+
+  - type: textarea
+    id: expected
+    attributes:
+      label: Expected Behavior
+      description: What did you expect to happen?
+    validations:
+      required: true
+
+  - type: textarea
+    id: actual
+    attributes:
+      label: Actual Behavior
+      description: What actually happened?
+    validations:
+      required: true
+
+  - type: input
+    id: version
+    attributes:
+      label: Version
+      description: What version are you using?
+      placeholder: e.g., 0.1.0 or commit hash
+    validations:
+      required: false
+
+  - type: dropdown
+    id: llm-provider
+    attributes:
+      label: LLM Provider
+      description: Which LLM provider are you using?
+      options:
+        - OpenAI
+        - Anthropic
+        - Gemini
+        - Groq
+        - Ollama
+        - LM Studio
+        - Other
+    validations:
+      required: false
+

.github/ISSUE_TEMPLATE/config.yml

@@ -0,0 +1,8 @@
+blank_issues_enabled: false
+contact_links:
+  - name: Questions & Help
+    url: https://github.com/vectorize-io/hindsight/discussions/categories/q-a
+    about: Please ask questions and get help in Discussions instead of opening an issue.
+  - name: Ideas & Feedback
+    url: https://github.com/vectorize-io/hindsight/discussions/categories/ideas
+    about: Share ideas or give feedback in Discussions.

.github/ISSUE_TEMPLATE/feature_request.yml

@@ -0,0 +1,82 @@
+name: Feature Request
+description: Suggest a new feature or enhancement
+labels: ["enhancement", "triage"]
+body:
+  - type: markdown
+    attributes:
+      value: |
+        Thanks for suggesting a feature! Please describe what you'd like to see added.
+
+  - type: textarea
+    id: use-case
+    attributes:
+      label: Use Case
+      description: Describe your specific use case. What are you building? What's your goal?
+      placeholder: |
+        I'm building an AI agent that needs to...
+        My application handles...
+    validations:
+      required: true
+
+  - type: textarea
+    id: problem
+    attributes:
+      label: Problem Statement
+      description: What problem are you facing? What's missing or difficult today?
+      placeholder: Currently I have to... which causes...
+    validations:
+      required: true
+
+  - type: textarea
+    id: benefit
+    attributes:
+      label: How This Feature Would Help
+      description: Explain how this feature would improve your workflow or solve your problem
+      placeholder: With this feature, I would be able to...
+    validations:
+      required: true
+
+  - type: textarea
+    id: solution
+    attributes:
+      label: Proposed Solution
+      description: Describe your ideal solution (optional - we may have ideas too!)
+      placeholder: It would be great if Hindsight could...
+    validations:
+      required: false
+
+  - type: textarea
+    id: alternatives
+    attributes:
+      label: Alternatives Considered
+      description: Have you considered any alternative solutions or workarounds?
+    validations:
+      required: false
+
+  - type: dropdown
+    id: priority
+    attributes:
+      label: Priority
+      description: How important is this feature to you?
+      options:
+        - Nice to have
+        - Important - affects my workflow
+        - Critical - blocking my use case
+    validations:
+      required: true
+
+  - type: textarea
+    id: additional
+    attributes:
+      label: Additional Context
+      description: Any other context, mockups, or examples?
+    validations:
+      required: false
+
+  - type: checkboxes
+    id: checklist
+    attributes:
+      label: Checklist
+      options:
+        - label: I would be willing to contribute this feature
+          required: false

.github/workflows/deploy-docs.yml

@@ -20,18 +20,17 @@ concurrency:
 jobs:
   build:
     runs-on: ubuntu-latest
-    defaults:
-      run:
-        working-directory: hindsight-docs
     steps:
       - uses: actions/checkout@v4
       - uses: actions/setup-node@v4
         with:
           node-version: 20
           cache: npm
-          cache-dependency-path: hindsight-docs/package-lock.json
-      - run: npm ci
-      - run: npm run build
+          cache-dependency-path: package-lock.json
+      - 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
       - uses: actions/upload-pages-artifact@v3
         with:
           path: hindsight-docs/build

.github/workflows/release.yml

@@ -38,6 +38,14 @@ jobs:
       working-directory: ./hindsight
       run: uv build --out-dir dist
 
+    - 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 first, then hindsight-all which depends on them)
     - name: Publish hindsight-client to PyPI
       uses: pypa/gh-action-pypi-publish@release/v1
@@ -57,6 +65,18 @@ jobs:
         packages-dir: ./hindsight/dist
         skip-existing: true
 
+    - name: Publish hindsight-litellm to PyPI
+      uses: pypa/gh-action-pypi-publish@release/v1
+      with:
+        packages-dir: ./hindsight-integrations/litellm/dist
+        skip-existing: true
+
+    - name: Publish hindsight-embed to PyPI
+      uses: pypa/gh-action-pypi-publish@release/v1
+      with:
+        packages-dir: ./hindsight-embed/dist
+        skip-existing: true
+
     # Upload artifacts for GitHub release
     - name: Upload artifacts
       uses: actions/upload-artifact@v4
@@ -66,6 +86,8 @@ jobs:
           hindsight-clients/python/dist/*
           hindsight-api/dist/*
           hindsight/dist/*
+          hindsight-integrations/litellm/dist/*
+          hindsight-embed/dist/*
         retention-days: 1
 
   release-typescript-client:
@@ -80,18 +102,29 @@ jobs:
       with:
         node-version: '20'
         registry-url: 'https://registry.npmjs.org'
+        cache: 'npm'
+        cache-dependency-path: package-lock.json
 
     - name: Install dependencies
-      working-directory: ./hindsight-clients/typescript
-      run: npm ci
+      run: npm ci --workspace=hindsight-clients/typescript
 
     - name: Build
-      working-directory: ./hindsight-clients/typescript
-      run: npm run build
+      run: npm run build --workspace=hindsight-clients/typescript
 
     - name: Publish to npm
       working-directory: ./hindsight-clients/typescript
-      run: npm publish --access public
+      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 }}
 
@@ -106,6 +139,163 @@ jobs:
         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: ${{ secrets.NPM_TOKEN }}
+
+    - name: Pack for GitHub release
+      working-directory: ./hindsight-integrations/ai-sdk
+      run: npm pack
+
+    - name: Upload artifacts
+      uses: actions/upload-artifact@v4
+      with:
+        name: ai-sdk-integration
+        path: hindsight-integrations/ai-sdk/*.tgz
+        retention-days: 1
+
+  release-control-plane:
+    runs-on: ubuntu-latest
+    environment: npm
+
+    steps:
+    - uses: actions/checkout@v4
+
+    - name: Set up Node.js
+      uses: actions/setup-node@v4
+      with:
+        node-version: '20'
+        registry-url: 'https://registry.npmjs.org'
+        cache: 'npm'
+        cache-dependency-path: package-lock.json
+
+    - name: Install dependencies
+      run: npm ci
+
+    - name: Build TypeScript client (dependency)
+      run: npm run build --workspace=hindsight-clients/typescript
+
+    - name: Fix platform-specific native modules
+      run: |
+        # npm ci installs from lockfile which may have wrong platform binaries
+        # Delete hoisted native modules and reinstall for current platform
+        rm -rf node_modules/lightningcss node_modules/@tailwindcss
+        npm install lightningcss @tailwindcss/postcss @tailwindcss/node
+
+    - name: Build
+      run: npm run build --workspace=hindsight-control-plane
+
+    - name: Publish to npm
+      working-directory: ./hindsight-control-plane
+      run: |
+        set +e
+        OUTPUT=$(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-control-plane
+      run: npm pack
+
+    - name: Upload artifacts
+      uses: actions/upload-artifact@v4
+      with:
+        name: control-plane
+        path: hindsight-control-plane/*.tgz
+        retention-days: 1
+
   release-rust-cli:
     runs-on: ${{ matrix.os }}
     strategy:
@@ -150,6 +340,7 @@ jobs:
         retention-days: 1
 
   release-docker-images:
+    name: Release Docker (${{ matrix.image_name }}${{ matrix.tag_suffix }})
     runs-on: ubuntu-latest
     permissions:
       contents: read
@@ -159,10 +350,28 @@ jobs:
         include:
           - target: api-only
             image_name: hindsight-api
+            tag_suffix: ""
+            build_args: ""
+          - target: api-only
+            image_name: hindsight-api
+            tag_suffix: "-slim"
+            build_args: |
+              INCLUDE_LOCAL_MODELS=false
+              PRELOAD_ML_MODELS=false
           - target: cp-only
             image_name: hindsight-control-plane
+            tag_suffix: ""
+            build_args: ""
           - target: standalone
             image_name: hindsight
+            tag_suffix: ""
+            build_args: ""
+          - target: standalone
+            image_name: hindsight
+            tag_suffix: "-slim"
+            build_args: |
+              INCLUDE_LOCAL_MODELS=false
+              PRELOAD_ML_MODELS=false
 
     steps:
     - uses: actions/checkout@v4
@@ -170,7 +379,7 @@ jobs:
     - name: Free Disk Space
       uses: jlumbroso/free-disk-space@main
       with:
-        tool-cache: false
+        tool-cache: true
         android: true
         dotnet: true
         haskell: true
@@ -195,23 +404,49 @@ jobs:
       id: get_version
       run: echo "VERSION=${GITHUB_REF#refs/tags/v}" >> $GITHUB_OUTPUT
 
-    - name: Extract metadata
+    - name: Extract metadata for release tags
       id: meta
       uses: docker/metadata-action@v5
       with:
         images: ghcr.io/${{ github.repository_owner }}/${{ matrix.image_name }}
+        flavor: |
+          latest=auto
+          suffix=${{ matrix.tag_suffix }}
         tags: |
           type=semver,pattern={{version}},value=${{ steps.get_version.outputs.VERSION }}
           type=semver,pattern={{major}}.{{minor}},value=${{ steps.get_version.outputs.VERSION }}
           type=semver,pattern={{major}},value=${{ steps.get_version.outputs.VERSION }}
           type=raw,value=latest
 
-    - name: Build and push
+    # TODO: Re-enable smoke test when disk space issue is resolved
+    # # 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@v6
+    #   with:
+    #     context: .
+    #     file: docker/standalone/Dockerfile
+    #     target: ${{ matrix.target }}
+    #     push: false
+    #     load: true
+    #     tags: ${{ matrix.image_name }}:test
+    #     cache-from: type=gha
+    #     cache-to: type=gha,mode=max
+
+    # # Step 2: Test the image before pushing anything
+    # - name: Smoke test - verify container starts
+    #   env:
+    #     GROQ_API_KEY: ${{ secrets.GROQ_API_KEY }}
+    #   run: ./docker/test-image.sh "${{ matrix.image_name }}:test" "${{ matrix.target }}"
+
+    # Build multi-platform and push to release tags
+    - name: Build and push release images
       uses: docker/build-push-action@v6
       with:
         context: .
         file: docker/standalone/Dockerfile
         target: ${{ matrix.target }}
+        build-args: ${{ matrix.build_args }}
         push: true
         platforms: linux/amd64,linux/arm64
         tags: ${{ steps.meta.outputs.tags }}
@@ -219,6 +454,9 @@ jobs:
 
   release-helm-chart:
     runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      packages: write
 
     steps:
     - uses: actions/checkout@v4
@@ -228,12 +466,18 @@ jobs:
       with:
         version: 'latest'
 
+    - name: Log in to GHCR
+      run: echo "${{ secrets.GITHUB_TOKEN }}" | helm registry login ghcr.io -u ${{ github.actor }} --password-stdin
+
     - name: Lint Helm chart
       run: helm lint helm/hindsight
 
     - name: Package Helm chart
       run: helm package helm/hindsight --destination ./helm-packages
 
+    - name: Push to GHCR OCI
+      run: helm push helm-packages/*.tgz oci://ghcr.io/${{ github.repository_owner }}/charts
+
     - name: Upload artifacts
       uses: actions/upload-artifact@v4
       with:
@@ -243,7 +487,7 @@ jobs:
 
   create-github-release:
     runs-on: ubuntu-latest
-    needs: [release-python-packages, release-typescript-client, 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
 
@@ -266,6 +510,24 @@ jobs:
         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@v4
+      with:
+        name: control-plane
+        path: ./artifacts/control-plane
+
     - name: Download Rust CLI (Linux)
       uses: actions/download-artifact@v4
       with:
@@ -297,8 +559,16 @@ jobs:
         cp artifacts/python-packages/hindsight-clients/python/dist/* release-assets/ || true
         cp artifacts/python-packages/hindsight-api/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
         cp artifacts/rust-cli-linux/hindsight-linux-amd64 release-assets/ || true
         cp artifacts/rust-cli-darwin-amd64/hindsight-darwin-amd64 release-assets/ || true
@@ -307,54 +577,11 @@ jobs:
         cp artifacts/helm-chart/*.tgz release-assets/ || true
         ls -la release-assets/
 
-    - name: Generate release notes
-      run: |
-        cat << 'EOF' > release-notes.md
-        ## Quick Start
-
-        ```bash
-        # Install the CLI
-        curl -fsSL https://raw.githubusercontent.com/vectorize-io/hindsight/refs/heads/main/hindsight-cli/install.sh | bash
-
-        # Start the server
-        docker run -p 8888:8888 -p 9999:9999 \
-          -e HINDSIGHT_API_LLM_PROVIDER=openai \
-          -e HINDSIGHT_API_LLM_API_KEY=$OPENAI_API_KEY \
-          -e HINDSIGHT_API_LLM_MODEL=gpt-4o-mini \
-          ghcr.io/${{ github.repository_owner }}/hindsight:${{ steps.get_version.outputs.VERSION }}
-        ```
-
-        ## Docker Images
-        - `ghcr.io/${{ github.repository_owner }}/hindsight:${{ steps.get_version.outputs.VERSION }}` - Standalone (recommended)
-        - `ghcr.io/${{ github.repository_owner }}/hindsight-api:${{ steps.get_version.outputs.VERSION }}` - API only
-        - `ghcr.io/${{ github.repository_owner }}/hindsight-control-plane:${{ steps.get_version.outputs.VERSION }}` - Web UI only
-
-        ## CLI
-        ```bash
-        curl -fsSL https://raw.githubusercontent.com/vectorize-io/hindsight/refs/heads/main/hindsight-cli/install.sh | bash
-        ```
-
-        ## Python
-        ```bash
-        pip install hindsight-all  # or hindsight-api, hindsight-client
-        ```
-
-        ## TypeScript/JavaScript
-        ```bash
-        npm install @vectorize-io/hindsight-client
-        ```
-
-        ## Helm
-        ```bash
-        helm install hindsight oci://ghcr.io/${{ github.repository_owner }}/charts/hindsight --version ${{ steps.get_version.outputs.VERSION }}
-        ```
-        EOF
-
     - name: Create GitHub Release
       uses: softprops/action-gh-release@v2
       with:
         files: release-assets/*
-        body_path: release-notes.md
+        generate_release_notes: true
         draft: false
         prerelease: false
       env:

.gitignore

@@ -5,12 +5,18 @@ build/
 dist/
 wheels/
 *.egg-info
-
+.mcp.json
+.osgrep
 # Virtual environments
 .venv
 
-# Environment variables
+# Node
+node_modules/
+
+# Environment variables and local config
 .env
+docker-compose.yml
+docker-compose.override.yml
 
 # IDE
 .idea/
@@ -21,6 +27,10 @@ wheels/
 # NLTK data (will be downloaded automatically)
 nltk_data/
 
+# Monitoring stack (Prometheus/Grafana binaries and data)
+.monitoring/
+.pgbouncer/
+
 # Large benchmark datasets (will be downloaded automatically)
 **/longmemeval_s_cleaned.json
 
@@ -29,8 +39,21 @@ logs/
 
 .DS_Store
 
+# Generated docs files
+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
+hindsight-clients/rust/target
+.claude
+whats-next.md
+TASK.md
+# Changelog is now tracked in hindsight-docs/src/pages/changelog.md
+# CHANGELOG.md
+
+blog-post*

AGENTS.md

@@ -0,0 +1,3 @@
+# AGENTS.md
+
+See [CLAUDE.md](./CLAUDE.md) for project documentation and coding conventions.

CLAUDE.md

@@ -0,0 +1,326 @@
+# CLAUDE.md
+
+This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
+
+## Project Overview
+
+Hindsight is an agent memory system that provides long-term memory for AI agents using biomimetic data structures. Memories are organized as:
+- **World facts**: General knowledge ("The sky is blue")
+- **Experience facts**: Personal experiences ("I visited Paris in 2023")
+- **Mental models**: Consolidated knowledge synthesized from facts ("User prefers functional programming patterns")
+
+## Development Commands
+
+### API Server (Python/FastAPI)
+```bash
+# Start API server (loads .env automatically)
+./scripts/dev/start-api.sh
+
+# Run all tests (parallelized with pytest-xdist)
+cd hindsight-api && uv run pytest tests/
+
+# Run specific test file
+cd hindsight-api && uv run pytest tests/test_http_api_integration.py -v
+
+# Run single test function
+cd hindsight-api && uv run pytest tests/test_retain.py::test_retain_simple -v
+
+# Lint and 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 && uv run ty check hindsight_api/
+```
+
+### Control Plane (Next.js)
+```bash
+./scripts/dev/start-control-plane.sh
+# Or manually:
+cd hindsight-control-plane && npm run dev
+```
+
+### Documentation Site (Docusaurus)
+```bash
+./scripts/dev/start-docs.sh
+```
+
+
+### Generating Clients/OpenAPI
+```bash
+# Regenerate OpenAPI spec after API changes (REQUIRED after changing endpoints)
+./scripts/generate-openapi.sh
+
+# Regenerate all client SDKs (Python, TypeScript, Rust)
+./scripts/generate-clients.sh
+```
+
+### 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/**: 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)
+- **hindsight-clients/**: Generated SDK clients (Python, TypeScript, Rust)
+- **hindsight-docs/**: Docusaurus documentation site
+- **hindsight-integrations/**: Framework integrations (LiteLLM, OpenAI)
+- **hindsight-dev/**: Development tools and benchmarks
+
+### 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, 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
+- `query_analyzer.py`: Query intent analysis
+
+**retain/**: Memory ingestion pipeline
+- `orchestrator.py`: Coordinates the retain flow
+- `fact_extraction.py`: LLM-based fact extraction from content
+- `link_utils.py`: Entity link creation and management
+
+**search/**: Multi-strategy retrieval
+- `retrieval.py`: Main retrieval orchestrator
+- `graph_retrieval.py`: Entity/relationship graph traversal
+- `mpfp_retrieval.py`: Multi-Path Fact Propagation retrieval
+- `fusion.py`: Reciprocal rank fusion for combining results
+- `reranking.py`: Cross-encoder reranking
+
+### API Layer (hindsight-api/hindsight_api/api/)
+- `http.py`: FastAPI HTTP routers (~80KB) for all REST endpoints
+- `mcp.py`: Model Context Protocol server implementation
+
+Main operations:
+- **Retain**: Store memories, extracts facts/entities/relationships
+- **Recall**: Retrieve memories via 4 parallel strategies (semantic, BM25, graph, temporal) + reranking
+- **Reflect**: Disposition-aware reasoning using memories and mental models.
+
+### Database
+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/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
+
+2. **Migration template**:
+   ```python
+   """Description of the migration
+
+   Revision ID: f1a2b3c4d5e6
+   Revises: <previous_revision_id>
+   Create Date: YYYY-MM-DD
+   """
+   from collections.abc import Sequence
+   from alembic import context, op
+
+   revision: str = "f1a2b3c4d5e6"
+   down_revision: str | Sequence[str] | None = "<previous_revision_id>"
+   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"CREATE INDEX ... ON {schema}table_name(...)")
+
+   def downgrade() -> None:
+       schema = _get_schema_prefix()
+       op.execute(f"DROP INDEX IF EXISTS {schema}index_name")
+   ```
+
+3. **Run migrations locally**:
+   ```bash
+   # Set database URL and run migrations
+   uv run hindsight-admin run-db-migration
+
+   # Run on a specific tenant schema
+   uv run hindsight-admin run-db-migration --schema tenant_xyz
+   ```
+
+## Key Conventions
+
+### Code Quality
+**Always run the lint script after making Python or TypeScript/Node changes:**
+```bash
+./scripts/hooks/lint.sh
+```
+This runs the same checks as the pre-commit hook (Ruff for Python, ESLint/Prettier for TypeScript).
+
+### Memory Banks
+- Each bank is an isolated memory store (like a "brain" for one user/agent)
+- Banks have dispositions (skepticism, literalism, empathy traits 1-5) affecting reflect
+- Banks can have background context
+- Bank isolation is strict - no cross-bank data leakage
+
+### API Design
+- All endpoints operate on a single bank per request
+- Multi-bank queries are client responsibility to orchestrate
+- Disposition traits only affect reflect, not recall
+
+### Control Plane API Routes
+
+When adding or modifying parameters in the dataplane API (hindsight-api), you must also update the control plane routes that proxy to it:
+
+1. **API Routes** (`hindsight-control-plane/src/app/api/`):
+   - `recall/route.ts` - proxies to `/v1/default/banks/{bank_id}/memories/recall`
+   - `reflect/route.ts` - proxies to `/v1/default/banks/{bank_id}/reflect`
+   - `memories/retain/route.ts` - proxies to `/v1/default/banks/{bank_id}/memories/retain`
+   - Other routes follow the same pattern
+
+2. **Client types** (`hindsight-control-plane/src/lib/api.ts`):
+   - Update the TypeScript type definitions for `recall()`, `reflect()`, `retain()` etc.
+
+3. **Checklist when adding new API parameters**:
+   - Add parameter extraction in the route handler (destructure from `body`)
+   - Pass the parameter to the SDK call
+   - Update the client type definition in `lib/api.ts`
+   - Update any UI components that need to use the new parameter
+
+### Python Style
+- Python 3.11+, type hints required
+- Async throughout (asyncpg, async FastAPI)
+- Pydantic models for request/response
+- Ruff for linting (line-length 120)
+- No Python files at project root - maintain clean directory structure
+- **Never use multi-item tuple return values** - prefer dataclass or Pydantic model for structured returns
+
+### Type Safety with Pydantic Models
+**NEVER use raw `dict` types for structured data.** Always use Pydantic models:
+- Use Pydantic `BaseModel` for all data structures passed between functions
+- Add `@field_validator` for type coercion (e.g., ensuring datetimes are timezone-aware)
+- Avoid `dict.get()` patterns - use typed model attributes instead
+- Parse external data (JSON, API responses) into Pydantic models at the boundary
+- This catches type errors at parse time, not deep in business logic
+
+```python
+# BAD - error-prone dict access
+def process(data: dict) -> str:
+    return data.get("name", "")  # No validation, silent failures
+
+# GOOD - typed and validated
+class UserData(BaseModel):
+    name: str
+    created_at: datetime
+
+    @field_validator("created_at", mode="before")
+    @classmethod
+    def ensure_tz_aware(cls, v):
+        if isinstance(v, str):
+            v = datetime.fromisoformat(v.replace("Z", "+00:00"))
+        if v.tzinfo is None:
+            return v.replace(tzinfo=timezone.utc)
+        return v
+
+def process(data: UserData) -> str:
+    return data.name  # Type-safe, validated at construction
+```
+
+### TypeScript Style
+- Next.js App Router for control plane
+- Tailwind CSS with shadcn/ui components
+
+### Adding New API Configuration Flags
+
+Configuration follows a hierarchical system: **Global (env vars) → Tenant (via extension) → Bank (database)**.
+
+Fields must be categorized as either **hierarchical** (can be overridden per-tenant/bank) or **static** (server-level only).
+
+#### Adding a New Configuration Field
+
+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
+   - **Mark as hierarchical or static** by adding to `_HIERARCHICAL_FIELDS` set (hierarchical) or leaving it out (static)
+   - Add initialization in `from_env()` method
+
+   ```python
+   # Hierarchical field (can be overridden per-bank)
+   _HIERARCHICAL_FIELDS = {
+       ...,
+       "my_setting",  # Add here for hierarchical
+   }
+
+   # Static field - just don't add to _HIERARCHICAL_FIELDS
+   ```
+
+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**:
+   ```python
+   # Config is resolved automatically per bank via ConfigResolver
+   config_dict = await self._config_resolver.get_bank_config(bank_id, context)
+   value = config_dict["my_setting"]
+   ```
+
+4. **Use static config** (non-hierarchical):
+   ```python
+   from ...config import get_config
+   config = get_config()
+   value = config.my_static_field
+   ```
+
+5. **Documentation** (`hindsight-docs/docs/developer/configuration.md`):
+   - Add to appropriate section table with Variable, Description, Default
+   - Mark if it's hierarchical (can be overridden per-bank)
+
+#### Hierarchical vs Static Guidelines
+
+**Hierarchical** (per-bank overridable):
+- LLM settings (provider, model, API key, base URL)
+- Operation-specific settings (retain mode, chunk size, etc.)
+- Feature flags that vary by customer/bank
+
+**Static** (server-level only):
+- Infrastructure settings (database URL, port, host)
+- Global limits (max concurrent operations)
+- System-wide feature flags
+
+## Environment Setup
+
+```bash
+cp .env.example .env
+# Edit .env with LLM API key
+
+# Python deps
+uv sync --directory hindsight-api/
+
+# Node deps (uses npm workspaces)
+npm install
+```
+
+Required env vars:
+- `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., 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: false, disabled for security)

CONTRIBUTING.md

@@ -5,13 +5,23 @@ Thanks for your interest in contributing to Hindsight!
 ## Getting Started
 
 1. Fork and clone the repository
-2. Install dependencies:
    ```bash
-   cd hindsight-api && uv sync
+   git clone [email protected]:vectorize-io/hindsight.git
+   cd hindsight
    ```
-3. Set up your environment:
+2. Set up your environment:
    ```bash
-   export OPENAI_API_KEY=your-key
+   cp .env.example .env
+   ```
+   Edit the .env to add LLM API key and config as required
+
+3. Install dependencies:
+   ```bash
+   # Python dependencies
+   uv sync --directory hindsight-api/
+
+   # Node dependencies (uses npm workspaces)
+   npm install
    ```
 
 ## Development
@@ -41,7 +51,36 @@ cd hindsight-api
 uv run pytest tests/
 ```
 
-### Code style
+### Code Style
+
+We use [Ruff](https://docs.astral.sh/ruff/) for Python linting and formatting, and ESLint/Prettier for TypeScript.
+
+#### Setting up git hooks (recommended)
+
+Set up git hooks to automatically lint and format code before each commit:
+
+```bash
+./scripts/setup-hooks.sh
+```
+
+This configures git to use the hooks in `.githooks/`, which run all scripts in `scripts/hooks/` on commit. The lint hook runs in parallel:
+- **Python**: `ruff check --fix`, `ruff format`, `ty check`
+- **TypeScript**: `eslint --fix`, `prettier`
+
+#### Manual linting and formatting
+
+```bash
+# Run all lints (same as pre-commit)
+./scripts/hooks/lint.sh
+
+# Or run individually for Python:
+cd hindsight-api
+uv run ruff check --fix .   # Lint and auto-fix
+uv run ruff format .        # Format code
+uv run ty check hindsight_api  # Type check
+```
+
+#### Style guidelines
 
 - Use Python type hints
 - Follow existing code patterns
@@ -54,6 +93,34 @@ uv run pytest tests/
 3. Run tests to ensure nothing breaks
 4. Submit a PR with a clear description of changes
 
+## Release Process
+
+The project uses `scripts/release.sh` for creating releases. This script automates the entire release workflow:
+
+1. Bumps version in all components (API, clients, CLI, control plane, Helm)
+2. **Regenerates OpenAPI spec and client SDKs** (Python, TypeScript, Rust)
+3. Updates documentation versioning
+4. Creates a commit and git tag
+5. Pushes to GitHub (triggers CI/CD to publish packages)
+
+### Usage
+
+```bash
+./scripts/release.sh <version>
+```
+
+**Example:**
+```bash
+./scripts/release.sh 0.5.0
+```
+
+### Important for Developers
+
+- During development, version bumps in `__init__.py` do NOT require client regeneration
+- Clients are only regenerated during releases
+- Do not manually run `./scripts/generate-clients.sh` unless testing generation changes
+- Client version comments will reflect the API version from the latest release
+
 ## Reporting Issues
 
 Open an issue on GitHub with:

NEW.md

@@ -1,268 +0,0 @@
-<div align="center">
-
-# Hindsight
-
-**Agent Memory that Works Like Human Memory**
-
-[![CI](https://github.com/vectorize-io/hindsight/actions/workflows/test.yml/badge.svg)](https://github.com/vectorize-io/hindsight/actions/workflows/test.yml)
-[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
-[![PyPI - hindsight-client](https://img.shields.io/pypi/v/hindsight-client?label=hindsight-client)](https://pypi.org/project/hindsight-client/)
-[![npm](https://img.shields.io/npm/v/@vectorize-io/hindsight-client)](https://www.npmjs.com/package/@vectorize-io/hindsight-client)
-[![Slack Community](https://img.shields.io/badge/Slack-Join%20Community-4A154B?logo=slack)](https://join.slack.com/t/hindsight-space/shared_invite/zt-3klo21kua-VUCC_zHP5rIcXFB1_5yw6A)
-
-[Documentation](https://vectorize-io.github.io/hindsight) • [Paper](./Hindsight.pdf) • [Examples](./examples)
-
-</div>
-
----
-
-## What is Hindsight?
-
-Hindsight is an agent memory system built to create smarter agents that learn over time. It eliminates the shortcomings of alternative techniques such as RAG and knowledge graph.
-
-Hindsight addresses common challenges that have frustrated AI engineers building agents to automate tasks and assist users with conversational interfaces. Many of these challenges stem directly from a lack of memory.
-
-- **Inconsistency:** Agents complete tasks successfully one time, then fail when asked to complete the same task again. Memory gives the agent a mechanism to remember what worked and what didn't and to use that information to reduce errors and improve consistency.
-- **Hallucinations:** Long term memory can be seeded with external knowledge to ground agent behavior in reliable sources to augment training data.
-- **Cognitive Overload:** As workflows get complex, retrievals, tool calls, user messages and agent responses can grow to fill the context window leading to context rot. Short term memory optimization allows agents to reduce tokens and focus context by removing irrelevant details.
-
-## How Hindsight Works
-
-![Overview](./hindsight-docs/static/img/hindsight-overview.png)
-
-Hindsight organizes memory into four networks to mimic the way 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")
-- **Opinion:** Beliefs with confidence scores ("I shouldn't touch the stove again" - .99 confidence)
-- **Observation:** Complex mental models derived by reflecting on facts and experiences ("Curling irons, ovens, and fire are also hot. I shouldn't touch those either.")
-
-Hindsight provides three simple methods to interact with the system:
-
-- **Retain:** Provide information to Hindsight that you want it to remember
-- **Recall:** Retrieve memories from Hindsight
-- **Reflect:** Reflect on memories and experiences to generate new observations and insights from existing memories.
-
-Memories in Hindsight are stored in banks (e.g. memory banks). When memories are retained, they are transformed to construct a series of search indexes, time series data, and entity/relationship graphs.
-
----
-
-## Quick Start
-
-### Docker (recommended)
-
-```bash
-export OPENAI_API_KEY=your-key
-
-docker run -p 8888:8888 -p 9999:9999 \
-  -e HINDSIGHT_API_LLM_PROVIDER=openai \
-  -e HINDSIGHT_API_LLM_API_KEY=$OPENAI_API_KEY \
-  -e HINDSIGHT_API_LLM_MODEL=gpt-4o-mini \
-  -v $HOME/.hindsight-docker:/home/hindsight/.pg0 \
-  ghcr.io/vectorize-io/hindsight
-```
-
-API: http://localhost:8888  
-UI: http://localhost:9999
-
-Install client:
-
-```bash
-pip install hindsight-client
-# or
-npm install @vectorize-io/hindsight-client
-```
-
-Python example:
-
-```python
-from hindsight import HindsightClient
-
-client = HindsightClient(base_url="http://localhost:8888")
-
-# Store
-client.retain(bank_id="my-agent", content="Alice works at Google as a software engineer")
-
-# Query
-results = client.recall(bank_id="my-agent", query="What does Alice do?")
-
-# Reflect
-response = client.reflect(bank_id="my-agent", query="Tell me about Alice")
-print(response.text)
-```
-
-### Python (embedded, no Docker)
-
-```bash
-pip install hindsight-all
-```
-
-```python
-import os
-from hindsight import HindsightServer, HindsightClient
-
-with HindsightServer(
-    llm_provider="openai",
-    llm_model="gpt-4o-mini", 
-    llm_api_key=os.environ["OPENAI_API_KEY"]
-) as server:
-    client = HindsightClient(base_url=server.url)
-    client.retain(bank_id="my-agent", content="Alice works at Google")
-    results = client.recall(bank_id="my-agent", query="Where does Alice work?")
-```
-
-### TypeScript
-
-```bash
-npm install @vectorize-io/hindsight-client
-```
-
-```typescript
-import { HindsightClient } from '@vectorize-io/hindsight-client';
-
-const client = new HindsightClient({ baseUrl: 'http://localhost:8888' });
-
-await client.retain('my-agent', 'Alice loves hiking in Yosemite');
-const response = await client.recall('my-agent', 'What does Alice like?');
-```
-
----
-
-## Architecture & Operations
-
-### Retain
-
-The `retain` operation is used to push new memories into Hindsight. It tells Hindsight to _retain_ the information you pass in as an input.
-
-```python
-from hindsight_client import Hindsight
-
-client = Hindsight(base_url="http://localhost:8888")
-
-# Simple
-client.retain(
-    bank_id="my-bank",
-    content="Alice works at Google as a software engineer"
-)
-
-# With context and timestamp
-client.retain(
-    bank_id="my-bank",
-    content="Alice got promoted to senior engineer",
-    context="career update",
-    timestamp="2025-06-15T10:00:00Z"
-)
-```
-
-Behind the scenes, the retain operation uses an LLM to extract key facts, temporal data, entities, and relationships. It passes these through a normalization process to transform extracted data into canonical entities, time series, and search indexes along with metadata. These representations create the pathways for accurate memory retrieval in the recall and reflect operations. 
-
-![Retain Operation](hindsight-docs/static/img/retain-operation.png)
-
-### Recall
-
-The recall operation is used to retrieve memories. These memories can come from any of the memory types (world, experiences, etc.)
-
-```python
-from hindsight_client import Hindsight
-
-client = Hindsight(base_url="http://localhost:8888")
-
-# Simple
-client.recall(bank_id="my-bank", query="What does Alice do?")
-
-# Temporal
-results = client.recall(bank_id="my-bank", query="What happened in June?")
-
-
-```
-
-Recall performs 4 retrieval strategies in parallel:
-- Semantic: Vector similarity
-- Keyword: BM25 exact matching
-- Graph: Entity/temporal/causal links
-- Temporal: Time range filtering
-
-![Retain Operation](hindsight-docs/static/img/recall-operation.png)
-
-The individual results from the retrievals are merged, then ordered by relevance using reciprocal rank fusion and a cross-encoder reranking model.
-
-The final output is trimmed as needed to fit within the token limit.
-
-### Reflect
-
-The reflect operation is used to perform a more thorough analysis of existing memories. This allows the agent to form new connections between memories which are then persisted as opinions and/or observations. When building agents, the reflect operation is a key capability to enable the agent to learn from its experiences. 
-
-For example, the `reflect` operation can be used to support use cases such as:
-
-- An **AI Project Manager** reflecting on what risks need to be mitigated on a project.
-- A **Sales Agent** reflecting on why certain outreach messages have gotten responses while others haven't.
-- A **Support Agent** reflecting on opportunities where customers have questions not answered by current product documentation.
-
-The `reflect` operation can also be used to handle on-demand question answering or analysis which require more deep thinking.
-
-```python
-from hindsight_client import Hindsight
-
-client = Hindsight(base_url="http://localhost:8888")
-
-client.reflect(bank_id="my-bank", query="What should I know about Alice?")
-```
-
-![Retain Operation](hindsight-docs/static/img/reflect-operation.png)
-
-## Integrations
-
-### OpenAI Drop-in Replacement
-
-```python
-from hindsight_openai import configure, OpenAI
-
-configure(hindsight_api_url="http://localhost:8888", agent_id="my-assistant")
-
-client = OpenAI(api_key="sk-...")
-response = client.chat.completions.create(
-    model="gpt-4",
-    messages=[{"role": "user", "content": "What did we discuss?"}]
-)
-# Memory automatically recalled and stored
-```
-
-### Examples
-
-[Examples directory](./examples) includes:
-
-- Basic usage
-- Multi-session conversations
-- Temporal queries
-- Entity reasoning
-- Opinion tracking
-- Production setup (Docker Compose + monitoring)
-
----
-
-## Resources
-
-**Documentation:** [vectorize-io.github.io/hindsight](https://vectorize-io.github.io/hindsight)
-
-**Clients:**
-- [Python](http://hindsight.vectorize.io/sdks/python)
-- [Node.js](http://hindsight.vectorize.io/sdks/nodejs)
-- [REST API](http://hindsight.vectorize.io/api-reference)
-
-**Community:**
-- [Slack](https://join.slack.com/t/hindsight-space/shared_invite/zt-3klo21kua-VUCC_zHP5rIcXFB1_5yw6A)
-- [GitHub Issues](https://github.com/vectorize-io/hindsight/issues)
-
----
-
-## Contributing
-
-See [CONTRIBUTING.md](./CONTRIBUTING.md).
-
-## License
-
-MIT — see [LICENSE](./LICENSE)
-
----
-
-Built by [Vectorize.io](https://vectorize.io)

README.md

@@ -1,101 +1,309 @@
-# Hindsight
+<div align="center">
 
-[![CI](https://github.com/vectorize-io/hindsight/actions/workflows/test.yml/badge.svg)](https://github.com/vectorize-io/hindsight/actions/workflows/test.yml)
+![Hindsight Banner](./hindsight-docs/static/img/hindsight-github-banner.png)
+
+[Documentation](https://hindsight.vectorize.io) • [Paper](https://arxiv.org/abs/2512.12818) • [Cookbook](https://hindsight.vectorize.io/cookbook) • [Hindsight Cloud](https://ui.hindsight.vectorize.io/signup)
+
+[![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)
-[![PyPI - hindsight-client](https://img.shields.io/pypi/v/hindsight-client?label=hindsight-client)](https://pypi.org/project/hindsight-client/)
-[![PyPI - hindsight-api](https://img.shields.io/pypi/v/hindsight-api?label=hindsight-api)](https://pypi.org/project/hindsight-api/)
-[![PyPI - hindsight-all](https://img.shields.io/pypi/v/hindsight-all?label=hindsight-all)](https://pypi.org/project/hindsight-all/)
-[![npm](https://img.shields.io/npm/v/@vectorize-io/hindsight-client)](https://www.npmjs.com/package/@vectorize-io/hindsight-client)
+![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)
 
-**Long-term memory for AI agents.**
 
-## Why Hindsight?
+</div>
 
-AI assistants forget everything between sessions. Every conversation starts from zero—no context about who you are, what you've discussed, or what the memory bank has learned. This isn't just inconvenient; it fundamentally limits what AI memory banks can do.
+---
 
-**The problem is harder than it looks:**
+## What is Hindsight?
 
-- **Simple vector search isn't enough** — "What did Alice do last spring?" requires temporal reasoning, not just semantic similarity
-- **Facts get disconnected** — Knowing "Alice works at Google" and "Google is in Mountain View" should let you answer "Where does Alice work?" even if you never stored that directly
-- **Memory banks need opinions** — A coding assistant that remembers "the user prefers functional programming" should weigh that when making recommendations
-- **Context matters** — The same information means different things to different memory banks with different personalities
+Hindsight™ is an agent memory system built to create smarter agents that learn over time. Most agent memory systems focus on recalling conversation history. Hindsight is focused on making agents that learn, not just remember.
 
-Hindsight solves these problems with a memory system designed specifically for AI memory banks.
+
+<video src="https://github.com/user-attachments/assets/923b798d-3581-4897-bb62-9cfa5a931682" controls></video>
+
+It eliminates the shortcomings of alternative techniques such as RAG and knowledge graph and delivers state-of-the-art performance on long term memory tasks.
+
+## Memory Performance & Accuracy
+
+Hindsight is the most accurate agent memory system ever tested according to benchmark performance. It has achieved state-of-the-art performance on the LongMemEval benchmark, widely used to assess memory system performance across a variety of conversational AI scenarios. The current reported performance of Hindsight and other agent memory solutions as of January 2026 is shown here:
+
+![Overview](./hindsight-docs/static/img/hindsight-bench.jpg)
+
+The benchmark performance data for Hindsight has been independently reproduced by research collaborators at the Virginia Tech [Sanghani Center for Artificial Intelligence and Data Analytics](https://sanghani.cs.vt.edu/) and The Washington Post. Other scores are self-reported by software vendors.
+
+Hindsight is being used in production at Fortune 500 enterprises and by a growing number of AI startups. 
+
+## Adding Hindsight to Your AI Agents
+
+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.
+
+![Hindsight Banner](./hindsight-docs/static/img/migration-code.png)
+
+---
+
+> 🤖 **Using a coding agent?** Install the Hindsight documentation skill for instant access to docs while you code:
+> ```bash
+> npx skills add https://github.com/vectorize-io/hindsight --skill hindsight-docs
+> ```
+> Works with Claude Code, Cursor, and other AI coding assistants.
+
+---
 
 
 ## Quick Start
 
-### Option 1: Docker (recommended)
-
-Get the full experience with the API and Control Plane UI:
+### Docker (recommended)
 
 ```bash
-export OPENAI_API_KEY=your-key
-docker run -p 8888:8888 -p 9999:9999 \
-  -e HINDSIGHT_API_LLM_PROVIDER=openai \
+export OPENAI_API_KEY=sk-xxx
+
+docker run --rm -it --pull always -p 8888:8888 -p 9999:9999 \
   -e HINDSIGHT_API_LLM_API_KEY=$OPENAI_API_KEY \
-  -e HINDSIGHT_API_LLM_MODEL=gpt-4o-mini \
   -v $HOME/.hindsight-docker:/home/hindsight/.pg0 \
-  ghcr.io/vectorize-io/hindsight
+  ghcr.io/vectorize-io/hindsight:latest
 ```
 
-- **API**: http://localhost:8888
-- **Control Plane UI**: http://localhost:9999
+>API: http://localhost:8888
+>UI: http://localhost:9999
 
-Then use the Python client:
+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).
+
+
+
+### Docker (external PostgreSQL)
 
 ```bash
-pip install hindsight-client
+export OPENAI_API_KEY=sk-xxx
+export HINDSIGHT_DB_PASSWORD=choose-a-password
+cd docker/docker-compose
+docker compose up 
 ```
 
+
+>API: http://localhost:8888
+>UI: http://localhost:9999
+
+### Client
+
+```bash
+pip install hindsight-client -U
+# or
+npm install @vectorize-io/hindsight-client
+```
+
+#### Python
+
 ```python
-from hindsight import HindsightClient
+from hindsight_client import Hindsight
 
-client = HindsightClient(base_url="http://localhost:8888")
+client = Hindsight(base_url="http://localhost:8888")
 
-# Store memories
-client.retain(bank_id="my-agent", content="Alice works at Google as a software engineer")
-client.retain(bank_id="my-agent", content="Alice mentioned she loves hiking in the mountains")
+# Retain: Store information
+client.retain(bank_id="my-bank", content="Alice works at Google as a software engineer")
 
-# Query with temporal reasoning
-results = client.recall(bank_id="my-agent", query="What does Alice do for work?")
+# Recall: Search memories
+client.recall(bank_id="my-bank", query="What does Alice do?")
 
-# Get a synthesized perspective
-response = client.reflect(bank_id="my-agent", query="Tell me about Alice")
-print(response.text)
+# Reflect: Generate disposition-aware response
+client.reflect(bank_id="my-bank", query="Tell me about Alice")
 ```
 
-### Option 2: Embedded (no docker/server required)
-
-For quick prototyping, run everything in-process:
+#### Node.js / TypeScript
 
 ```bash
-pip install hindsight-all
-export OPENAI_API_KEY=your-key
+npm install @vectorize-io/hindsight-client
+```
+
+```javascript
+const { HindsightClient } = require('@vectorize-io/hindsight-client');
+
+const main = async () => {
+  const client = new HindsightClient({ baseUrl: 'http://localhost:8888' });
+
+  await client.retain('my-bank', 'Alice loves hiking in Yosemite');
+
+  const results = await client.recall('my-bank', 'What does Alice like?');
+  console.log(results);
+}
+
+main();
+```
+
+
+### Python Embedded (no server required)
+
+```bash
+pip install hindsight-all -U
 ```
 
 ```python
 import os
 from hindsight import HindsightServer, HindsightClient
 
-with HindsightServer(llm_provider="openai", llm_model="gpt-4o-mini", llm_api_key=os.environ["OPENAI_API_KEY"]) as server:
+with HindsightServer(
+    llm_provider="openai",
+    llm_model="gpt-5-mini", 
+    llm_api_key=os.environ["OPENAI_API_KEY"]
+) as server:
     client = HindsightClient(base_url=server.url)
-
-    client.retain(bank_id="my-user", content="User prefers functional programming")
-    response = client.reflect(bank_id="my-user", query="What coding style should I use?")
-    print(response.text)
+    client.retain(bank_id="my-bank", content="Alice works at Google")
+    results = client.recall(bank_id="my-bank", query="Where does Alice work?")
 ```
 
 
+---
 
-## Documentation
+## Use Cases
 
-Full documentation: [hindsight.vectorize.io](https://hindsight.vectorize.io)
+
+Hindsight is built to support conversational AI agents as well as agents that are intended to perform tasks autonomously. The ideal use case for Hindsight are agents that require a blend of these features such as AI employees that need to handle open-ended tasks, change behavior based on user feedback, and learn to perform complex tasks to automate work at a level that approximates a human work. Hindsight can be used with simple AI workflows like those built with n8n and other similar tools, but may be overkill for such applications.
+
+### Per-User Memories and Chat History
+
+One of the simpler use cases you can use Hindsight for is to personalize AI chatbots and other conversational agents by storing and recalling memories associated with individual users.
+
+The requirements for this use case usually look something like this:
+
+![Per-User Memories](./hindsight-docs/static/img/per-user-memory-requirements.png)
+
+<video src="https://github.com/user-attachments/assets/4805e8e1-e7d1-47c6-a4f8-2344a5ec8906" controls></video>
+
+Satisfying these requirements in Hindsight is straightforward. When new user inputs and tool calls are ingested into Hindsight using the retain operation, custom metadata can be used to enrich the new memories. Metadata provides a convenient way to isolate memories that need to be restricted to a given user. Once these are fed into the retain operation, any raw memories and mental models that get created can be filtered when retrieving relevant memories. 
+
+![Per-User Memories](./hindsight-docs/static/img/per-user-memory-howto.png)
+
+---
+
+## Architecture & Operations
+
+![Overview](./hindsight-docs/static/img/hindsight-overview.webp)
+
+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")
+- **Mental Models:** Learned understanding of the agent's world formed by reflecting on raw memories and experiences.
+
+Memories in Hindsight are stored in banks (i.e. memory banks). When memories are added to Hindsight, they are pushed into either the world facts or experiences memory pathway. They are then represented as a combination of entities, relationships, and time series with sparse/dense vector representations to aid in later recall.
+
+Hindsight provides three simple methods to interact with the system:
+
+- **Retain:** Provide information to Hindsight that you want it to remember
+- **Recall:** Retrieve memories from Hindsight
+- **Reflect:** Reflect on memories and experiences to generate new observations and insights from existing memories.
+
+### Retain
+
+The `retain` operation is used to push new memories into Hindsight. It tells Hindsight to _retain_ the information you pass in as an input.
+
+```python
+from hindsight_client import Hindsight
+
+client = Hindsight(base_url="http://localhost:8888")
+
+# Simple
+client.retain(
+    bank_id="my-bank",
+    content="Alice works at Google as a software engineer"
+)
+
+# With context and timestamp
+client.retain(
+    bank_id="my-bank",
+    content="Alice got promoted to senior engineer",
+    context="career update",
+    timestamp="2025-06-15T10:00:00Z"
+)
+```
+
+Behind the scenes, the retain operation uses an LLM to extract key facts, temporal data, entities, and relationships. It passes these through a normalization process to transform extracted data into canonical entities, time series, and search indexes along with metadata. These representations create the pathways for accurate memory retrieval in the recall and reflect operations. 
+
+![Retain Operation](hindsight-docs/static/img/retain-operation.webp)
+
+### Recall
+
+The recall operation is used to retrieve memories. These memories can come from any of the memory types (world, experiences, etc.)
+
+```python
+from hindsight_client import Hindsight
+
+client = Hindsight(base_url="http://localhost:8888")
+
+# Simple
+client.recall(bank_id="my-bank", query="What does Alice do?")
+
+# Temporal
+client.recall(bank_id="my-bank", query="What happened in June?")
+```
+
+Recall performs 4 retrieval strategies in parallel:
+- Semantic: Vector similarity
+- Keyword: BM25 exact matching
+- Graph: Entity/temporal/causal links
+- Temporal: Time range filtering
+
+![Retain Operation](hindsight-docs/static/img/recall-operation.webp)
+
+The individual results from the retrievals are merged, then ordered by relevance using reciprocal rank fusion and a cross-encoder reranking model.
+
+The final output is trimmed as needed to fit within the token limit.
+
+### Reflect
+
+The reflect operation is used to perform a more thorough analysis of existing memories. This allows the agent to form new connections between memories and build a more thorough understanding of its world.
+
+For example, the `reflect` operation can be used to support use cases such as:
+
+- An **AI Project Manager** reflecting on what risks need to be mitigated on a project.
+- A **Sales Agent** reflecting on why certain outreach messages have gotten responses while others haven't.
+- A **Support Agent** reflecting on opportunities where customers have questions not answered by current product documentation.
+
+The `reflect` operation can also be used to handle on-demand question answering or analysis which require more deep thinking.
+
+```python
+from hindsight_client import Hindsight
+
+client = Hindsight(base_url="http://localhost:8888")
+
+client.reflect(bank_id="my-bank", query="What should I know about Alice?")
+```
+
+![Retain Operation](hindsight-docs/static/img/reflect-operation.webp)
+
+---
+
+## Resources
+
+**Documentation:** 
+- [https://hindsight.vectorize.io](https://hindsight.vectorize.io)
+
+**Clients:**
+- [Python](http://hindsight.vectorize.io/sdks/python)
+- [Node.js](http://hindsight.vectorize.io/sdks/nodejs)
+- [REST API](https://hindsight.vectorize.io/api-reference)
+- [CLI](https://hindsight.vectorize.io/sdks/cli)
+
+**Community:**
+- [Slack](https://join.slack.com/t/hindsight-space/shared_invite/zt-3nhbm4w29-LeSJ5Ixi6j8PdiYOCPlOgg)
+- [GitHub Issues](https://github.com/vectorize-io/hindsight/issues)
+
+---
+## Star History
+
+[![Star History Chart](https://api.star-history.com/svg?repos=vectorize-io/hindsight&type=date&legend=top-left)](https://www.star-history.com/#vectorize-io/hindsight&type=date&legend=top-left)
+---
 
 ## Contributing
 
-We welcome contributions! See [CONTRIBUTING.md](./CONTRIBUTING.md) for guidelines.
+See [CONTRIBUTING.md](./CONTRIBUTING.md).
 
 ## License
 
-MIT
+MIT — see [LICENSE](./LICENSE)
+
+---
+
+Built by [Vectorize.io](https://vectorize.io)

docker/docker-compose/external-pg/docker-compose.yaml

@@ -0,0 +1,54 @@
+# Docker Compose file for Hindsight with PostgreSQL and pgvector
+#
+# 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)
+
+services:
+  db:
+    # Use a PostgreSQL-Image with pgvector extension pre-installed
+    # see https://hub.docker.com/r/pgvector/pgvector 
+    image: pgvector/pgvector:pg${HINDSIGHT_DB_VERSION:-18}
+    container_name: hindsight-db
+    restart: always
+    # Expose PostgreSQL port
+    # ports:
+      # - "5432:5432"
+    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
+
+  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}
+    depends_on:
+      - db
+    networks:
+      - hindsight-net
+
+networks:
+  hindsight-net:
+    driver: bridge
+
+volumes:
+  pg_data:

docker/docker-compose/nginx/README.md

@@ -0,0 +1,96 @@
+# Nginx Reverse Proxy with Custom Base Path
+
+Deploy Hindsight API under `/hindsight` (or any custom path) using Nginx reverse proxy.
+
+## Quick Start (Published Image - API Only)
+
+```bash
+docker-compose up
+```
+
+- **API:** http://localhost:8080/hindsight/docs
+- **Control Plane:** http://localhost:9999 (direct access, not proxied)
+
+## Full Stack with Custom Base Path (Requires Build)
+
+**Important:** You cannot rebuild from the published image with build args. You must build from source.
+
+### Build from Source with Custom Base Path
+
+1. **Clone the repository** (if you haven't):
+```bash
+git clone https://github.com/vectorize-io/hindsight.git
+cd hindsight
+```
+
+2. **Build with base path**:
+```bash
+docker build \
+  --build-arg NEXT_PUBLIC_BASE_PATH=/hindsight \
+  -f docker/standalone/Dockerfile \
+  -t hindsight:custom \
+  .
+```
+
+3. **Update docker-compose.yml** to use your built image:
+```yaml
+services:
+  hindsight:
+    image: hindsight:custom  # ← Change this
+    environment:
+      HINDSIGHT_API_BASE_PATH: /hindsight
+      NEXT_PUBLIC_BASE_PATH: /hindsight
+```
+
+4. **Update nginx.conf** to handle Control Plane routes (see below)
+
+5. **Run**:
+```bash
+docker-compose up
+```
+
+### Required nginx.conf for Full Stack
+
+Replace the current `nginx.conf` with this to proxy both API and Control Plane:
+
+```nginx
+events { worker_connections 1024; }
+
+http {
+    include /etc/nginx/mime.types;
+    default_type application/octet-stream;
+
+    upstream hindsight_api { server hindsight:8888; }
+    upstream hindsight_cp { server hindsight:9999; }
+
+    server {
+        listen 80;
+
+        # API
+        location ~ ^/hindsight/(docs|openapi\.json|health|metrics|v1|mcp) {
+            proxy_pass http://hindsight_api;
+            proxy_set_header Host $http_host;
+        }
+
+        # Control Plane static files
+        location ~ ^/hindsight/_next/ {
+            proxy_pass http://hindsight_cp;
+            proxy_set_header Host $http_host;
+        }
+
+        # Control Plane UI
+        location /hindsight {
+            proxy_pass http://hindsight_cp;
+            proxy_set_header Host $http_host;
+        }
+
+        location = / { return 301 /hindsight; }
+    }
+}
+```
+
+### Why Build is Required
+
+Next.js requires `basePath` at **build time**. The published image was built without a custom base path, so you must rebuild from source with the `NEXT_PUBLIC_BASE_PATH` build arg to deploy the Control Plane under a subpath.
+
+The API works without rebuild because `HINDSIGHT_API_BASE_PATH` is a runtime environment variable.

docker/docker-compose/nginx/docker-compose.yml

@@ -0,0 +1,88 @@
+# Hindsight API deployment with Nginx reverse proxy (API-only)
+#
+# This example deploys Hindsight API under the path /hindsight with:
+# - Hindsight standalone image (API + Control Plane + embedded pg0)
+# - Nginx reverse proxy (API only)
+#
+# Quick Start:
+#   docker-compose -f docker/docker-compose/nginx/docker-compose.yml up
+#
+# Access:
+#   API (via nginx):         http://localhost:8080/hindsight/docs
+#   Control Plane (direct):  http://localhost:9999
+#
+# For full stack deployment (API + Control Plane both under /hindsight):
+#   See README.md in this directory for instructions on building with basePath.
+#
+# Note: This configuration uses the published image (no build required).
+#       Control Plane is served directly because Next.js basePath requires
+#       build-time configuration. See README.md for the full stack option.
+
+services:
+  # Hindsight (API + Control Plane + embedded pg0)
+  hindsight:
+    image: ghcr.io/vectorize-io/hindsight:latest
+    ports:
+      - "9999:9999"  # Control Plane (direct access, not proxied)
+    environment:
+      # API base path for reverse proxy
+      HINDSIGHT_API_BASE_PATH: /hindsight
+
+      # LLM configuration
+      # Using mock provider for testing (no API key needed)
+      # For production, set OPENAI_API_KEY or ANTHROPIC_API_KEY and use a real provider
+      HINDSIGHT_API_LLM_PROVIDER: ${HINDSIGHT_API_LLM_PROVIDER:-mock}
+      HINDSIGHT_API_LLM_API_KEY: ${OPENAI_API_KEY:-not-needed-for-mock}
+      HINDSIGHT_API_LLM_MODEL: ${HINDSIGHT_API_LLM_MODEL:-mock-model}
+
+      # Production examples (uncomment and set appropriate API key):
+      # HINDSIGHT_API_LLM_PROVIDER: openai
+      # HINDSIGHT_API_LLM_API_KEY: ${OPENAI_API_KEY}
+      # HINDSIGHT_API_LLM_MODEL: gpt-4o-mini
+
+      # HINDSIGHT_API_LLM_PROVIDER: anthropic
+      # HINDSIGHT_API_LLM_API_KEY: ${ANTHROPIC_API_KEY}
+      # HINDSIGHT_API_LLM_MODEL: claude-sonnet-4-20250514
+
+      # Server config
+      HINDSIGHT_API_HOST: 0.0.0.0
+      HINDSIGHT_API_PORT: 8888
+      HINDSIGHT_API_LOG_LEVEL: info
+
+      # Control Plane config
+      HINDSIGHT_CP_DATAPLANE_API_URL: http://localhost:8888
+    volumes:
+      # Persist embedded pg0 database
+      - hindsight_data:/app/data
+    # Note: Ports not exposed - access via Nginx at localhost:8080/hindsight/
+    # To debug directly, uncomment these ports:
+    # ports:
+    #   - "8888:8888"  # API
+    #   - "9999:9999"  # Control Plane
+    healthcheck:
+      test: ["CMD", "curl", "-f", "http://localhost:8888/hindsight/health"]
+      interval: 10s
+      timeout: 5s
+      retries: 3
+      start_period: 30s
+    networks:
+      - hindsight
+
+  # Nginx reverse proxy
+  nginx:
+    image: nginx:alpine
+    ports:
+      - "8080:80"
+    volumes:
+      - ./nginx.conf:/etc/nginx/nginx.conf:ro
+    depends_on:
+      hindsight:
+        condition: service_healthy
+    networks:
+      - hindsight
+
+volumes:
+  hindsight_data:
+
+networks:
+  hindsight:

docker/docker-compose/nginx/nginx.conf

@@ -0,0 +1,40 @@
+# Nginx configuration for API-only reverse proxy
+# Control Plane accessed directly (not through nginx)
+
+events {
+    worker_connections 1024;
+}
+
+http {
+    include /etc/nginx/mime.types;
+    default_type application/octet-stream;
+
+    # Logging
+    access_log /var/log/nginx/access.log;
+    error_log /var/log/nginx/error.log;
+
+    # Upstream - Hindsight API
+    upstream hindsight_api {
+        server hindsight:8888;
+    }
+
+    server {
+        listen 80;
+        server_name _;
+
+        # API endpoints - forward with /hindsight prefix
+        location /hindsight/ {
+            proxy_pass http://hindsight_api;
+
+            proxy_set_header Host $http_host;
+            proxy_set_header X-Real-IP $remote_addr;
+            proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+            proxy_set_header X-Forwarded-Proto $scheme;
+        }
+
+        # Redirect root to API docs
+        location = / {
+            return 301 /hindsight/docs;
+        }
+    }
+}

docker/docker-compose/pg_textsearch/Dockerfile

@@ -0,0 +1,32 @@
+# 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

@@ -0,0 +1,91 @@
+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/vchord/docker-compose.yaml

@@ -0,0 +1,93 @@
+name: hindsight
+# Docker Compose file for Hindsight with PostgreSQL and vectorchord
+# docker compose -f docker/docker-compose/docker-compose.yaml down && sleep 2 && docker compose -f docker/docker-compose/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)
+# - HINDSIGHT_DB_VERSION: PostgreSQL version (default: 18)
+
+services:
+  db:
+    # Use a PostgreSQL-Image with vectorchord extension pre-installed
+    image: tensorchord/vchord-suite:pg${HINDSIGHT_DB_VERSION:-18-latest}
+    container_name: hindsight-db
+    restart: always
+    # Expose PostgreSQL port
+    ports:
+      - "5436:5432"
+    environment:
+      POSTGRES_USER: ${HINDSIGHT_DB_USER:-hindsight_user}
+      POSTGRES_PASSWORD: ${HINDSIGHT_DB_PASSWORD:-hindsight_password}
+      POSTGRES_DB: ${HINDSIGHT_DB_NAME:-hindsight_db}
+    volumes:
+      - pg_data:/var/lib/postgresql/${HINDSIGHT_DB_VERSION:-18}/docker
+    networks:
+      - hindsight-net
+
+  vectorchord-init:
+    image: tensorchord/vchord-suite:pg18-latest
+    #container_name: vectorchord-init
+    depends_on:
+      - db
+    environment:
+      - PGPASSWORD=${HINDSIGHT_DB_PASSWORD:-hindsight_password}
+    command: >
+      bash -c "
+        echo 'Waiting for PostgreSQL to be ready...';
+        until pg_isready -h hindsight-db -p 5432 -U hindsight_user; do
+          echo 'PostgreSQL is unavailable - sleeping';
+          sleep 2;
+        done;
+        echo 'PostgreSQL is ready - creating hindsight_db database';
+        psql -h hindsight-db -p 5432 -U hindsight_user -c 'CREATE DATABASE hindsight_db;' 2>/dev/null || echo 'Database already exists';
+        echo 'Creating extensions in hindsight_db database';
+        psql -h hindsight-db -p 5432 -U hindsight_user -d hindsight_db -c 'CREATE EXTENSION IF NOT EXISTS vchord CASCADE;';
+        psql -h hindsight-db -p 5432 -U hindsight_user -d hindsight_db -c 'CREATE EXTENSION IF NOT EXISTS pg_tokenizer CASCADE;';
+        psql -h hindsight-db -p 5432 -U hindsight_user -d hindsight_db -c 'CREATE EXTENSION IF NOT EXISTS vchord_bm25 CASCADE;';
+        echo 'Creating llmlingua2 tokenizer';
+        psql -h hindsight-db -p 5432 -U hindsight_user -d hindsight_db -c \"SELECT create_tokenizer('llmlingua2', \\$\\$ model = \\\"llmlingua2\\\" \\$\\$);\" 2>/dev/null || echo 'Tokenizer already exists or creation skipped';
+        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 (uses OpenAI for testing vchord)
+      # 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: vchord
+      HINDSIGHT_API_TEXT_SEARCH_EXTENSION: vchord
+
+    depends_on:
+      - db
+    networks:
+      - hindsight-net
+
+
+networks:
+  hindsight-net:
+    driver: bridge
+
+volumes:
+  pg_data:

docker/standalone/Dockerfile

@@ -2,16 +2,25 @@
 # Supports building API-only, Control Plane-only, or both
 #
 # Build args:
-#   INCLUDE_API=true/false  - Include API (default: true)
-#   INCLUDE_CP=true/false   - Include Control Plane (default: true)
+#   INCLUDE_API=true/false         - Include API (default: true)
+#   INCLUDE_CP=true/false          - Include Control Plane (default: true)
+#   INCLUDE_LOCAL_MODELS=true/false - Include local ML models for embeddings/reranking (default: true)
+#                                     Set to false when using external providers (TEI, OpenAI, Cohere)
+#   PRELOAD_ML_MODELS=true/false   - Pre-download ML models during build (default: true)
+#                                     Only effective when INCLUDE_LOCAL_MODELS=true
+#                                     NOTE: tiktoken encodings are ALWAYS preloaded (required for air-gapped deployments)
 #
 # Examples:
-#   docker build -t hindsight .                                    # Both (standalone)
-#   docker build -t hindsight-api --build-arg INCLUDE_CP=false .   # API only
-#   docker build -t hindsight-cp --build-arg INCLUDE_API=false .   # Control Plane only
+#   docker build -t hindsight .                                          # Both (standalone)
+#   docker build -t hindsight-api --build-arg INCLUDE_CP=false .         # API only
+#   docker build -t hindsight-cp --build-arg INCLUDE_API=false .         # Control Plane only
+#   docker build -t hindsight --build-arg PRELOAD_ML_MODELS=false .      # Skip ML model preload
+#   docker build -t hindsight --build-arg INCLUDE_LOCAL_MODELS=false .   # Skip local ML deps (for external providers)
 
 ARG INCLUDE_API=true
 ARG INCLUDE_CP=true
+ARG PRELOAD_ML_MODELS=true
+ARG INCLUDE_LOCAL_MODELS=true
 
 # =============================================================================
 # Stage: API Builder
@@ -19,6 +28,7 @@ ARG INCLUDE_CP=true
 FROM python:3.11-slim AS api-builder
 
 ARG INCLUDE_API
+ARG INCLUDE_LOCAL_MODELS
 RUN if [ "$INCLUDE_API" != "true" ]; then echo "Skipping API build" && exit 0; fi
 
 WORKDIR /app
@@ -37,12 +47,23 @@ COPY hindsight-api/README.md ./api/
 
 WORKDIR /app/api
 
+# 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 and alembic migrations
+# Copy source code (alembic migrations are inside hindsight_api/)
 COPY hindsight-api/hindsight_api ./hindsight_api
-COPY hindsight-api/alembic ./alembic
+
+# Install the local package (uv sync only installed dependencies, not the package itself)
+RUN uv pip install -e .
 
 # =============================================================================
 # Stage: SDK Builder (needed for Control Plane)
@@ -52,13 +73,15 @@ FROM node:20-slim AS sdk-builder
 ARG INCLUDE_CP
 RUN if [ "$INCLUDE_CP" != "true" ]; then echo "Skipping SDK build" && exit 0; fi
 
-WORKDIR /app/sdk
+WORKDIR /app
 
-COPY hindsight-clients/typescript/package*.json ./
-RUN npm ci
+# Copy root package files for npm workspaces
+COPY package.json package-lock.json ./
+COPY hindsight-clients/typescript/ ./hindsight-clients/typescript/
 
-COPY hindsight-clients/typescript/ ./
-RUN npm run build
+# Install and build SDK using workspace (--ignore-scripts skips git hooks setup)
+RUN npm ci --ignore-scripts -w @vectorize-io/hindsight-client
+RUN npm run build -w @vectorize-io/hindsight-client
 
 # =============================================================================
 # Stage: Control Plane Builder
@@ -68,30 +91,52 @@ FROM node:20-slim AS cp-builder
 ARG INCLUDE_CP
 RUN if [ "$INCLUDE_CP" != "true" ]; then echo "Skipping CP build" && exit 0; fi
 
-WORKDIR /app
-
-# Copy built SDK
-COPY --from=sdk-builder /app/sdk /app/sdk
+# Create directory structure matching the monorepo layout
+# This is required because build:standalone script expects .next/standalone/memory-poc/hindsight-control-plane
+WORKDIR /app/memory-poc/hindsight-control-plane
 
 # Install Control Plane dependencies
 # Only copy package.json (not package-lock.json) to ensure npm installs
 # correct platform-specific native bindings for lightningcss/tailwindcss
 COPY hindsight-control-plane/package.json ./
+# Remove the file: dependency on SDK (we'll copy it directly later)
+RUN sed -i '/"@vectorize-io\/hindsight-client":/d' package.json
 RUN npm install
 
 # Copy Control Plane source (excluding node_modules via .dockerignore)
 COPY hindsight-control-plane/ ./
 # Remove package-lock.json to avoid conflicts with installed native bindings
-RUN rm -f package-lock.json
+# Also remove the file: dependency from package.json (restored by COPY above)
+RUN rm -f package-lock.json && sed -i '/"@vectorize-io\/hindsight-client":/d' package.json
 
-# Link SDK (temporary for build)
-RUN cd /app/sdk && npm link && cd /app && npm link @vectorize-io/hindsight-client
+# Copy built SDK directly into node_modules (more reliable than npm link in Docker)
+COPY --from=sdk-builder /app/hindsight-clients/typescript ./node_modules/@vectorize-io/hindsight-client
 
-# Build Control Plane
-RUN npm run build
+# Accept base path as build argument for reverse proxy deployments
+# Usage: docker build --build-arg NEXT_PUBLIC_BASE_PATH=/hindsight ...
+ARG NEXT_PUBLIC_BASE_PATH=""
 
-# Create public directory if it doesn't exist
-RUN mkdir -p public
+# Build Control Plane - run next build first, then custom standalone copy
+# (The build:standalone script expects a specific path structure that differs in Docker)
+RUN npm exec -- next build
+
+# Create standalone directory structure manually
+# Note: Must exclude node_modules from find to avoid wrong server.js from next/dist/experimental/testmode/
+# Note: Must explicitly copy .next since glob * doesn't match hidden directories
+RUN STANDALONE_ROOT=$(find .next/standalone -path '*/node_modules' -prune -o -name 'server.js' -print | head -1 | xargs dirname) && \
+    mkdir -p standalone && \
+    cp -r "$STANDALONE_ROOT"/* standalone/ && \
+    cp -r "$STANDALONE_ROOT"/.next standalone/.next && \
+    # Copy node_modules if separate from app dir (monorepo structure)
+    if [ -d ".next/standalone/node_modules" ] && [ "$STANDALONE_ROOT" != ".next/standalone" ]; then \
+      cp -r .next/standalone/node_modules standalone/node_modules; \
+    fi && \
+    cp -r .next/static standalone/.next/static && \
+    mkdir -p standalone/public && \
+    cp -r public/* standalone/public/ 2>/dev/null || true && \
+    # Verify required files exist
+    test -f standalone/server.js || (echo "ERROR: server.js missing!" && exit 1) && \
+    test -f standalone/.next/BUILD_ID || (echo "ERROR: BUILD_ID missing!" && exit 1)
 
 # =============================================================================
 # Stage: Final Image - API Only
@@ -100,18 +145,18 @@ FROM python:3.11-slim AS api-only
 
 WORKDIR /app
 
-# Install pg0 dependencies
+# Note: libicu version varies by Debian version - try common versions in order
 RUN apt-get update && apt-get install -y \
     curl \
+    procps \
     libxml2 \
     libssl3 \
     libgssapi-krb5-2 \
     libossp-uuid16 \
-    && apt-get install -y libicu72 || apt-get install -y libicu74 || apt-get install -y libicu* \
+    && (apt-get install -y libicu72 2>/dev/null || apt-get install -y libicu74 2>/dev/null || apt-get install -y libicu76 2>/dev/null || true) \
     && rm -rf /var/lib/apt/lists/* \
     && pip install --no-cache-dir uv
 
-# Create non-root user (PostgreSQL cannot run as root)
 RUN useradd -m -s /bin/bash hindsight
 
 # Copy API with virtual environment from builder
@@ -121,52 +166,65 @@ COPY --from=api-builder /app/api /app/api
 COPY docker/standalone/start-all.sh /app/start-all.sh
 RUN chmod +x /app/start-all.sh
 
-# Create data directory for pg0 and set ownership
-RUN mkdir -p /app/data && chown -R hindsight:hindsight /app
+RUN chown -R hindsight:hindsight /app
 
-# Switch to non-root user
 USER hindsight
 
-# Set PATH for hindsight user
-ENV PATH="/home/hindsight/.hindsight/bin:/app/api/.venv/bin:${PATH}"
+ENV PATH="/app/api/.venv/bin:${PATH}"
 
-# Install pg0 binary
-RUN mkdir -p /home/hindsight/.hindsight/bin && \
-    ARCH=$(uname -m) && \
-    if [ "$ARCH" = "aarch64" ] || [ "$ARCH" = "arm64" ]; then \
-        PG0_BINARY="pg0-linux-aarch64-gnu"; \
-    elif [ "$ARCH" = "x86_64" ]; then \
-        PG0_BINARY="pg0-linux-x86_64-gnu"; \
-    else \
-        echo "Unsupported architecture: $ARCH" && exit 1; \
-    fi && \
-    echo "Installing pg0 binary: $PG0_BINARY" && \
-    for i in 1 2 3 4 5; do \
-        curl -fsSL -o /home/hindsight/.hindsight/bin/pg0 \
-            "https://github.com/vectorize-io/pg0/releases/latest/download/$PG0_BINARY" && \
-        chmod +x /home/hindsight/.hindsight/bin/pg0 && \
-        break || (echo "Retry $i failed, waiting..." && sleep 10); \
-    done && \
-    /home/hindsight/.hindsight/bin/pg0 --version
+# Pre-download tiktoken encoding (ALWAYS - required for token counting even in air-gapped envs)
+# Tiktoken is a core runtime dependency, not an optional ML model
+RUN MAX_RETRIES=3; \
+    RETRY_DELAY=5; \
+    for i in $(seq 1 $MAX_RETRIES); do \
+      echo "Attempt $i/$MAX_RETRIES: Downloading tiktoken encoding..."; \
+      /app/api/.venv/bin/python -c "\
+import tiktoken; \
+print('Downloading cl100k_base encoding...'); \
+tiktoken.get_encoding('cl100k_base'); \
+print('Tiktoken encoding cached successfully')" && break; \
+      if [ $i -lt $MAX_RETRIES ]; then \
+        echo "Attempt $i failed, retrying in ${RETRY_DELAY}s..."; \
+        sleep $RETRY_DELAY; \
+        RETRY_DELAY=$((RETRY_DELAY * 2)); \
+      fi; \
+    done; \
+    if [ $i -eq $MAX_RETRIES ]; then \
+      echo "ERROR: Failed to download tiktoken encoding after $MAX_RETRIES attempts"; \
+      exit 1; \
+    fi
 
-# Pre-download PostgreSQL binaries
-ENV PG0_HOME=/home/hindsight/.pg0-cache
-RUN pg0 start --help && \
-    (pg0 start --name hindsight --port 5555 --username hindsight --password hindsight --database hindsight && \
-    sleep 2 && \
-    pg0 stop --name hindsight && \
-    echo "PostgreSQL pre-cached to $PG0_HOME") || echo "Pre-download skipped"
-
-ENV PG0_HOME=/home/hindsight/.pg0
-
-# Pre-download ML models to avoid runtime download
-RUN /app/api/.venv/bin/python -c "\
+# Pre-download ML models to avoid runtime download (conditional)
+# Only runs if both PRELOAD_ML_MODELS=true AND INCLUDE_LOCAL_MODELS=true
+# Includes retry logic with exponential backoff for transient network failures
+ARG PRELOAD_ML_MODELS
+ARG INCLUDE_LOCAL_MODELS
+ENV HF_HUB_DOWNLOAD_TIMEOUT=600
+RUN if [ "$PRELOAD_ML_MODELS" = "true" ] && [ "$INCLUDE_LOCAL_MODELS" = "true" ]; then \
+    MAX_RETRIES=3; \
+    RETRY_DELAY=10; \
+    for i in $(seq 1 $MAX_RETRIES); do \
+      echo "Attempt $i/$MAX_RETRIES: Downloading ML models..."; \
+      /app/api/.venv/bin/python -c "\
+import os; os.environ['HF_HUB_DOWNLOAD_TIMEOUT'] = '600'; \
 from sentence_transformers import SentenceTransformer, CrossEncoder; \
 print('Downloading embedding model...'); \
 SentenceTransformer('BAAI/bge-small-en-v1.5'); \
 print('Downloading cross-encoder model...'); \
 CrossEncoder('cross-encoder/ms-marco-MiniLM-L-6-v2'); \
-print('Models cached successfully')"
+print('Models cached successfully')" && break; \
+      if [ $i -lt $MAX_RETRIES ]; then \
+        echo "Attempt $i failed, retrying in ${RETRY_DELAY}s..."; \
+        sleep $RETRY_DELAY; \
+        RETRY_DELAY=$((RETRY_DELAY * 2)); \
+      fi; \
+    done; \
+    if [ $i -eq $MAX_RETRIES ] && ! /app/api/.venv/bin/python -c "from sentence_transformers import SentenceTransformer; SentenceTransformer('BAAI/bge-small-en-v1.5')" 2>/dev/null; then \
+      echo "ERROR: Failed to download models after $MAX_RETRIES attempts"; \
+      exit 1; \
+    fi; \
+    elif [ "$INCLUDE_LOCAL_MODELS" != "true" ]; then echo "Skipping ML model preload (local-models not included)"; \
+    else echo "Skipping ML model preload"; fi
 
 EXPOSE 8888
 
@@ -175,6 +233,11 @@ ENV HINDSIGHT_API_PORT=8888
 ENV HINDSIGHT_API_LOG_LEVEL=info
 ENV HINDSIGHT_ENABLE_API=true
 ENV HINDSIGHT_ENABLE_CP=false
+ENV PYTHONUNBUFFERED=1
+# Suppress verbose transformers/HuggingFace model loading warnings
+ENV TRANSFORMERS_VERBOSITY=error
+ENV HF_HUB_VERBOSITY=error
+ENV TOKENIZERS_PARALLELISM=false
 
 CMD ["/app/start-all.sh"]
 
@@ -186,13 +249,13 @@ FROM node:20-alpine AS cp-only
 WORKDIR /app
 
 # Copy built SDK
-COPY --from=sdk-builder /app/sdk /app/sdk
+COPY --from=sdk-builder /app/hindsight-clients/typescript /app/sdk
 
 # Copy Control Plane standalone build
 WORKDIR /app/control-plane
-COPY --from=cp-builder /app/.next/standalone ./
-COPY --from=cp-builder /app/.next/static ./.next/static
-COPY --from=cp-builder /app/public ./public
+COPY --from=cp-builder /app/memory-poc/hindsight-control-plane/standalone ./
+COPY --from=cp-builder /app/memory-poc/hindsight-control-plane/.next/static ./.next/static
+COPY --from=cp-builder /app/memory-poc/hindsight-control-plane/public ./public
 
 WORKDIR /app
 
@@ -219,33 +282,34 @@ FROM python:3.11-slim AS standalone
 
 WORKDIR /app
 
-# Install Node.js, curl, uv, and pg0 dependencies
+# Install Node.js, curl, uv, and system dependencies
+# Note: libicu version varies by Debian version - try common versions in order
 RUN apt-get update && apt-get install -y \
     curl \
+    procps \
     libxml2 \
     libssl3 \
     libgssapi-krb5-2 \
     libossp-uuid16 \
-    && apt-get install -y libicu72 || apt-get install -y libicu74 || apt-get install -y libicu* \
+    && (apt-get install -y libicu72 2>/dev/null || apt-get install -y libicu74 2>/dev/null || apt-get install -y libicu76 2>/dev/null || true) \
     && curl -fsSL https://deb.nodesource.com/setup_20.x | bash - \
     && apt-get install -y nodejs \
     && rm -rf /var/lib/apt/lists/* \
     && pip install --no-cache-dir uv
 
-# Create non-root user (PostgreSQL cannot run as root)
 RUN useradd -m -s /bin/bash hindsight
 
 # Copy API with virtual environment from builder
 COPY --from=api-builder /app/api /app/api
 
 # Copy built SDK
-COPY --from=sdk-builder /app/sdk /app/sdk
+COPY --from=sdk-builder /app/hindsight-clients/typescript /app/sdk
 
 # Copy Control Plane standalone build
 WORKDIR /app/control-plane
-COPY --from=cp-builder /app/.next/standalone ./
-COPY --from=cp-builder /app/.next/static ./.next/static
-COPY --from=cp-builder /app/public ./public
+COPY --from=cp-builder /app/memory-poc/hindsight-control-plane/standalone ./
+COPY --from=cp-builder /app/memory-poc/hindsight-control-plane/.next/static ./.next/static
+COPY --from=cp-builder /app/memory-poc/hindsight-control-plane/public ./public
 
 WORKDIR /app
 
@@ -253,52 +317,65 @@ WORKDIR /app
 COPY docker/standalone/start-all.sh /app/start-all.sh
 RUN chmod +x /app/start-all.sh
 
-# Create data directory for pg0 and set ownership
-RUN mkdir -p /app/data && chown -R hindsight:hindsight /app
+RUN chown -R hindsight:hindsight /app
 
-# Switch to non-root user
 USER hindsight
 
-# Set PATH for hindsight user
-ENV PATH="/home/hindsight/.hindsight/bin:/app/api/.venv/bin:${PATH}"
+ENV PATH="/app/api/.venv/bin:${PATH}"
 
-# Install pg0 binary
-RUN mkdir -p /home/hindsight/.hindsight/bin && \
-    ARCH=$(uname -m) && \
-    if [ "$ARCH" = "aarch64" ] || [ "$ARCH" = "arm64" ]; then \
-        PG0_BINARY="pg0-linux-aarch64-gnu"; \
-    elif [ "$ARCH" = "x86_64" ]; then \
-        PG0_BINARY="pg0-linux-x86_64-gnu"; \
-    else \
-        echo "Unsupported architecture: $ARCH" && exit 1; \
-    fi && \
-    echo "Installing pg0 binary: $PG0_BINARY" && \
-    for i in 1 2 3 4 5; do \
-        curl -fsSL -o /home/hindsight/.hindsight/bin/pg0 \
-            "https://github.com/vectorize-io/pg0/releases/latest/download/$PG0_BINARY" && \
-        chmod +x /home/hindsight/.hindsight/bin/pg0 && \
-        break || (echo "Retry $i failed, waiting..." && sleep 10); \
-    done && \
-    /home/hindsight/.hindsight/bin/pg0 --version
+# Pre-download tiktoken encoding (ALWAYS - required for token counting even in air-gapped envs)
+# Tiktoken is a core runtime dependency, not an optional ML model
+RUN MAX_RETRIES=3; \
+    RETRY_DELAY=5; \
+    for i in $(seq 1 $MAX_RETRIES); do \
+      echo "Attempt $i/$MAX_RETRIES: Downloading tiktoken encoding..."; \
+      /app/api/.venv/bin/python -c "\
+import tiktoken; \
+print('Downloading cl100k_base encoding...'); \
+tiktoken.get_encoding('cl100k_base'); \
+print('Tiktoken encoding cached successfully')" && break; \
+      if [ $i -lt $MAX_RETRIES ]; then \
+        echo "Attempt $i failed, retrying in ${RETRY_DELAY}s..."; \
+        sleep $RETRY_DELAY; \
+        RETRY_DELAY=$((RETRY_DELAY * 2)); \
+      fi; \
+    done; \
+    if [ $i -eq $MAX_RETRIES ]; then \
+      echo "ERROR: Failed to download tiktoken encoding after $MAX_RETRIES attempts"; \
+      exit 1; \
+    fi
 
-# Pre-download PostgreSQL binaries
-ENV PG0_HOME=/home/hindsight/.pg0-cache
-RUN pg0 start --help && \
-    (pg0 start --name hindsight --port 5555 --username hindsight --password hindsight --database hindsight && \
-    sleep 2 && \
-    pg0 stop --name hindsight && \
-    echo "PostgreSQL pre-cached to $PG0_HOME") || echo "Pre-download skipped"
-
-ENV PG0_HOME=/home/hindsight/.pg0
-
-# Pre-download ML models to avoid runtime download
-RUN /app/api/.venv/bin/python -c "\
+# Pre-download ML models to avoid runtime download (conditional)
+# Only runs if both PRELOAD_ML_MODELS=true AND INCLUDE_LOCAL_MODELS=true
+# Includes retry logic with exponential backoff for transient network failures
+ARG PRELOAD_ML_MODELS
+ARG INCLUDE_LOCAL_MODELS
+ENV HF_HUB_DOWNLOAD_TIMEOUT=600
+RUN if [ "$PRELOAD_ML_MODELS" = "true" ] && [ "$INCLUDE_LOCAL_MODELS" = "true" ]; then \
+    MAX_RETRIES=3; \
+    RETRY_DELAY=10; \
+    for i in $(seq 1 $MAX_RETRIES); do \
+      echo "Attempt $i/$MAX_RETRIES: Downloading ML models..."; \
+      /app/api/.venv/bin/python -c "\
+import os; os.environ['HF_HUB_DOWNLOAD_TIMEOUT'] = '600'; \
 from sentence_transformers import SentenceTransformer, CrossEncoder; \
 print('Downloading embedding model...'); \
 SentenceTransformer('BAAI/bge-small-en-v1.5'); \
 print('Downloading cross-encoder model...'); \
 CrossEncoder('cross-encoder/ms-marco-MiniLM-L-6-v2'); \
-print('Models cached successfully')"
+print('Models cached successfully')" && break; \
+      if [ $i -lt $MAX_RETRIES ]; then \
+        echo "Attempt $i failed, retrying in ${RETRY_DELAY}s..."; \
+        sleep $RETRY_DELAY; \
+        RETRY_DELAY=$((RETRY_DELAY * 2)); \
+      fi; \
+    done; \
+    if [ $i -eq $MAX_RETRIES ] && ! /app/api/.venv/bin/python -c "from sentence_transformers import SentenceTransformer; SentenceTransformer('BAAI/bge-small-en-v1.5')" 2>/dev/null; then \
+      echo "ERROR: Failed to download models after $MAX_RETRIES attempts"; \
+      exit 1; \
+    fi; \
+    elif [ "$INCLUDE_LOCAL_MODELS" != "true" ]; then echo "Skipping ML model preload (local-models not included)"; \
+    else echo "Skipping ML model preload"; fi
 
 EXPOSE 8888 9999
 
@@ -309,6 +386,11 @@ ENV NODE_ENV=production
 ENV HINDSIGHT_CP_DATAPLANE_API_URL=http://localhost:8888
 ENV HINDSIGHT_ENABLE_API=true
 ENV HINDSIGHT_ENABLE_CP=true
+ENV PYTHONUNBUFFERED=1
+# Suppress verbose transformers/HuggingFace model loading warnings
+ENV TRANSFORMERS_VERBOSITY=error
+ENV HF_HUB_VERBOSITY=error
+ENV TOKENIZERS_PARALLELISM=false
 
 CMD ["/app/start-all.sh"]
 

docker/standalone/start-all.sh

@@ -1,23 +1,74 @@
 #!/bin/bash
 set -e
 
-echo "🚀 Starting Hindsight..."
-echo ""
-
 # Service flags (default to true if not set)
 ENABLE_API="${HINDSIGHT_ENABLE_API:-true}"
 ENABLE_CP="${HINDSIGHT_ENABLE_CP:-true}"
 
-# Copy pre-cached PostgreSQL data if runtime directory is empty (first run with volume)
-if [ "$ENABLE_API" = "true" ]; then
-    PG0_CACHE="/home/hindsight/.pg0-cache"
-    PG0_HOME="/home/hindsight/.pg0"
-    if [ -d "$PG0_CACHE" ] && [ "$(ls -A $PG0_CACHE 2>/dev/null)" ]; then
-        if [ ! "$(ls -A $PG0_HOME 2>/dev/null)" ]; then
-            echo "📦 Copying pre-cached PostgreSQL data..."
-            cp -r "$PG0_CACHE"/* "$PG0_HOME"/ 2>/dev/null || true
-        fi
+# =============================================================================
+# Dependency waiting (opt-in via HINDSIGHT_WAIT_FOR_DEPS=true)
+#
+# Problem: When running with LM Studio, the LLM may take time to load models.
+# If Hindsight starts before LM Studio is ready, it fails on LLM verification.
+# This wait loop ensures dependencies are ready before starting.
+# =============================================================================
+if [ "${HINDSIGHT_WAIT_FOR_DEPS:-false}" = "true" ]; then
+    LLM_BASE_URL="${HINDSIGHT_API_LLM_BASE_URL:-http://host.docker.internal:1234/v1}"
+    MAX_RETRIES="${HINDSIGHT_RETRY_MAX:-0}"  # 0 = infinite
+    RETRY_INTERVAL="${HINDSIGHT_RETRY_INTERVAL:-10}"
+
+    # Check if external database is configured (skip check for embedded pg0)
+    SKIP_DB_CHECK=false
+    if [ -z "${HINDSIGHT_API_DATABASE_URL}" ]; then
+        SKIP_DB_CHECK=true
+    else
+        DB_CHECK_HOST=$(echo "$HINDSIGHT_API_DATABASE_URL" | sed -E 's|.*@([^:/]+):([0-9]+)/.*|\1 \2|')
     fi
+
+    check_db() {
+        if $SKIP_DB_CHECK; then
+            return 0
+        fi
+        if command -v pg_isready &> /dev/null; then
+            pg_isready -h $(echo $DB_CHECK_HOST | cut -d' ' -f1) -p $(echo $DB_CHECK_HOST | cut -d' ' -f2) &>/dev/null
+        else
+            python3 -c "import socket; s=socket.socket(); s.settimeout(5); exit(0 if s.connect_ex(('$(echo $DB_CHECK_HOST | cut -d' ' -f1)', $(echo $DB_CHECK_HOST | cut -d' ' -f2))) == 0 else 1)" 2>/dev/null
+        fi
+    }
+
+    check_llm() {
+        curl -sf "${LLM_BASE_URL}/models" --connect-timeout 5 &>/dev/null
+    }
+
+    echo "⏳ Waiting for dependencies to be ready..."
+    attempt=1
+
+    while true; do
+        db_ok=false
+        llm_ok=false
+
+        if check_db; then
+            db_ok=true
+        fi
+
+        if check_llm; then
+            llm_ok=true
+        fi
+
+        if $db_ok && $llm_ok; then
+            echo "✅ Dependencies ready!"
+            break
+        fi
+
+        if [ "$MAX_RETRIES" -ne 0 ] && [ "$attempt" -ge "$MAX_RETRIES" ]; then
+            echo "❌ Max retries ($MAX_RETRIES) reached. Dependencies not available."
+            exit 1
+        fi
+
+        echo "   Attempt $attempt: DB=$( $db_ok && echo 'ok' || echo 'waiting' ), LLM=$( $llm_ok && echo 'ok' || echo 'waiting' )"
+        sleep "$RETRY_INTERVAL"
+        ((attempt++))
+    done
 fi
 
 # Track PIDs for wait
@@ -26,32 +77,31 @@ PIDS=()
 # Start API if enabled
 if [ "$ENABLE_API" = "true" ]; then
     cd /app/api
-    python -m hindsight_api.web.server 2>&1 | sed -u 's/^/[api] /' &
+    # Run API directly - Python's PYTHONUNBUFFERED=1 handles output buffering
+    hindsight-api &
     API_PID=$!
     PIDS+=($API_PID)
 
     # Wait for API to be ready
-    echo "⏳ Waiting for API..."
     for i in {1..60}; do
         if curl -sf http://localhost:8888/health &>/dev/null; then
-            echo "✅ API is ready"
             break
         fi
         sleep 1
     done
 else
-    echo "⏭️  API disabled (HINDSIGHT_ENABLE_API=false)"
+    echo "API disabled (HINDSIGHT_ENABLE_API=false)"
 fi
 
 # Start Control Plane if enabled
 if [ "$ENABLE_CP" = "true" ]; then
     echo "🎛️  Starting Control Plane..."
     cd /app/control-plane
-    PORT=9999 node server.js 2>&1 | grep -v -E "^[[:space:]]*(▲|✓|-|$)" | sed -u 's/^/[control-plane] /' &
+    PORT=9999 node server.js &
     CP_PID=$!
     PIDS+=($CP_PID)
 else
-    echo "⏭️  Control Plane disabled (HINDSIGHT_ENABLE_CP=false)"
+    echo "Control Plane disabled (HINDSIGHT_ENABLE_CP=false)"
 fi
 
 # Print status

docker/test-image.sh

@@ -0,0 +1,201 @@
+#!/bin/bash
+#
+# Docker Smoke Test Script
+#
+# Tests that a Hindsight Docker image starts correctly and becomes healthy.
+# Can be run locally or in CI pipelines.
+#
+# Usage:
+#   ./docker/test-image.sh <image> [target]
+#
+# Arguments:
+#   image   - Docker image to test (e.g., hindsight-api:test, ghcr.io/vectorize-io/hindsight:latest)
+#   target  - Optional: 'cp-only' for control plane, otherwise assumes API image (default: api)
+#
+# Environment variables:
+#   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)
+#   HINDSIGHT_API_COHERE_API_KEY                - Cohere API key for reranking (optional)
+#   SMOKE_TEST_TIMEOUT                          - Timeout in seconds (default: 120)
+#   SMOKE_TEST_CONTAINER_NAME                   - Container name (default: hindsight-smoke-test)
+#
+# Examples:
+#   # Test a locally built full image
+#   ./docker/test-image.sh hindsight-api:test
+#
+#   # Test a released image
+#   ./docker/test-image.sh ghcr.io/vectorize-io/hindsight:latest
+#
+#   # Test control plane image
+#   ./docker/test-image.sh hindsight-control-plane:test cp-only
+#
+#   # Test slim image with external providers
+#   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
+#   export HINDSIGHT_API_COHERE_API_KEY=xxx
+#   ./docker/test-image.sh hindsight-slim:test
+#
+# Exit codes:
+#   0 - Success (container healthy)
+#   1 - Failure (container not healthy within timeout)
+#   2 - Invalid arguments
+#
+
+set -euo pipefail
+
+# Colors for output
+RED='\033[0;31m'
+GREEN='\033[0;32m'
+YELLOW='\033[0;33m'
+NC='\033[0m' # No Color
+
+# Configuration
+IMAGE="${1:-}"
+TARGET="${2:-api}"
+TIMEOUT="${SMOKE_TEST_TIMEOUT:-120}"
+CONTAINER_NAME="${SMOKE_TEST_CONTAINER_NAME:-hindsight-smoke-test}"
+LLM_PROVIDER="${HINDSIGHT_API_LLM_PROVIDER:-groq}"
+LLM_MODEL="${HINDSIGHT_API_LLM_MODEL:-llama-3.3-70b-versatile}"
+
+# Validate arguments
+if [ -z "$IMAGE" ]; then
+    echo -e "${RED}Error: Image argument is required${NC}"
+    echo ""
+    echo "Usage: $0 <image> [target]"
+    echo ""
+    echo "Examples:"
+    echo "  $0 hindsight-api:test"
+    echo "  $0 ghcr.io/vectorize-io/hindsight:latest"
+    echo "  $0 hindsight-control-plane:test cp-only"
+    exit 2
+fi
+
+# Determine health endpoint based on target
+if [ "$TARGET" = "cp-only" ]; then
+    HEALTH_PORT=9999
+    HEALTH_PATH="/api/health"
+    NEEDS_LLM=false
+else
+    HEALTH_PORT=8888
+    HEALTH_PATH="/health"
+    NEEDS_LLM=true
+fi
+
+# Check for required environment variables
+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
+
+# Cleanup function
+cleanup() {
+    echo "Cleaning up..."
+    docker stop "$CONTAINER_NAME" 2>/dev/null || true
+    docker rm "$CONTAINER_NAME" 2>/dev/null || true
+}
+
+# Set trap to cleanup on exit
+trap cleanup EXIT
+
+echo -e "${YELLOW}Starting smoke test for: ${IMAGE}${NC}"
+echo "  Target: $TARGET"
+echo "  Health endpoint: http://localhost:${HEALTH_PORT}${HEALTH_PATH}"
+echo "  Timeout: ${TIMEOUT}s"
+echo ""
+
+# Remove any existing container with the same name
+docker rm -f "$CONTAINER_NAME" 2>/dev/null || true
+
+# Start container based on target type
+echo "Starting container..."
+if [ "$TARGET" = "cp-only" ]; then
+    docker run -d --name "$CONTAINER_NAME" \
+        -p "${HEALTH_PORT}:${HEALTH_PORT}" \
+        "$IMAGE"
+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=$LLM_PROVIDER"
+    DOCKER_CMD="$DOCKER_CMD -e HINDSIGHT_API_LLM_API_KEY=${GROQ_API_KEY}"
+    DOCKER_CMD="$DOCKER_CMD -e HINDSIGHT_API_LLM_MODEL=$LLM_MODEL"
+
+    # Add optional embeddings provider config
+    if [ -n "${HINDSIGHT_API_EMBEDDINGS_PROVIDER:-}" ]; then
+        DOCKER_CMD="$DOCKER_CMD -e HINDSIGHT_API_EMBEDDINGS_PROVIDER=${HINDSIGHT_API_EMBEDDINGS_PROVIDER}"
+    fi
+    if [ -n "${HINDSIGHT_API_EMBEDDINGS_OPENAI_API_KEY:-}" ]; then
+        DOCKER_CMD="$DOCKER_CMD -e HINDSIGHT_API_EMBEDDINGS_OPENAI_API_KEY=${HINDSIGHT_API_EMBEDDINGS_OPENAI_API_KEY}"
+    fi
+
+    # Add optional reranker provider config
+    if [ -n "${HINDSIGHT_API_RERANKER_PROVIDER:-}" ]; then
+        DOCKER_CMD="$DOCKER_CMD -e HINDSIGHT_API_RERANKER_PROVIDER=${HINDSIGHT_API_RERANKER_PROVIDER}"
+    fi
+    if [ -n "${HINDSIGHT_API_COHERE_API_KEY:-}" ]; then
+        DOCKER_CMD="$DOCKER_CMD -e HINDSIGHT_API_COHERE_API_KEY=${HINDSIGHT_API_COHERE_API_KEY}"
+    fi
+
+    DOCKER_CMD="$DOCKER_CMD -p ${HEALTH_PORT}:${HEALTH_PORT}"
+    DOCKER_CMD="$DOCKER_CMD $IMAGE"
+
+    eval $DOCKER_CMD
+fi
+
+# Wait for health endpoint
+echo "Waiting for health endpoint at http://localhost:${HEALTH_PORT}${HEALTH_PATH}..."
+start_time=$(date +%s)
+
+for i in $(seq 1 "$TIMEOUT"); do
+    if curl -sf "http://localhost:${HEALTH_PORT}${HEALTH_PATH}" > /dev/null 2>&1; then
+        end_time=$(date +%s)
+        duration=$((end_time - start_time))
+        echo ""
+        echo -e "${GREEN}Container is healthy after ${duration}s${NC}"
+        echo ""
+        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 ""
+        echo ""
+        echo "=== Container Logs (last 50 lines) ==="
+        docker logs "$CONTAINER_NAME" 2>&1 | tail -50
+        echo ""
+        echo -e "${GREEN}Smoke test PASSED${NC}"
+        exit 0
+    fi
+
+    # Show progress every 10 seconds
+    if [ $((i % 10)) -eq 0 ]; then
+        echo "  Still waiting... (${i}s)"
+    fi
+
+    # Check if container is still running
+    if ! docker ps -q -f "name=$CONTAINER_NAME" | grep -q .; then
+        echo ""
+        echo -e "${RED}Container exited unexpectedly!${NC}"
+        echo ""
+        echo "=== Container Logs ==="
+        docker logs "$CONTAINER_NAME" 2>&1
+        echo ""
+        echo -e "${RED}Smoke test FAILED${NC}"
+        exit 1
+    fi
+
+    sleep 1
+done
+
+# Timeout reached
+echo ""
+echo -e "${RED}Container failed to become healthy after ${TIMEOUT}s${NC}"
+echo ""
+echo "=== Container Logs ==="
+docker logs "$CONTAINER_NAME" 2>&1
+echo ""
+echo -e "${RED}Smoke test FAILED${NC}"
+exit 1

docker/test-slim-local.sh

@@ -0,0 +1,51 @@
+#!/bin/bash
+#
+# Local Test Script for Slim Docker Images
+#
+# This script makes it easy to test slim images locally with external providers.
+# 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:
+#   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"
+    exit 1
+fi
+
+if [ -z "${COHERE_API_KEY:-}" ]; then
+    echo "❌ Error: COHERE_API_KEY environment variable is required"
+    echo "Set it with: export COHERE_API_KEY=xxx"
+    exit 1
+fi
+
+# Configuration
+IMAGE="${1:-hindsight-slim:test}"
+echo "Testing image: $IMAGE"
+echo ""
+
+# Set up external providers
+export HINDSIGHT_API_EMBEDDINGS_PROVIDER=openai
+export HINDSIGHT_API_EMBEDDINGS_OPENAI_API_KEY=$OPENAI_API_KEY
+export HINDSIGHT_API_RERANKER_PROVIDER=cohere
+export HINDSIGHT_API_COHERE_API_KEY=$COHERE_API_KEY
+
+# Run the test
+exec "$(dirname "$0")/test-image.sh" "$IMAGE" standalone

helm/INSTALL.txt

@@ -1,135 +0,0 @@
-HINDSIGHT HELM CHART INSTALLATION GUIDE
-=====================================
-
-PREREQUISITES
--------------
-- Kubernetes cluster (1.19+)
-- kubectl configured
-- Helm 3.x installed
-- PostgreSQL database with pgvector extension (if not using bundled PostgreSQL)
-
-BASIC INSTALLATION
-------------------
-
-1. Install with default values (requires external PostgreSQL):
-
-   helm install hindsight ./hindsight \
-     --set postgresql.external.host=your-postgres-host \
-     --set postgresql.external.password=your-password \
-     --set api.secrets.MEMORY_LLM_API_KEY=your-api-key
-
-2. Install with custom values file:
-
-   helm install hindsight ./hindsight -f hindsight/values-production.yaml
-
-3. Install in a specific namespace:
-
-   kubectl create namespace hindsight
-   helm install hindsight ./hindsight -n hindsight
-
-CONFIGURATION OPTIONS
----------------------
-
-Development setup (using values-development.yaml):
-   helm install hindsight ./hindsight -f hindsight/values-development.yaml
-
-Production setup (using values-production.yaml):
-   helm install hindsight ./hindsight -f hindsight/values-production.yaml
-
-Custom LLM provider:
-   helm install hindsight ./hindsight \
-     --set api.env.MEMORY_LLM_PROVIDER=openai \
-     --set api.env.MEMORY_LLM_MODEL=gpt-4 \
-     --set api.secrets.MEMORY_LLM_API_KEY=sk-your-key
-
-Enable ingress:
-   helm install hindsight ./hindsight \
-     --set ingress.enabled=true \
-     --set ingress.hosts[0].host=hindsight.example.com
-
-Enable autoscaling:
-   helm install hindsight ./hindsight \
-     --set autoscaling.enabled=true \
-     --set autoscaling.minReplicas=2 \
-     --set autoscaling.maxReplicas=10
-
-UPGRADE
--------
-
-Upgrade existing installation:
-   helm upgrade hindsight ./hindsight
-
-Upgrade with new values:
-   helm upgrade hindsight ./hindsight -f hindsight/values-production.yaml
-
-UNINSTALL
----------
-
-Remove the Helm release:
-   helm uninstall hindsight
-
-Remove with namespace:
-   helm uninstall hindsight -n hindsight
-
-TESTING
--------
-
-Test the installation with dry-run:
-   helm install hindsight ./hindsight --dry-run --debug
-
-Validate templates:
-   helm template hindsight ./hindsight
-
-Lint the chart:
-   helm lint ./hindsight
-
-ACCESSING THE SERVICES
-----------------------
-
-Port-forward control plane:
-   kubectl port-forward svc/hindsight-control-plane 3000:3000
-
-Port-forward API:
-   kubectl port-forward svc/hindsight-api 8888:8888
-
-Get service URLs:
-   helm status hindsight
-
-DATABASE INITIALIZATION
------------------------
-
-NOTE: Database migrations now run automatically when the API service starts.
-      You typically don't need to run migrations manually.
-
-If you want to pre-initialize the database before deploying (optional):
-   kubectl run hindsight-init --rm -it --restart=Never \
-     --image=hindsight/api:latest \
-     --env="DATABASE_URL=postgresql://user:pass@host:5432/hindsight" \
-     -- python -c "from hindsight.migrations import run_migrations; run_migrations()"
-
-TROUBLESHOOTING
----------------
-
-Check pod status:
-   kubectl get pods -l app.kubernetes.io/name=hindsight
-
-View logs for API:
-   kubectl logs -l app.kubernetes.io/component=api
-
-View logs for control plane:
-   kubectl logs -l app.kubernetes.io/component=control-plane
-
-Describe a pod:
-   kubectl describe pod <pod-name>
-
-Check configuration:
-   kubectl get configmap hindsight-config -o yaml
-   kubectl get secret hindsight-secret -o yaml
-
-NOTES
------
-- Make sure PostgreSQL has pgvector extension enabled
-- Run database migrations before first use
-- Configure proper resource limits for production
-- Use external secrets management for production
-- Enable TLS/SSL for production deployments

helm/hindsight/Chart.lock

@@ -0,0 +1,6 @@
+dependencies:
+- name: postgresql
+  repository: https://charts.bitnami.com/bitnami
+  version: 15.5.38
+digest: sha256:f67c7612736803ece8a669f8ca6b0555f3b78557bc0ecb732aa2e43f0df7750d
+generated: "2025-12-10T17:20:57.058794+01:00"

helm/hindsight/Chart.yaml

@@ -1,9 +1,9 @@
 apiVersion: v2
 name: hindsight
-description: A Helm chart for Hindsight - temporal-semantic-entity memory system for AI agents
+description: Hindsight helm chart
 type: application
-version: 0.0.21
-appVersion: "0.0.21"
+version: 0.4.11
+appVersion: "0.4.11"
 keywords:
   - ai
   - memory

helm/hindsight/README.md

@@ -0,0 +1,182 @@
+# Hindsight Helm Chart
+
+Helm chart for deploying Hindsight - a temporal-semantic-entity memory system for AI agents.
+
+## Prerequisites
+
+- Kubernetes 1.19+
+- Helm 3.0+
+- PostgreSQL database (external or bundled)
+
+## Quick Start
+
+```bash
+# Update dependencies first
+helm dependency update ./helm/hindsight
+
+# Install (PostgreSQL included by default)
+export OPENAI_API_KEY="sk-your-openai-key"
+helm upgrade hindsight --install ./helm/hindsight -n hindsight --create-namespace \
+  --set api.secrets.HINDSIGHT_API_LLM_API_KEY="$OPENAI_API_KEY"
+```
+
+To use an external database instead:
+
+```bash
+helm install hindsight ./helm/hindsight -n hindsight --create-namespace \
+  --set api.secrets.HINDSIGHT_API_LLM_API_KEY="sk-your-openai-key" \
+  --set postgresql.enabled=false \
+  --set postgresql.external.host=my-postgres.example.com \
+  --set postgresql.external.password=mypassword
+```
+
+## Installation
+
+### Add the repository (if published)
+
+```bash
+helm repo add hindsight https://your-helm-repo.com
+helm repo update
+```
+
+### Install with custom values file
+
+Create a `values-override.yaml`:
+
+```yaml
+api:
+  secrets:
+    HINDSIGHT_API_LLM_API_KEY: "sk-your-openai-key"
+
+postgresql:
+  external:
+    host: "my-postgres.example.com"
+    password: "mypassword"
+```
+
+Then install:
+
+```bash
+helm install hindsight ./helm/hindsight -n hindsight --create-namespace -f values-override.yaml
+```
+
+## Configuration
+
+### Key Values
+
+| Parameter | Description | Default |
+|-----------|-------------|---------|
+| `version` | Default image tag for all components | `0.1.0` |
+| `api.enabled` | Enable the API component | `true` |
+| `api.image.repository` | API image repository | `hindsight/api` |
+| `api.image.tag` | API image tag (defaults to `version`) | - |
+| `api.service.port` | API service port | `8888` |
+| `controlPlane.enabled` | Enable the control plane | `true` |
+| `controlPlane.image.repository` | Control plane image repository | `hindsight/control-plane` |
+| `controlPlane.image.tag` | Control plane image tag (defaults to `version`) | - |
+| `controlPlane.service.port` | Control plane service port | `3000` |
+| `postgresql.enabled` | Deploy PostgreSQL as subchart | `true` |
+| `postgresql.external.host` | External PostgreSQL host | `postgresql` |
+| `postgresql.external.port` | External PostgreSQL port | `5432` |
+| `postgresql.external.database` | Database name | `hindsight` |
+| `postgresql.external.username` | Database username | `hindsight` |
+| `ingress.enabled` | Enable ingress | `false` |
+| `autoscaling.enabled` | Enable HPA | `false` |
+
+### Environment Variables
+
+All environment variables in `api.env` and `controlPlane.env` are automatically added to the respective pods. Sensitive values should go in `api.secrets` or `controlPlane.secrets`.
+
+```yaml
+api:
+  env:
+    HINDSIGHT_API_LLM_PROVIDER: "openai"
+    HINDSIGHT_API_LLM_MODEL: "gpt-4"
+  secrets:
+    HINDSIGHT_API_LLM_API_KEY: "your-api-key"
+    HINDSIGHT_API_LLM_BASE_URL: "https://api.openai.com/v1"
+
+controlPlane:
+  env:
+    NODE_ENV: "production"
+  secrets: {}
+```
+
+### External Database
+
+To connect to an external PostgreSQL database:
+
+```yaml
+postgresql:
+  enabled: false
+  external:
+    host: "my-postgres.example.com"
+    port: 5432
+    database: "hindsight"
+    username: "hindsight"
+    password: "your-password"
+```
+
+### Ingress
+
+To expose the services via ingress:
+
+```yaml
+ingress:
+  enabled: true
+  className: "nginx"
+  annotations:
+    cert-manager.io/cluster-issuer: "letsencrypt-prod"
+  hosts:
+    - host: hindsight.example.com
+      paths:
+        - path: /
+          pathType: Prefix
+          service: controlPlane
+        - path: /api
+          pathType: Prefix
+          service: api
+  tls:
+    - secretName: hindsight-tls
+      hosts:
+        - hindsight.example.com
+```
+
+## Upgrading
+
+```bash
+helm upgrade hindsight ./helm/hindsight -n hindsight
+```
+
+## Uninstalling
+
+```bash
+helm uninstall hindsight -n hindsight
+```
+
+## Components
+
+The chart deploys:
+
+- **API**: The main Hindsight API server for memory operations
+- **Control Plane**: Web UI for managing agents and viewing memories
+
+## Development
+
+### Lint the chart
+
+```bash
+helm lint ./helm/hindsight
+```
+
+### Template locally
+
+```bash
+helm template hindsight ./helm/hindsight --debug
+```
+
+### Dry run installation
+
+```bash
+helm install hindsight ./helm/hindsight --dry-run --debug
+```

helm/hindsight/templates/NOTES.txt

@@ -1,71 +1,2 @@
-Thank you for installing {{ .Chart.Name }}!
-
-Your release is named {{ .Release.Name }}.
-
-To learn more about the release, try:
-
-  $ helm status {{ .Release.Name }}
-  $ helm get all {{ .Release.Name }}
-
-{{- if .Values.ingress.enabled }}
-
-The application is accessible via the following URL(s):
-{{- range .Values.ingress.hosts }}
-  - http{{ if $.Values.ingress.tls }}s{{ end }}://{{ .host }}
-{{- end }}
-
-{{- else }}
-
-1. Get the Control Plane URL by running these commands:
-{{- if contains "NodePort" .Values.controlPlane.service.type }}
-  export NODE_PORT=$(kubectl get --namespace {{ .Release.Namespace }} -o jsonpath="{.spec.ports[0].nodePort}" services {{ include "hindsight.fullname" . }}-control-plane)
-  export NODE_IP=$(kubectl get nodes --namespace {{ .Release.Namespace }} -o jsonpath="{.items[0].status.addresses[0].address}")
-  echo "Control Plane URL: http://$NODE_IP:$NODE_PORT"
-{{- else if contains "LoadBalancer" .Values.controlPlane.service.type }}
-  NOTE: It may take a few minutes for the LoadBalancer IP to be available.
-  You can watch the status by running 'kubectl get --namespace {{ .Release.Namespace }} svc -w {{ include "hindsight.fullname" . }}-control-plane'
-  export SERVICE_IP=$(kubectl get svc --namespace {{ .Release.Namespace }} {{ include "hindsight.fullname" . }}-control-plane --template "{{"{{ range (index .status.loadBalancer.ingress 0) }}{{.}}{{ end }}"}}")
-  echo "Control Plane URL: http://$SERVICE_IP:{{ .Values.controlPlane.service.port }}"
-{{- else if contains "ClusterIP" .Values.controlPlane.service.type }}
-  export POD_NAME=$(kubectl get pods --namespace {{ .Release.Namespace }} -l "app.kubernetes.io/component=control-plane,app.kubernetes.io/instance={{ .Release.Name }}" -o jsonpath="{.items[0].metadata.name}")
-  export CONTAINER_PORT=$(kubectl get pod --namespace {{ .Release.Namespace }} $POD_NAME -o jsonpath="{.spec.containers[0].ports[0].containerPort}")
-  echo "Control Plane URL: http://127.0.0.1:3000"
-  kubectl --namespace {{ .Release.Namespace }} port-forward $POD_NAME 3000:$CONTAINER_PORT
-{{- end }}
-
-2. Get the API URL by running these commands:
-{{- if contains "NodePort" .Values.api.service.type }}
-  export NODE_PORT=$(kubectl get --namespace {{ .Release.Namespace }} -o jsonpath="{.spec.ports[0].nodePort}" services {{ include "hindsight.fullname" . }}-api)
-  export NODE_IP=$(kubectl get nodes --namespace {{ .Release.Namespace }} -o jsonpath="{.items[0].status.addresses[0].address}")
-  echo "API URL: http://$NODE_IP:$NODE_PORT"
-{{- else if contains "LoadBalancer" .Values.api.service.type }}
-  NOTE: It may take a few minutes for the LoadBalancer IP to be available.
-  You can watch the status by running 'kubectl get --namespace {{ .Release.Namespace }} svc -w {{ include "hindsight.fullname" . }}-api'
-  export SERVICE_IP=$(kubectl get svc --namespace {{ .Release.Namespace }} {{ include "hindsight.fullname" . }}-api --template "{{"{{ range (index .status.loadBalancer.ingress 0) }}{{.}}{{ end }}"}}")
-  echo "API URL: http://$SERVICE_IP:{{ .Values.api.service.port }}"
-{{- else if contains "ClusterIP" .Values.api.service.type }}
-  export POD_NAME=$(kubectl get pods --namespace {{ .Release.Namespace }} -l "app.kubernetes.io/component=api,app.kubernetes.io/instance={{ .Release.Name }}" -o jsonpath="{.items[0].metadata.name}")
-  export CONTAINER_PORT=$(kubectl get pod --namespace {{ .Release.Namespace }} $POD_NAME -o jsonpath="{.spec.containers[0].ports[0].containerPort}")
-  echo "API URL: http://127.0.0.1:8888"
-  kubectl --namespace {{ .Release.Namespace }} port-forward $POD_NAME 8888:$CONTAINER_PORT
-{{- end }}
-
-{{- end }}
-
-{{- if not .Values.postgresql.enabled }}
-
-NOTE: You are using an external PostgreSQL database.
-Please ensure that:
-  1. The database is accessible from the cluster
-  2. The pgvector extension is enabled
-
-Database migrations run automatically when the API service starts.
-
-If you want to pre-initialize the database before deploying (optional):
-  kubectl run --namespace {{ .Release.Namespace }} hindsight-init --rm -it --restart=Never \
-    --image={{ .Values.api.image.repository }}:{{ .Values.api.image.tag }} \
-    --env="DATABASE_URL={{ include "hindsight.databaseUrl" . }}" \
-    -- python -c "from hindsight.migrations import run_migrations; run_migrations()"
-{{- end }}
-
-For more information, visit: https://github.com/yourusername/hindsight
+Hindsight installed. Access the control plane:
+  kubectl port-forward -n {{ .Release.Namespace }} svc/{{ include "hindsight.fullname" . }}-control-plane 3000:3000

helm/hindsight/templates/_helpers.tpl

@@ -80,6 +80,22 @@ Control plane selector labels
 app.kubernetes.io/component: control-plane
 {{- end }}
 
+{{/*
+Worker labels
+*/}}
+{{- define "hindsight.worker.labels" -}}
+{{ include "hindsight.labels" . }}
+app.kubernetes.io/component: worker
+{{- end }}
+
+{{/*
+Worker selector labels
+*/}}
+{{- define "hindsight.worker.selectorLabels" -}}
+{{ include "hindsight.selectorLabels" . }}
+app.kubernetes.io/component: worker
+{{- end }}
+
 {{/*
 Create the name of the service account to use
 */}}
@@ -98,7 +114,7 @@ Generate database URL
 {{- if .Values.databaseUrl }}
 {{- .Values.databaseUrl }}
 {{- else if .Values.postgresql.enabled }}
-{{- printf "postgresql://%s:%s@%s-postgresql:%d/%s" .Values.postgresql.auth.username .Values.postgresql.auth.password (include "hindsight.fullname" .) (.Values.postgresql.primary.service.port | int) .Values.postgresql.auth.database }}
+{{- printf "postgresql://%s:%s@%s-postgresql:%d/%s" .Values.postgresql.auth.username .Values.postgresql.auth.password (include "hindsight.fullname" .) (.Values.postgresql.service.port | int) .Values.postgresql.auth.database }}
 {{- else }}
 {{- printf "postgresql://%s:$(POSTGRES_PASSWORD)@%s:%d/%s" .Values.postgresql.external.username .Values.postgresql.external.host (.Values.postgresql.external.port | int) .Values.postgresql.external.database }}
 {{- end }}
@@ -110,3 +126,46 @@ API URL for control plane
 {{- define "hindsight.apiUrl" -}}
 {{- printf "http://%s-api:%d" (include "hindsight.fullname" .) (.Values.api.service.port | int) }}
 {{- end }}
+
+{{/*
+TEI reranker labels
+*/}}
+{{- define "hindsight.tei.reranker.labels" -}}
+{{ include "hindsight.labels" . }}
+app.kubernetes.io/component: tei-reranker
+{{- end }}
+
+{{/*
+TEI reranker selector labels
+*/}}
+{{- define "hindsight.tei.reranker.selectorLabels" -}}
+{{ include "hindsight.selectorLabels" . }}
+app.kubernetes.io/component: tei-reranker
+{{- end }}
+
+{{/*
+TEI embedding labels
+*/}}
+{{- define "hindsight.tei.embedding.labels" -}}
+{{ include "hindsight.labels" . }}
+app.kubernetes.io/component: tei-embedding
+{{- end }}
+
+{{/*
+TEI embedding selector labels
+*/}}
+{{- define "hindsight.tei.embedding.selectorLabels" -}}
+{{ include "hindsight.selectorLabels" . }}
+app.kubernetes.io/component: tei-embedding
+{{- end }}
+
+{{/*
+Get the name of the secret to use
+*/}}
+{{- define "hindsight.secretName" -}}
+{{- if .Values.existingSecret }}
+{{- .Values.existingSecret }}
+{{- else }}
+{{- printf "%s-secret" (include "hindsight.fullname" .) }}
+{{- end }}
+{{- end }}

helm/hindsight/templates/api-deployment.yaml

@@ -15,8 +15,9 @@ spec:
   template:
     metadata:
       annotations:
-        checksum/config: {{ include (print $.Template.BasePath "/configmap.yaml") . | sha256sum }}
+        {{- if not .Values.existingSecret }}
         checksum/secret: {{ include (print $.Template.BasePath "/secret.yaml") . | sha256sum }}
+        {{- end }}
         {{- with .Values.podAnnotations }}
         {{- toYaml . | nindent 8 }}
         {{- end }}
@@ -32,45 +33,61 @@ spec:
       - name: api
         securityContext:
           {{- toYaml .Values.securityContext | nindent 10 }}
-        image: "{{ .Values.api.image.repository }}:{{ .Values.api.image.tag }}"
+        image: "{{ .Values.api.image.repository }}:{{ .Values.api.image.tag | default .Values.version | default .Chart.AppVersion }}"
         imagePullPolicy: {{ .Values.api.image.pullPolicy }}
         ports:
         - name: http
           containerPort: {{ .Values.api.service.targetPort }}
           protocol: TCP
+        {{- if .Values.existingSecret }}
+        envFrom:
+        - secretRef:
+            name: {{ .Values.existingSecret }}
+        {{- end }}
         env:
-        - name: HINDSIGHT_API_DATABASE_URL
-          value: {{ include "hindsight.databaseUrl" . | quote }}
+        {{- /* POSTGRES_PASSWORD must be defined before DATABASE_URL for $(VAR) interpolation */}}
         {{- if not .Values.postgresql.enabled }}
         - name: POSTGRES_PASSWORD
           valueFrom:
             secretKeyRef:
-              name: {{ include "hindsight.fullname" . }}-secret
+              name: {{ include "hindsight.secretName" . }}
               key: postgres-password
         {{- end }}
-        - name: HINDSIGHT_API_LLM_PROVIDER
-          valueFrom:
-            configMapKeyRef:
-              name: {{ include "hindsight.fullname" . }}-config
-              key: llm-provider
-        - name: HINDSIGHT_API_LLM_MODEL
-          valueFrom:
-            configMapKeyRef:
-              name: {{ include "hindsight.fullname" . }}-config
-              key: llm-model
-        {{- if and .Values.api.secrets (hasKey .Values.api.secrets "HINDSIGHT_API_LLM_API_KEY") }}
-        - name: HINDSIGHT_API_LLM_API_KEY
-          valueFrom:
-            secretKeyRef:
-              name: {{ include "hindsight.fullname" . }}-secret
-              key: llm-api-key
+        - name: HINDSIGHT_API_DATABASE_URL
+          value: {{ include "hindsight.databaseUrl" . | quote }}
+        {{- /* Disable internal worker when dedicated workers are enabled */}}
+        {{- if .Values.worker.enabled }}
+        - name: HINDSIGHT_API_WORKER_ENABLED
+          value: "false"
         {{- end }}
-        {{- if and .Values.api.secrets (hasKey .Values.api.secrets "HINDSIGHT_API_LLM_BASE_URL") }}
-        - name: HINDSIGHT_API_LLM_BASE_URL
+        {{- /* Explicitly set port to override K8s service discovery env var (HINDSIGHT_API_PORT) */}}
+        - name: HINDSIGHT_API_PORT
+          value: {{ .Values.api.service.targetPort | quote }}
+        {{- range $key, $value := .Values.api.env }}
+        - name: {{ $key }}
+          value: {{ $value | quote }}
+        {{- end }}
+        {{- if .Values.tei.reranker.enabled }}
+        - name: HINDSIGHT_API_RERANKER_PROVIDER
+          value: "tei"
+        - name: HINDSIGHT_API_RERANKER_TEI_URL
+          value: "http://{{ include "hindsight.fullname" . }}-tei-reranker:{{ .Values.tei.reranker.port }}"
+        {{- end }}
+        {{- if .Values.tei.embedding.enabled }}
+        - name: HINDSIGHT_API_EMBEDDINGS_PROVIDER
+          value: "tei"
+        - name: HINDSIGHT_API_EMBEDDINGS_TEI_URL
+          value: "http://{{ include "hindsight.fullname" . }}-tei-embedding:{{ .Values.tei.embedding.port }}"
+        {{- end }}
+        {{- /* Only use api.secrets when not using existingSecret (for chart-managed secrets) */}}
+        {{- if not .Values.existingSecret }}
+        {{- range $key, $value := .Values.api.secrets }}
+        - name: {{ $key }}
           valueFrom:
             secretKeyRef:
-              name: {{ include "hindsight.fullname" . }}-secret
-              key: llm-base-url
+              name: {{ include "hindsight.secretName" $ }}
+              key: {{ $key }}
+        {{- end }}
         {{- end }}
         livenessProbe:
           {{- toYaml .Values.api.livenessProbe | nindent 10 }}
@@ -82,7 +99,7 @@ spec:
       nodeSelector:
         {{- toYaml . | nindent 8 }}
       {{- end }}
-      {{- with .Values.affinity }}
+      {{- with (.Values.api.affinity | default .Values.affinity) }}
       affinity:
         {{- toYaml . | nindent 8 }}
       {{- end }}

helm/hindsight/templates/configmap.yaml

@@ -1,15 +0,0 @@
-apiVersion: v1
-kind: ConfigMap
-metadata:
-  name: {{ include "hindsight.fullname" . }}-config
-  labels:
-    {{- include "hindsight.labels" . | nindent 4 }}
-data:
-  # API configuration
-  llm-provider: {{ .Values.api.env.HINDSIGHT_API_LLM_PROVIDER | quote }}
-  llm-model: {{ .Values.api.env.HINDSIGHT_API_LLM_MODEL | quote }}
-
-  # Control plane configuration
-  node-env: {{ .Values.controlPlane.env.NODE_ENV | quote }}
-  hostname: {{ .Values.controlPlane.env.HINDSIGHT_CP_HOSTNAME | quote }}
-  control-plane-port: {{ .Values.controlPlane.env.HINDSIGHT_CP_PORT | quote }}

helm/hindsight/templates/controlplane-deployment.yaml

@@ -15,7 +15,9 @@ spec:
   template:
     metadata:
       annotations:
-        checksum/config: {{ include (print $.Template.BasePath "/configmap.yaml") . | sha256sum }}
+        {{- if not .Values.existingSecret }}
+        checksum/secret: {{ include (print $.Template.BasePath "/secret.yaml") . | sha256sum }}
+        {{- end }}
         {{- with .Values.podAnnotations }}
         {{- toYaml . | nindent 8 }}
         {{- end }}
@@ -31,30 +33,34 @@ spec:
       - name: control-plane
         securityContext:
           {{- toYaml .Values.securityContext | nindent 10 }}
-        image: "{{ .Values.controlPlane.image.repository }}:{{ .Values.controlPlane.image.tag }}"
+        image: "{{ .Values.controlPlane.image.repository }}:{{ .Values.controlPlane.image.tag | default .Values.version | default .Chart.AppVersion }}"
         imagePullPolicy: {{ .Values.controlPlane.image.pullPolicy }}
         ports:
         - name: http
           containerPort: {{ .Values.controlPlane.service.targetPort }}
           protocol: TCP
+        {{- if .Values.existingSecret }}
+        envFrom:
+        - secretRef:
+            name: {{ .Values.existingSecret }}
+        {{- end }}
         env:
-        - name: NODE_ENV
-          valueFrom:
-            configMapKeyRef:
-              name: {{ include "hindsight.fullname" . }}-config
-              key: node-env
-        - name: HINDSIGHT_CP_HOSTNAME
-          valueFrom:
-            configMapKeyRef:
-              name: {{ include "hindsight.fullname" . }}-config
-              key: hostname
-        - name: HINDSIGHT_CP_PORT
-          valueFrom:
-            configMapKeyRef:
-              name: {{ include "hindsight.fullname" . }}-config
-              key: control-plane-port
         - name: HINDSIGHT_CP_DATAPLANE_API_URL
           value: {{ include "hindsight.apiUrl" . | quote }}
+        {{- range $key, $value := .Values.controlPlane.env }}
+        - name: {{ $key }}
+          value: {{ $value | quote }}
+        {{- end }}
+        {{- /* Only use controlPlane.secrets when not using existingSecret (for chart-managed secrets) */}}
+        {{- if not .Values.existingSecret }}
+        {{- range $key, $value := .Values.controlPlane.secrets }}
+        - name: {{ $key }}
+          valueFrom:
+            secretKeyRef:
+              name: {{ include "hindsight.secretName" $ }}
+              key: {{ $key }}
+        {{- end }}
+        {{- end }}
         livenessProbe:
           {{- toYaml .Values.controlPlane.livenessProbe | nindent 10 }}
         readinessProbe:
@@ -65,7 +71,7 @@ spec:
       nodeSelector:
         {{- toYaml . | nindent 8 }}
       {{- end }}
-      {{- with .Values.affinity }}
+      {{- with (.Values.controlPlane.affinity | default .Values.affinity) }}
       affinity:
         {{- toYaml . | nindent 8 }}
       {{- end }}

helm/hindsight/templates/pdb.yaml

@@ -0,0 +1,56 @@
+{{- if and .Values.api.enabled .Values.api.podDisruptionBudget.enabled }}
+apiVersion: policy/v1
+kind: PodDisruptionBudget
+metadata:
+  name: {{ include "hindsight.fullname" . }}-api
+  labels:
+    {{- include "hindsight.api.labels" . | nindent 4 }}
+spec:
+  {{- if .Values.api.podDisruptionBudget.minAvailable }}
+  minAvailable: {{ .Values.api.podDisruptionBudget.minAvailable }}
+  {{- end }}
+  {{- if .Values.api.podDisruptionBudget.maxUnavailable }}
+  maxUnavailable: {{ .Values.api.podDisruptionBudget.maxUnavailable }}
+  {{- end }}
+  selector:
+    matchLabels:
+      {{- include "hindsight.api.selectorLabels" . | nindent 6 }}
+{{- end }}
+---
+{{- if and .Values.controlPlane.enabled .Values.controlPlane.podDisruptionBudget.enabled }}
+apiVersion: policy/v1
+kind: PodDisruptionBudget
+metadata:
+  name: {{ include "hindsight.fullname" . }}-control-plane
+  labels:
+    {{- include "hindsight.controlPlane.labels" . | nindent 4 }}
+spec:
+  {{- if .Values.controlPlane.podDisruptionBudget.minAvailable }}
+  minAvailable: {{ .Values.controlPlane.podDisruptionBudget.minAvailable }}
+  {{- end }}
+  {{- if .Values.controlPlane.podDisruptionBudget.maxUnavailable }}
+  maxUnavailable: {{ .Values.controlPlane.podDisruptionBudget.maxUnavailable }}
+  {{- end }}
+  selector:
+    matchLabels:
+      {{- include "hindsight.controlPlane.selectorLabels" . | nindent 6 }}
+{{- end }}
+---
+{{- if and .Values.worker.enabled .Values.worker.podDisruptionBudget.enabled }}
+apiVersion: policy/v1
+kind: PodDisruptionBudget
+metadata:
+  name: {{ include "hindsight.fullname" . }}-worker
+  labels:
+    {{- include "hindsight.worker.labels" . | nindent 4 }}
+spec:
+  {{- if .Values.worker.podDisruptionBudget.minAvailable }}
+  minAvailable: {{ .Values.worker.podDisruptionBudget.minAvailable }}
+  {{- end }}
+  {{- if .Values.worker.podDisruptionBudget.maxUnavailable }}
+  maxUnavailable: {{ .Values.worker.podDisruptionBudget.maxUnavailable }}
+  {{- end }}
+  selector:
+    matchLabels:
+      {{- include "hindsight.worker.selectorLabels" . | nindent 6 }}
+{{- end }}

helm/hindsight/templates/postgresql-service.yaml

@@ -0,0 +1,19 @@
+{{- if .Values.postgresql.enabled }}
+apiVersion: v1
+kind: Service
+metadata:
+  name: {{ include "hindsight.fullname" . }}-postgresql
+  labels:
+    {{- include "hindsight.labels" . | nindent 4 }}
+    app.kubernetes.io/component: postgresql
+spec:
+  type: ClusterIP
+  ports:
+  - port: {{ .Values.postgresql.service.port }}
+    targetPort: postgresql
+    protocol: TCP
+    name: postgresql
+  selector:
+    {{- include "hindsight.selectorLabels" . | nindent 4 }}
+    app.kubernetes.io/component: postgresql
+{{- end }}

helm/hindsight/templates/postgresql-statefulset.yaml

@@ -0,0 +1,85 @@
+{{- if .Values.postgresql.enabled }}
+apiVersion: apps/v1
+kind: StatefulSet
+metadata:
+  name: {{ include "hindsight.fullname" . }}-postgresql
+  labels:
+    {{- include "hindsight.labels" . | nindent 4 }}
+    app.kubernetes.io/component: postgresql
+spec:
+  serviceName: {{ include "hindsight.fullname" . }}-postgresql
+  replicas: 1
+  selector:
+    matchLabels:
+      {{- include "hindsight.selectorLabels" . | nindent 6 }}
+      app.kubernetes.io/component: postgresql
+  template:
+    metadata:
+      labels:
+        {{- include "hindsight.selectorLabels" . | nindent 8 }}
+        app.kubernetes.io/component: postgresql
+    spec:
+      containers:
+      - name: postgresql
+        image: "{{ .Values.postgresql.image.repository }}:{{ .Values.postgresql.image.tag }}"
+        imagePullPolicy: {{ .Values.postgresql.image.pullPolicy }}
+        ports:
+        - name: postgresql
+          containerPort: 5432
+          protocol: TCP
+        env:
+        - name: POSTGRES_USER
+          value: {{ .Values.postgresql.auth.username | quote }}
+        - name: POSTGRES_PASSWORD
+          value: {{ .Values.postgresql.auth.password | quote }}
+        - name: POSTGRES_DB
+          value: {{ .Values.postgresql.auth.database | quote }}
+        - name: PGDATA
+          value: /var/lib/postgresql/data/pgdata
+        livenessProbe:
+          exec:
+            command:
+            - pg_isready
+            - -U
+            - {{ .Values.postgresql.auth.username }}
+            - -d
+            - {{ .Values.postgresql.auth.database }}
+          initialDelaySeconds: 30
+          periodSeconds: 10
+          timeoutSeconds: 5
+          failureThreshold: 3
+        readinessProbe:
+          exec:
+            command:
+            - pg_isready
+            - -U
+            - {{ .Values.postgresql.auth.username }}
+            - -d
+            - {{ .Values.postgresql.auth.database }}
+          initialDelaySeconds: 5
+          periodSeconds: 5
+          timeoutSeconds: 3
+          failureThreshold: 3
+        resources:
+          {{- toYaml .Values.postgresql.resources | nindent 10 }}
+        volumeMounts:
+        - name: data
+          mountPath: /var/lib/postgresql/data
+  {{- if .Values.postgresql.persistence.enabled }}
+  volumeClaimTemplates:
+  - metadata:
+      name: data
+    spec:
+      accessModes: ["ReadWriteOnce"]
+      {{- if .Values.postgresql.persistence.storageClass }}
+      storageClassName: {{ .Values.postgresql.persistence.storageClass | quote }}
+      {{- end }}
+      resources:
+        requests:
+          storage: {{ .Values.postgresql.persistence.size }}
+  {{- else }}
+      volumes:
+      - name: data
+        emptyDir: {}
+  {{- end }}
+{{- end }}

helm/hindsight/templates/secret.yaml

@@ -1,19 +1,19 @@
+{{- if not .Values.existingSecret }}
 apiVersion: v1
 kind: Secret
 metadata:
-  name: {{ include "hindsight.fullname" . }}-secret
+  name: {{ include "hindsight.secretName" . }}
   labels:
     {{- include "hindsight.labels" . | nindent 4 }}
 type: Opaque
 data:
-  {{- if and .Values.api.secrets (hasKey .Values.api.secrets "MEMORY_LLM_API_KEY") }}
-  llm-api-key: {{ .Values.api.secrets.MEMORY_LLM_API_KEY | b64enc | quote }}
+  {{- range $key, $value := .Values.api.secrets }}
+  {{ $key }}: {{ $value | b64enc | quote }}
   {{- end }}
-  {{- if and .Values.api.secrets (hasKey .Values.api.secrets "MEMORY_LLM_BASE_URL") }}
-  llm-base-url: {{ .Values.api.secrets.MEMORY_LLM_BASE_URL | b64enc | quote }}
+  {{- range $key, $value := .Values.controlPlane.secrets }}
+  {{ $key }}: {{ $value | b64enc | quote }}
   {{- end }}
-  {{- if not .Values.postgresql.enabled }}
-  {{- if .Values.postgresql.external.password }}
+  {{- if and (not .Values.postgresql.enabled) .Values.postgresql.external.password }}
   postgres-password: {{ .Values.postgresql.external.password | b64enc | quote }}
   {{- end }}
-  {{- end }}
+{{- end }}

helm/hindsight/templates/tei-embedding-deployment.yaml

@@ -0,0 +1,76 @@
+{{- if .Values.tei.embedding.enabled }}
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: {{ include "hindsight.fullname" . }}-tei-embedding
+  labels:
+    {{- include "hindsight.tei.embedding.labels" . | nindent 4 }}
+spec:
+  replicas: {{ .Values.tei.embedding.replicaCount }}
+  selector:
+    matchLabels:
+      {{- include "hindsight.tei.embedding.selectorLabels" . | nindent 6 }}
+  template:
+    metadata:
+      {{- with .Values.podAnnotations }}
+      annotations:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      labels:
+        {{- include "hindsight.tei.embedding.selectorLabels" . | nindent 8 }}
+    spec:
+      {{- if .Values.serviceAccount.create }}
+      serviceAccountName: {{ include "hindsight.serviceAccountName" . }}
+      {{- end }}
+      securityContext:
+        {{- toYaml .Values.podSecurityContext | nindent 8 }}
+      containers:
+      - name: tei-embedding
+        securityContext:
+          {{- toYaml .Values.securityContext | nindent 10 }}
+        image: "{{ .Values.tei.embedding.image.repository }}:{{ .Values.tei.embedding.image.tag }}"
+        imagePullPolicy: {{ .Values.tei.embedding.image.pullPolicy }}
+        args:
+          - "--model-id"
+          - {{ .Values.tei.embedding.model | quote }}
+          - "--hostname"
+          - "0.0.0.0"
+          {{- range .Values.tei.embedding.args }}
+          - {{ . | quote }}
+          {{- end }}
+        ports:
+        - name: http
+          containerPort: {{ .Values.tei.embedding.port }}
+          protocol: TCP
+        env:
+        - name: PORT
+          value: {{ .Values.tei.embedding.port | quote }}
+        {{- range $key, $value := .Values.tei.embedding.env }}
+        - name: {{ $key }}
+          value: {{ $value | quote }}
+        {{- end }}
+        livenessProbe:
+          {{- toYaml .Values.tei.embedding.livenessProbe | nindent 10 }}
+        readinessProbe:
+          {{- toYaml .Values.tei.embedding.readinessProbe | nindent 10 }}
+        resources:
+          {{- toYaml .Values.tei.embedding.resources | nindent 10 }}
+        volumeMounts:
+        - name: model-cache
+          mountPath: /data
+      volumes:
+      - name: model-cache
+        emptyDir: {}
+      {{- with .Values.nodeSelector }}
+      nodeSelector:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      {{- with .Values.affinity }}
+      affinity:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      {{- with .Values.tolerations }}
+      tolerations:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+{{- end }}

helm/hindsight/templates/tei-embedding-service.yaml

@@ -0,0 +1,17 @@
+{{- if .Values.tei.embedding.enabled }}
+apiVersion: v1
+kind: Service
+metadata:
+  name: {{ include "hindsight.fullname" . }}-tei-embedding
+  labels:
+    {{- include "hindsight.tei.embedding.labels" . | nindent 4 }}
+spec:
+  type: ClusterIP
+  ports:
+  - port: {{ .Values.tei.embedding.port }}
+    targetPort: http
+    protocol: TCP
+    name: http
+  selector:
+    {{- include "hindsight.tei.embedding.selectorLabels" . | nindent 4 }}
+{{- end }}

helm/hindsight/templates/tei-reranker-deployment.yaml

@@ -0,0 +1,76 @@
+{{- if .Values.tei.reranker.enabled }}
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: {{ include "hindsight.fullname" . }}-tei-reranker
+  labels:
+    {{- include "hindsight.tei.reranker.labels" . | nindent 4 }}
+spec:
+  replicas: {{ .Values.tei.reranker.replicaCount }}
+  selector:
+    matchLabels:
+      {{- include "hindsight.tei.reranker.selectorLabels" . | nindent 6 }}
+  template:
+    metadata:
+      {{- with .Values.podAnnotations }}
+      annotations:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      labels:
+        {{- include "hindsight.tei.reranker.selectorLabels" . | nindent 8 }}
+    spec:
+      {{- if .Values.serviceAccount.create }}
+      serviceAccountName: {{ include "hindsight.serviceAccountName" . }}
+      {{- end }}
+      securityContext:
+        {{- toYaml .Values.podSecurityContext | nindent 8 }}
+      containers:
+      - name: tei-reranker
+        securityContext:
+          {{- toYaml .Values.securityContext | nindent 10 }}
+        image: "{{ .Values.tei.reranker.image.repository }}:{{ .Values.tei.reranker.image.tag }}"
+        imagePullPolicy: {{ .Values.tei.reranker.image.pullPolicy }}
+        args:
+          - "--model-id"
+          - {{ .Values.tei.reranker.model | quote }}
+          - "--hostname"
+          - "0.0.0.0"
+          {{- range .Values.tei.reranker.args }}
+          - {{ . | quote }}
+          {{- end }}
+        ports:
+        - name: http
+          containerPort: {{ .Values.tei.reranker.port }}
+          protocol: TCP
+        env:
+        - name: PORT
+          value: {{ .Values.tei.reranker.port | quote }}
+        {{- range $key, $value := .Values.tei.reranker.env }}
+        - name: {{ $key }}
+          value: {{ $value | quote }}
+        {{- end }}
+        livenessProbe:
+          {{- toYaml .Values.tei.reranker.livenessProbe | nindent 10 }}
+        readinessProbe:
+          {{- toYaml .Values.tei.reranker.readinessProbe | nindent 10 }}
+        resources:
+          {{- toYaml .Values.tei.reranker.resources | nindent 10 }}
+        volumeMounts:
+        - name: model-cache
+          mountPath: /data
+      volumes:
+      - name: model-cache
+        emptyDir: {}
+      {{- with .Values.nodeSelector }}
+      nodeSelector:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      {{- with .Values.affinity }}
+      affinity:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      {{- with .Values.tolerations }}
+      tolerations:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+{{- end }}

helm/hindsight/templates/tei-reranker-service.yaml

@@ -0,0 +1,17 @@
+{{- if .Values.tei.reranker.enabled }}
+apiVersion: v1
+kind: Service
+metadata:
+  name: {{ include "hindsight.fullname" . }}-tei-reranker
+  labels:
+    {{- include "hindsight.tei.reranker.labels" . | nindent 4 }}
+spec:
+  type: ClusterIP
+  ports:
+  - port: {{ .Values.tei.reranker.port }}
+    targetPort: http
+    protocol: TCP
+    name: http
+  selector:
+    {{- include "hindsight.tei.reranker.selectorLabels" . | nindent 4 }}
+{{- end }}

helm/hindsight/templates/worker-service.yaml

@@ -0,0 +1,25 @@
+{{- if .Values.worker.enabled }}
+apiVersion: v1
+kind: Service
+metadata:
+  name: {{ include "hindsight.fullname" . }}-worker
+  labels:
+    {{- include "hindsight.worker.labels" . | nindent 4 }}
+  {{- if .Values.podAnnotations }}
+  annotations:
+    {{- /* Common Prometheus annotations for metrics scraping */}}
+    prometheus.io/scrape: "true"
+    prometheus.io/port: {{ .Values.worker.service.port | quote }}
+    prometheus.io/path: "/metrics"
+  {{- end }}
+spec:
+  # Headless service for StatefulSet (enables stable DNS names like worker-0.worker.namespace)
+  clusterIP: None
+  ports:
+  - port: {{ .Values.worker.service.port }}
+    targetPort: {{ .Values.worker.service.targetPort }}
+    protocol: TCP
+    name: http
+  selector:
+    {{- include "hindsight.worker.selectorLabels" . | nindent 4 }}
+{{- end }}

helm/hindsight/templates/worker-statefulset.yaml

@@ -0,0 +1,110 @@
+{{- if .Values.worker.enabled }}
+apiVersion: apps/v1
+kind: StatefulSet
+metadata:
+  name: {{ include "hindsight.fullname" . }}-worker
+  labels:
+    {{- include "hindsight.worker.labels" . | nindent 4 }}
+spec:
+  serviceName: {{ include "hindsight.fullname" . }}-worker
+  replicas: {{ .Values.worker.replicaCount }}
+  selector:
+    matchLabels:
+      {{- include "hindsight.worker.selectorLabels" . | nindent 6 }}
+  template:
+    metadata:
+      annotations:
+        {{- if not .Values.existingSecret }}
+        checksum/secret: {{ include (print $.Template.BasePath "/secret.yaml") . | sha256sum }}
+        {{- end }}
+        {{- with .Values.podAnnotations }}
+        {{- toYaml . | nindent 8 }}
+        {{- end }}
+      labels:
+        {{- include "hindsight.worker.selectorLabels" . | nindent 8 }}
+    spec:
+      {{- if .Values.serviceAccount.create }}
+      serviceAccountName: {{ include "hindsight.serviceAccountName" . }}
+      {{- end }}
+      securityContext:
+        {{- toYaml .Values.podSecurityContext | nindent 8 }}
+      containers:
+      - name: worker
+        securityContext:
+          {{- toYaml .Values.securityContext | nindent 10 }}
+        image: "{{ .Values.worker.image.repository }}:{{ .Values.worker.image.tag | default .Values.version | default .Chart.AppVersion }}"
+        imagePullPolicy: {{ .Values.worker.image.pullPolicy }}
+        command: ["hindsight-worker"]
+        ports:
+        - name: http
+          containerPort: {{ .Values.worker.service.targetPort }}
+          protocol: TCP
+        {{- if .Values.existingSecret }}
+        envFrom:
+        - secretRef:
+            name: {{ .Values.existingSecret }}
+        {{- end }}
+        env:
+        {{- /* POSTGRES_PASSWORD must be defined before DATABASE_URL for $(VAR) interpolation */}}
+        {{- if not .Values.postgresql.enabled }}
+        - name: POSTGRES_PASSWORD
+          valueFrom:
+            secretKeyRef:
+              name: {{ include "hindsight.secretName" . }}
+              key: postgres-password
+        {{- end }}
+        - name: HINDSIGHT_API_DATABASE_URL
+          value: {{ include "hindsight.databaseUrl" . | quote }}
+        {{- /* Worker ID uses pod name (StatefulSet provides stable names like worker-0, worker-1) */}}
+        - name: HINDSIGHT_API_WORKER_ID
+          valueFrom:
+            fieldRef:
+              fieldPath: metadata.name
+        {{- /* Inherit LLM config from api.env */}}
+        {{- range $key, $value := .Values.api.env }}
+        - name: {{ $key }}
+          value: {{ $value | quote }}
+        {{- end }}
+        {{- /* Worker-specific env vars */}}
+        {{- range $key, $value := .Values.worker.env }}
+        - name: {{ $key }}
+          value: {{ $value | quote }}
+        {{- end }}
+        {{- /* Only use secrets when not using existingSecret */}}
+        {{- if not .Values.existingSecret }}
+        {{- /* Inherit secrets from api.secrets */}}
+        {{- range $key, $value := .Values.api.secrets }}
+        - name: {{ $key }}
+          valueFrom:
+            secretKeyRef:
+              name: {{ include "hindsight.secretName" $ }}
+              key: {{ $key }}
+        {{- end }}
+        {{- /* Worker-specific secrets (can override api.secrets) */}}
+        {{- range $key, $value := .Values.worker.secrets }}
+        - name: {{ $key }}
+          valueFrom:
+            secretKeyRef:
+              name: {{ include "hindsight.secretName" $ }}
+              key: {{ $key }}
+        {{- end }}
+        {{- end }}
+        livenessProbe:
+          {{- toYaml .Values.worker.livenessProbe | nindent 10 }}
+        readinessProbe:
+          {{- toYaml .Values.worker.readinessProbe | nindent 10 }}
+        resources:
+          {{- toYaml .Values.worker.resources | nindent 10 }}
+      {{- with .Values.nodeSelector }}
+      nodeSelector:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      {{- with (.Values.worker.affinity | default .Values.affinity) }}
+      affinity:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      {{- with .Values.tolerations }}
+      tolerations:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+{{- end }}

helm/hindsight/values.yaml

@@ -1,5 +1,18 @@
 # Default values for hindsight
 
+# Global version override - use this to set a consistent image tag across all components
+# If not set, defaults to Chart.appVersion from Chart.yaml
+# version: ""
+
+# Use an existing secret instead of creating one from values
+# When set, all keys from this secret are injected as environment variables via envFrom
+# Required keys:
+#   - postgres-password: PostgreSQL password (when postgresql.enabled=false)
+# Optional keys (any key becomes an env var):
+#   - HINDSIGHT_API_LLM_API_KEY: API key for LLM provider
+#   - Any other env vars you want to inject
+# existingSecret: "my-hindsight-secret"
+
 # Global settings
 replicaCount: 1
 
@@ -8,9 +21,9 @@ api:
   enabled: true
   replicaCount: 1
   image:
-    repository: hindsight/api
+    repository: ghcr.io/vectorize-io/hindsight-api
     pullPolicy: IfNotPresent
-    tag: "latest"
+    # tag defaults to .Values.version if not specified
 
   service:
     type: ClusterIP
@@ -29,7 +42,7 @@ api:
   # Liveness and readiness probes
   livenessProbe:
     httpGet:
-      path: /
+      path: /health
       port: 8888
     initialDelaySeconds: 30
     periodSeconds: 10
@@ -38,16 +51,25 @@ api:
 
   readinessProbe:
     httpGet:
-      path: /
+      path: /health
       port: 8888
     initialDelaySeconds: 10
     periodSeconds: 5
     timeoutSeconds: 3
     failureThreshold: 3
 
+  # Pod disruption budget
+  podDisruptionBudget:
+    enabled: false
+    minAvailable: 1
+    # maxUnavailable: 1
+
+  # Pod affinity/anti-affinity (overrides global affinity for this component)
+  # affinity: {}
+
   # Environment variables
   env:
-    HINDSIGHT_API_LLM_PROVIDER: "groq"
+    #HINDSIGHT_API_LLM_PROVIDER: "groq"
     HINDSIGHT_API_LLM_MODEL: "openai/gpt-oss-120b"
 
   # Secret environment variables
@@ -55,14 +77,80 @@ api:
     # HINDSIGHT_API_LLM_API_KEY: "your-api-key"
     # HINDSIGHT_API_LLM_BASE_URL: "https://api.groq.com/openai/v1"
 
+# Worker settings (distributed task processing)
+# When enabled, dedicated worker pods process tasks and the API's internal worker is disabled
+worker:
+  enabled: false
+  replicaCount: 2
+  image:
+    repository: ghcr.io/vectorize-io/hindsight-api
+    pullPolicy: IfNotPresent
+    # tag: ""  # defaults to .Values.version, then Chart.appVersion if not specified
+
+  service:
+    # Service for metrics scraping (headless for StatefulSet)
+    port: 8889
+    targetPort: 8889
+
+  # Resource limits and requests
+  resources:
+    limits:
+      cpu: 2000m
+      memory: 4Gi
+    requests:
+      cpu: 500m
+      memory: 1Gi
+
+  # Liveness and readiness probes
+  livenessProbe:
+    httpGet:
+      path: /health
+      port: 8889
+    initialDelaySeconds: 30
+    periodSeconds: 10
+    timeoutSeconds: 5
+    failureThreshold: 3
+
+  readinessProbe:
+    httpGet:
+      path: /health
+      port: 8889
+    initialDelaySeconds: 10
+    periodSeconds: 5
+    timeoutSeconds: 3
+    failureThreshold: 3
+
+  # Worker-specific environment variables
+  env:
+    # Poll interval in milliseconds (how often to check for new tasks)
+    HINDSIGHT_API_WORKER_POLL_INTERVAL_MS: "500"
+    # Number of tasks to claim per poll cycle
+    HINDSIGHT_API_WORKER_BATCH_SIZE: "10"
+    # Max retries before marking a task as failed
+    HINDSIGHT_API_WORKER_MAX_RETRIES: "3"
+    # HTTP port for metrics/health (matches service.targetPort)
+    HINDSIGHT_API_WORKER_HTTP_PORT: "8889"
+
+  # Pod disruption budget
+  podDisruptionBudget:
+    enabled: false
+    minAvailable: 1
+    # maxUnavailable: 1
+
+  # Pod affinity/anti-affinity (overrides global affinity for this component)
+  # affinity: {}
+
+  # Secret environment variables (inherited from api.secrets if not specified)
+  secrets: {}
+
 # Image settings for control plane
 controlPlane:
   enabled: true
   replicaCount: 1
   image:
-    repository: hindsight/hindsight-control-plane
+    repository: ghcr.io/vectorize-io/hindsight-control-plane
     pullPolicy: IfNotPresent
-    tag: "latest"
+    # tag defaults to .Values.version if not specified
 
   service:
     type: ClusterIP
@@ -78,10 +166,9 @@ controlPlane:
       cpu: 250m
       memory: 512Mi
 
-  # Liveness and readiness probes
+  # Liveness and readiness probes (TCP check)
   livenessProbe:
-    httpGet:
-      path: /
+    tcpSocket:
       port: 3000
     initialDelaySeconds: 30
     periodSeconds: 10
@@ -89,14 +176,22 @@ controlPlane:
     failureThreshold: 3
 
   readinessProbe:
-    httpGet:
-      path: /
+    tcpSocket:
       port: 3000
     initialDelaySeconds: 10
     periodSeconds: 5
     timeoutSeconds: 3
     failureThreshold: 3
 
+  # Pod disruption budget
+  podDisruptionBudget:
+    enabled: false
+    minAvailable: 1
+    # maxUnavailable: 1
+
+  # Pod affinity/anti-affinity (overrides global affinity for this component)
+  # affinity: {}
+
   # Environment variables
   env:
     NODE_ENV: "production"
@@ -106,21 +201,43 @@ controlPlane:
 # PostgreSQL configuration
 postgresql:
   # Set to true to deploy PostgreSQL as part of this chart
-  enabled: false
+  enabled: true
+
+  image:
+    repository: ankane/pgvector
+    tag: latest
+    pullPolicy: IfNotPresent
+
+  auth:
+    username: "hindsight"
+    password: "hindsight"
+    database: "hindsight"
+
+  service:
+    port: 5432
+
+  persistence:
+    enabled: true
+    size: 8Gi
+    # storageClass: ""
+
+  resources:
+    limits:
+      cpu: 1000m
+      memory: 1Gi
+    requests:
+      cpu: 250m
+      memory: 256Mi
 
   # External PostgreSQL connection details
-  # If postgresql.enabled is false, provide external database details
+  # Only used if postgresql.enabled is false
   external:
     host: "postgresql"
     port: 5432
     database: "hindsight"
     username: "hindsight"
-    # Password should be provided via secret
     # password: ""
 
-# Database URL (auto-generated from postgresql config if not provided)
-# databaseUrl: "postgresql://user:pass@host:5432/database"
-
 # Ingress configuration
 ingress:
   enabled: false
@@ -173,9 +290,87 @@ nodeSelector: {}
 # Tolerations
 tolerations: []
 
-# Affinity
+# Affinity (applied to all components unless overridden per-component)
 affinity: {}
 
+# TEI (Text Embeddings Inference) - optional standalone deployments
+# for reranking and/or embedding models
+tei:
+  reranker:
+    enabled: false
+    replicaCount: 1
+    image:
+      repository: ghcr.io/huggingface/text-embeddings-inference
+      tag: cpu-1.8.3
+      pullPolicy: IfNotPresent
+    model: "cross-encoder/ms-marco-MiniLM-L-6-v2"
+    port: 8090
+    args:
+      - "--auto-truncate"
+    env:
+      PAYLOAD_LIMIT: "10000000"
+      MAX_CLIENT_BATCH_SIZE: "256"
+    resources:
+      limits:
+        cpu: 2000m
+        memory: 2Gi
+      requests:
+        cpu: 500m
+        memory: 1Gi
+    livenessProbe:
+      httpGet:
+        path: /health
+        port: 8090
+      initialDelaySeconds: 30
+      periodSeconds: 10
+      timeoutSeconds: 5
+      failureThreshold: 6
+    readinessProbe:
+      httpGet:
+        path: /health
+        port: 8090
+      initialDelaySeconds: 15
+      periodSeconds: 5
+      timeoutSeconds: 3
+      failureThreshold: 3
+
+  embedding:
+    enabled: false
+    replicaCount: 1
+    image:
+      repository: ghcr.io/huggingface/text-embeddings-inference
+      tag: cpu-1.8.3
+      pullPolicy: IfNotPresent
+    model: "sentence-transformers/all-MiniLM-L6-v2"
+    port: 8091
+    args: []
+    env:
+      PAYLOAD_LIMIT: "10000000"
+      MAX_CLIENT_BATCH_SIZE: "256"
+    resources:
+      limits:
+        cpu: 2000m
+        memory: 2Gi
+      requests:
+        cpu: 500m
+        memory: 1Gi
+    livenessProbe:
+      httpGet:
+        path: /health
+        port: 8091
+      initialDelaySeconds: 30
+      periodSeconds: 10
+      timeoutSeconds: 5
+      failureThreshold: 6
+    readinessProbe:
+      httpGet:
+        path: /health
+        port: 8091
+      initialDelaySeconds: 15
+      periodSeconds: 5
+      timeoutSeconds: 3
+      failureThreshold: 3
+
 # Autoscaling
 autoscaling:
   enabled: false

hindsight-api/README.md

@@ -1 +1,137 @@
-# Memory 
+# 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/alembic/versions/5a366d414dce_initial_schema.py

@@ -1,274 +0,0 @@
-"""initial_schema
-
-Revision ID: 5a366d414dce
-Revises:
-Create Date: 2025-11-27 11:54:19.228030
-
-"""
-from typing import Sequence, Union
-
-from alembic import op
-import sqlalchemy as sa
-from sqlalchemy.dialects import postgresql
-from pgvector.sqlalchemy import Vector
-
-
-# revision identifiers, used by Alembic.
-revision: str = '5a366d414dce'
-down_revision: Union[str, Sequence[str], None] = None
-branch_labels: Union[str, Sequence[str], None] = None
-depends_on: Union[str, Sequence[str], None] = None
-
-
-def upgrade() -> None:
-    """Upgrade schema - create all tables from scratch."""
-
-    # Enable required extensions
-    op.execute('CREATE EXTENSION IF NOT EXISTS vector')
-
-    # Create banks table
-    op.create_table(
-        'banks',
-        sa.Column('bank_id', sa.Text(), nullable=False),
-        sa.Column('name', sa.Text(), nullable=True),
-        sa.Column('personality', postgresql.JSONB(astext_type=sa.Text()), server_default=sa.text("'{}'::jsonb"), nullable=False),
-        sa.Column('background', sa.Text(), nullable=True),
-        sa.Column('created_at', postgresql.TIMESTAMP(timezone=True), server_default=sa.text('now()'), nullable=False),
-        sa.Column('updated_at', postgresql.TIMESTAMP(timezone=True), server_default=sa.text('now()'), nullable=False),
-        sa.PrimaryKeyConstraint('bank_id', name=op.f('pk_banks'))
-    )
-
-    # Create documents table
-    op.create_table(
-        'documents',
-        sa.Column('id', sa.Text(), nullable=False),
-        sa.Column('bank_id', sa.Text(), nullable=False),
-        sa.Column('original_text', sa.Text(), nullable=True),
-        sa.Column('content_hash', sa.Text(), nullable=True),
-        sa.Column('metadata', postgresql.JSONB(astext_type=sa.Text()), server_default=sa.text("'{}'::jsonb"), nullable=False),
-        sa.Column('created_at', postgresql.TIMESTAMP(timezone=True), server_default=sa.text('now()'), nullable=False),
-        sa.Column('updated_at', postgresql.TIMESTAMP(timezone=True), server_default=sa.text('now()'), nullable=False),
-        sa.PrimaryKeyConstraint('id', 'bank_id', name=op.f('pk_documents'))
-    )
-    op.create_index('idx_documents_bank_id', 'documents', ['bank_id'])
-    op.create_index('idx_documents_content_hash', 'documents', ['content_hash'])
-
-    # Create async_operations table
-    op.create_table(
-        'async_operations',
-        sa.Column('operation_id', postgresql.UUID(as_uuid=True), server_default=sa.text('gen_random_uuid()'), nullable=False),
-        sa.Column('bank_id', sa.Text(), nullable=False),
-        sa.Column('operation_type', sa.Text(), nullable=False),
-        sa.Column('status', sa.Text(), server_default='pending', nullable=False),
-        sa.Column('created_at', postgresql.TIMESTAMP(timezone=True), server_default=sa.text('now()'), nullable=False),
-        sa.Column('updated_at', postgresql.TIMESTAMP(timezone=True), server_default=sa.text('now()'), nullable=False),
-        sa.Column('completed_at', postgresql.TIMESTAMP(timezone=True), nullable=True),
-        sa.Column('error_message', sa.Text(), nullable=True),
-        sa.Column('result_metadata', postgresql.JSONB(astext_type=sa.Text()), server_default=sa.text("'{}'::jsonb"), nullable=False),
-        sa.PrimaryKeyConstraint('operation_id', name=op.f('pk_async_operations')),
-        sa.CheckConstraint("status IN ('pending', 'processing', 'completed', 'failed')", name='async_operations_status_check')
-    )
-    op.create_index('idx_async_operations_bank_id', 'async_operations', ['bank_id'])
-    op.create_index('idx_async_operations_status', 'async_operations', ['status'])
-    op.create_index('idx_async_operations_bank_status', 'async_operations', ['bank_id', 'status'])
-
-    # Create entities table
-    op.create_table(
-        'entities',
-        sa.Column('id', postgresql.UUID(as_uuid=True), server_default=sa.text('gen_random_uuid()'), nullable=False),
-        sa.Column('canonical_name', sa.Text(), nullable=False),
-        sa.Column('bank_id', sa.Text(), nullable=False),
-        sa.Column('metadata', postgresql.JSONB(astext_type=sa.Text()), server_default=sa.text("'{}'::jsonb"), nullable=False),
-        sa.Column('first_seen', postgresql.TIMESTAMP(timezone=True), server_default=sa.text('now()'), nullable=False),
-        sa.Column('last_seen', postgresql.TIMESTAMP(timezone=True), server_default=sa.text('now()'), nullable=False),
-        sa.Column('mention_count', sa.Integer(), server_default='1', nullable=False),
-        sa.PrimaryKeyConstraint('id', name=op.f('pk_entities'))
-    )
-    op.create_index('idx_entities_bank_id', 'entities', ['bank_id'])
-    op.create_index('idx_entities_canonical_name', 'entities', ['canonical_name'])
-    op.create_index('idx_entities_bank_name', 'entities', ['bank_id', 'canonical_name'])
-    # Create unique index on (bank_id, LOWER(canonical_name)) for entity resolution
-    op.execute('CREATE UNIQUE INDEX idx_entities_bank_lower_name ON entities (bank_id, LOWER(canonical_name))')
-
-    # Create memory_units table
-    op.create_table(
-        'memory_units',
-        sa.Column('id', postgresql.UUID(as_uuid=True), server_default=sa.text('gen_random_uuid()'), nullable=False),
-        sa.Column('bank_id', sa.Text(), nullable=False),
-        sa.Column('document_id', sa.Text(), nullable=True),
-        sa.Column('text', sa.Text(), nullable=False),
-        sa.Column('embedding', Vector(384), nullable=True),
-        sa.Column('context', sa.Text(), nullable=True),
-        sa.Column('event_date', postgresql.TIMESTAMP(timezone=True), nullable=False),
-        sa.Column('occurred_start', postgresql.TIMESTAMP(timezone=True), nullable=True),
-        sa.Column('occurred_end', postgresql.TIMESTAMP(timezone=True), nullable=True),
-        sa.Column('mentioned_at', postgresql.TIMESTAMP(timezone=True), nullable=True),
-        sa.Column('fact_type', sa.Text(), server_default='world', nullable=False),
-        sa.Column('confidence_score', sa.Float(), nullable=True),
-        sa.Column('access_count', sa.Integer(), server_default='0', nullable=False),
-        sa.Column('metadata', postgresql.JSONB(astext_type=sa.Text()), server_default=sa.text("'{}'::jsonb"), nullable=False),
-        sa.Column('created_at', postgresql.TIMESTAMP(timezone=True), server_default=sa.text('now()'), nullable=False),
-        sa.Column('updated_at', postgresql.TIMESTAMP(timezone=True), server_default=sa.text('now()'), nullable=False),
-        sa.ForeignKeyConstraint(['document_id', 'bank_id'], ['documents.id', 'documents.bank_id'], name='memory_units_document_fkey', ondelete='CASCADE'),
-        sa.PrimaryKeyConstraint('id', name=op.f('pk_memory_units')),
-        sa.CheckConstraint("fact_type IN ('world', 'bank', 'opinion', 'observation')", name='memory_units_fact_type_check'),
-        sa.CheckConstraint("confidence_score IS NULL OR (confidence_score >= 0.0 AND confidence_score <= 1.0)", name='memory_units_confidence_range_check'),
-        sa.CheckConstraint(
-            "(fact_type = 'opinion' AND confidence_score IS NOT NULL) OR "
-            "(fact_type = 'observation') OR "
-            "(fact_type NOT IN ('opinion', 'observation') AND confidence_score IS NULL)",
-            name='confidence_score_fact_type_check'
-        )
-    )
-
-    # Add search_vector column for full-text search
-    op.execute("""
-        ALTER TABLE memory_units
-        ADD COLUMN search_vector tsvector
-        GENERATED ALWAYS AS (to_tsvector('english', COALESCE(text, '') || ' ' || COALESCE(context, ''))) STORED
-    """)
-
-    op.create_index('idx_memory_units_bank_id', 'memory_units', ['bank_id'])
-    op.create_index('idx_memory_units_document_id', 'memory_units', ['document_id'])
-    op.create_index('idx_memory_units_event_date', 'memory_units', [sa.text('event_date DESC')])
-    op.create_index('idx_memory_units_bank_date', 'memory_units', ['bank_id', sa.text('event_date DESC')])
-    op.create_index('idx_memory_units_access_count', 'memory_units', [sa.text('access_count DESC')])
-    op.create_index('idx_memory_units_fact_type', 'memory_units', ['fact_type'])
-    op.create_index('idx_memory_units_bank_fact_type', 'memory_units', ['bank_id', 'fact_type'])
-    op.create_index('idx_memory_units_bank_type_date', 'memory_units', ['bank_id', 'fact_type', sa.text('event_date DESC')])
-    op.create_index('idx_memory_units_opinion_confidence', 'memory_units', ['bank_id', sa.text('confidence_score DESC')], postgresql_where=sa.text("fact_type = 'opinion'"))
-    op.create_index('idx_memory_units_opinion_date', 'memory_units', ['bank_id', sa.text('event_date DESC')], postgresql_where=sa.text("fact_type = 'opinion'"))
-    op.create_index('idx_memory_units_observation_date', 'memory_units', ['bank_id', sa.text('event_date DESC')], postgresql_where=sa.text("fact_type = 'observation'"))
-    op.create_index('idx_memory_units_embedding', 'memory_units', ['embedding'], postgresql_using='hnsw', postgresql_ops={'embedding': 'vector_cosine_ops'})
-
-    # Create BM25 full-text search index on search_vector
-    op.execute("""
-        CREATE INDEX idx_memory_units_text_search ON memory_units
-        USING gin(search_vector)
-    """)
-
-    op.execute("""
-        CREATE MATERIALIZED VIEW memory_units_bm25 AS
-        SELECT
-            id,
-            bank_id,
-            text,
-            to_tsvector('english', text) AS text_vector,
-            log(1.0 + length(text)::float / (SELECT avg(length(text)) FROM memory_units)) AS doc_length_factor
-        FROM memory_units
-    """)
-
-    op.create_index('idx_memory_units_bm25_bank', 'memory_units_bm25', ['bank_id'])
-    op.create_index('idx_memory_units_bm25_text_vector', 'memory_units_bm25', ['text_vector'], postgresql_using='gin')
-
-    # Create entity_cooccurrences table
-    op.create_table(
-        'entity_cooccurrences',
-        sa.Column('entity_id_1', postgresql.UUID(as_uuid=True), nullable=False),
-        sa.Column('entity_id_2', postgresql.UUID(as_uuid=True), nullable=False),
-        sa.Column('cooccurrence_count', sa.Integer(), server_default='1', nullable=False),
-        sa.Column('last_cooccurred', postgresql.TIMESTAMP(timezone=True), server_default=sa.text('now()'), nullable=False),
-        sa.ForeignKeyConstraint(['entity_id_1'], ['entities.id'], name=op.f('fk_entity_cooccurrences_entity_id_1_entities'), ondelete='CASCADE'),
-        sa.ForeignKeyConstraint(['entity_id_2'], ['entities.id'], name=op.f('fk_entity_cooccurrences_entity_id_2_entities'), ondelete='CASCADE'),
-        sa.PrimaryKeyConstraint('entity_id_1', 'entity_id_2', name=op.f('pk_entity_cooccurrences')),
-        sa.CheckConstraint('entity_id_1 < entity_id_2', name='entity_cooccurrence_order_check')
-    )
-    op.create_index('idx_entity_cooccurrences_entity1', 'entity_cooccurrences', ['entity_id_1'])
-    op.create_index('idx_entity_cooccurrences_entity2', 'entity_cooccurrences', ['entity_id_2'])
-    op.create_index('idx_entity_cooccurrences_count', 'entity_cooccurrences', [sa.text('cooccurrence_count DESC')])
-
-    # Create memory_links table
-    op.create_table(
-        'memory_links',
-        sa.Column('from_unit_id', postgresql.UUID(as_uuid=True), nullable=False),
-        sa.Column('to_unit_id', postgresql.UUID(as_uuid=True), nullable=False),
-        sa.Column('link_type', sa.Text(), nullable=False),
-        sa.Column('entity_id', postgresql.UUID(as_uuid=True), nullable=True),
-        sa.Column('weight', sa.Float(), server_default='1.0', nullable=False),
-        sa.Column('created_at', postgresql.TIMESTAMP(timezone=True), server_default=sa.text('now()'), nullable=False),
-        sa.ForeignKeyConstraint(['entity_id'], ['entities.id'], name=op.f('fk_memory_links_entity_id_entities'), ondelete='CASCADE'),
-        sa.ForeignKeyConstraint(['from_unit_id'], ['memory_units.id'], name=op.f('fk_memory_links_from_unit_id_memory_units'), ondelete='CASCADE'),
-        sa.ForeignKeyConstraint(['to_unit_id'], ['memory_units.id'], name=op.f('fk_memory_links_to_unit_id_memory_units'), ondelete='CASCADE'),
-        sa.CheckConstraint("link_type IN ('temporal', 'semantic', 'entity', 'causes', 'caused_by', 'enables', 'prevents')", name='memory_links_link_type_check'),
-        sa.CheckConstraint('weight >= 0.0 AND weight <= 1.0', name='memory_links_weight_check')
-    )
-    # Create unique constraint using COALESCE for nullable entity_id
-    op.execute("CREATE UNIQUE INDEX idx_memory_links_unique ON memory_links (from_unit_id, to_unit_id, link_type, COALESCE(entity_id, '00000000-0000-0000-0000-000000000000'::uuid))")
-    op.create_index('idx_memory_links_from_unit', 'memory_links', ['from_unit_id'])
-    op.create_index('idx_memory_links_to_unit', 'memory_links', ['to_unit_id'])
-    op.create_index('idx_memory_links_entity', 'memory_links', ['entity_id'])
-    op.create_index('idx_memory_links_link_type', 'memory_links', ['link_type'])
-
-    # Create unit_entities table
-    op.create_table(
-        'unit_entities',
-        sa.Column('unit_id', postgresql.UUID(as_uuid=True), nullable=False),
-        sa.Column('entity_id', postgresql.UUID(as_uuid=True), nullable=False),
-        sa.ForeignKeyConstraint(['entity_id'], ['entities.id'], name=op.f('fk_unit_entities_entity_id_entities'), ondelete='CASCADE'),
-        sa.ForeignKeyConstraint(['unit_id'], ['memory_units.id'], name=op.f('fk_unit_entities_unit_id_memory_units'), ondelete='CASCADE'),
-        sa.PrimaryKeyConstraint('unit_id', 'entity_id', name=op.f('pk_unit_entities'))
-    )
-    op.create_index('idx_unit_entities_unit', 'unit_entities', ['unit_id'])
-    op.create_index('idx_unit_entities_entity', 'unit_entities', ['entity_id'])
-
-
-def downgrade() -> None:
-    """Downgrade schema - drop all tables."""
-
-    # Drop tables in reverse dependency order
-    op.drop_index('idx_unit_entities_entity', table_name='unit_entities')
-    op.drop_index('idx_unit_entities_unit', table_name='unit_entities')
-    op.drop_table('unit_entities')
-
-    op.drop_index('idx_memory_links_link_type', table_name='memory_links')
-    op.drop_index('idx_memory_links_entity', table_name='memory_links')
-    op.drop_index('idx_memory_links_to_unit', table_name='memory_links')
-    op.drop_index('idx_memory_links_from_unit', table_name='memory_links')
-    op.execute('DROP INDEX IF EXISTS idx_memory_links_unique')
-    op.drop_table('memory_links')
-
-    op.drop_index('idx_entity_cooccurrences_count', table_name='entity_cooccurrences')
-    op.drop_index('idx_entity_cooccurrences_entity2', table_name='entity_cooccurrences')
-    op.drop_index('idx_entity_cooccurrences_entity1', table_name='entity_cooccurrences')
-    op.drop_table('entity_cooccurrences')
-
-    # Drop BM25 materialized view and index
-    op.drop_index('idx_memory_units_bm25_text_vector', table_name='memory_units_bm25')
-    op.drop_index('idx_memory_units_bm25_bank', table_name='memory_units_bm25')
-    op.execute('DROP MATERIALIZED VIEW IF EXISTS memory_units_bm25')
-
-    op.drop_index('idx_memory_units_embedding', table_name='memory_units')
-    op.drop_index('idx_memory_units_observation_date', table_name='memory_units')
-    op.drop_index('idx_memory_units_opinion_date', table_name='memory_units')
-    op.drop_index('idx_memory_units_opinion_confidence', table_name='memory_units')
-    op.drop_index('idx_memory_units_bank_type_date', table_name='memory_units')
-    op.drop_index('idx_memory_units_bank_fact_type', table_name='memory_units')
-    op.drop_index('idx_memory_units_fact_type', table_name='memory_units')
-    op.drop_index('idx_memory_units_access_count', table_name='memory_units')
-    op.drop_index('idx_memory_units_bank_date', table_name='memory_units')
-    op.drop_index('idx_memory_units_event_date', table_name='memory_units')
-    op.drop_index('idx_memory_units_document_id', table_name='memory_units')
-    op.drop_index('idx_memory_units_bank_id', table_name='memory_units')
-    op.execute('DROP INDEX IF EXISTS idx_memory_units_text_search')
-    op.drop_table('memory_units')
-
-    op.execute('DROP INDEX IF EXISTS idx_entities_bank_lower_name')
-    op.drop_index('idx_entities_bank_name', table_name='entities')
-    op.drop_index('idx_entities_canonical_name', table_name='entities')
-    op.drop_index('idx_entities_bank_id', table_name='entities')
-    op.drop_table('entities')
-
-    op.drop_index('idx_async_operations_bank_status', table_name='async_operations')
-    op.drop_index('idx_async_operations_status', table_name='async_operations')
-    op.drop_index('idx_async_operations_bank_id', table_name='async_operations')
-    op.drop_table('async_operations')
-
-    op.drop_index('idx_documents_content_hash', table_name='documents')
-    op.drop_index('idx_documents_bank_id', table_name='documents')
-    op.drop_table('documents')
-
-    op.drop_table('banks')
-
-    # Drop extensions (optional - comment out if you want to keep them)
-    # op.execute('DROP EXTENSION IF EXISTS vector')
-    # op.execute('DROP EXTENSION IF EXISTS "uuid-ossp"')

hindsight-api/alembic/versions/b7c4d8e9f1a2_add_chunks_table.py

@@ -1,70 +0,0 @@
-"""add_chunks_table
-
-Revision ID: b7c4d8e9f1a2
-Revises: 5a366d414dce
-Create Date: 2025-11-28 00:00:00.000000
-
-"""
-from typing import Sequence, Union
-
-from alembic import op
-import sqlalchemy as sa
-from sqlalchemy.dialects import postgresql
-
-
-# revision identifiers, used by Alembic.
-revision: str = 'b7c4d8e9f1a2'
-down_revision: Union[str, Sequence[str], None] = '5a366d414dce'
-branch_labels: Union[str, Sequence[str], None] = None
-depends_on: Union[str, Sequence[str], None] = None
-
-
-def upgrade() -> None:
-    """Add chunks table and link memory_units to chunks."""
-
-    # Create chunks table with single text PK (bank_id_document_id_chunk_index)
-    op.create_table(
-        'chunks',
-        sa.Column('chunk_id', sa.Text(), nullable=False),
-        sa.Column('document_id', sa.Text(), nullable=False),
-        sa.Column('bank_id', sa.Text(), nullable=False),
-        sa.Column('chunk_index', sa.Integer(), nullable=False),
-        sa.Column('chunk_text', sa.Text(), nullable=False),
-        sa.Column('created_at', postgresql.TIMESTAMP(timezone=True), server_default=sa.text('now()'), nullable=False),
-        sa.ForeignKeyConstraint(['document_id', 'bank_id'], ['documents.id', 'documents.bank_id'], name='chunks_document_fkey', ondelete='CASCADE'),
-        sa.PrimaryKeyConstraint('chunk_id', name=op.f('pk_chunks'))
-    )
-
-    # Add indexes for efficient queries
-    op.create_index('idx_chunks_document_id', 'chunks', ['document_id'])
-    op.create_index('idx_chunks_bank_id', 'chunks', ['bank_id'])
-
-    # Add chunk_id column to memory_units (nullable, as existing records won't have chunks)
-    op.add_column('memory_units', sa.Column('chunk_id', sa.Text(), nullable=True))
-
-    # Add foreign key constraint to chunks table
-    op.create_foreign_key(
-        'memory_units_chunk_fkey',
-        'memory_units',
-        'chunks',
-        ['chunk_id'],
-        ['chunk_id'],
-        ondelete='SET NULL'
-    )
-
-    # Add index on chunk_id for efficient lookups
-    op.create_index('idx_memory_units_chunk_id', 'memory_units', ['chunk_id'])
-
-
-def downgrade() -> None:
-    """Remove chunks table and chunk_id from memory_units."""
-
-    # Drop index and foreign key from memory_units
-    op.drop_index('idx_memory_units_chunk_id', table_name='memory_units')
-    op.drop_constraint('memory_units_chunk_fkey', 'memory_units', type_='foreignkey')
-    op.drop_column('memory_units', 'chunk_id')
-
-    # Drop chunks table indexes and table
-    op.drop_index('idx_chunks_bank_id', table_name='chunks')
-    op.drop_index('idx_chunks_document_id', table_name='chunks')
-    op.drop_table('chunks')

hindsight-api/alembic/versions/d9f6a3b4c5e2_rename_bank_to_interactions.py

@@ -1,48 +0,0 @@
-"""Rename fact_type 'bank' to 'experience'
-
-Revision ID: d9f6a3b4c5e2
-Revises: c8e5f2a3b4d1
-Create Date: 2024-12-04 15:00:00.000000
-
-"""
-from alembic import op
-import sqlalchemy as sa
-
-
-# revision identifiers, used by Alembic.
-revision = 'd9f6a3b4c5e2'
-down_revision = 'c8e5f2a3b4d1'
-branch_labels = None
-depends_on = None
-
-
-def upgrade():
-    # Drop old check constraint FIRST (before updating data)
-    op.drop_constraint('memory_units_fact_type_check', 'memory_units', type_='check')
-
-    # Update existing 'bank' values to 'experience'
-    op.execute("UPDATE memory_units SET fact_type = 'experience' WHERE fact_type = 'bank'")
-    # Also update any 'interactions' values (in case of partial migration)
-    op.execute("UPDATE memory_units SET fact_type = 'experience' WHERE fact_type = 'interactions'")
-
-    # Create new check constraint with 'experience' instead of 'bank'
-    op.create_check_constraint(
-        'memory_units_fact_type_check',
-        'memory_units',
-        "fact_type IN ('world', 'experience', 'opinion', 'observation')"
-    )
-
-
-def downgrade():
-    # Drop new check constraint FIRST
-    op.drop_constraint('memory_units_fact_type_check', 'memory_units', type_='check')
-
-    # Update 'experience' back to 'bank'
-    op.execute("UPDATE memory_units SET fact_type = 'bank' WHERE fact_type = 'experience'")
-
-    # Recreate old check constraint
-    op.create_check_constraint(
-        'memory_units_fact_type_check',
-        'memory_units',
-        "fact_type IN ('world', 'bank', 'opinion', 'observation')"
-    )

hindsight-api/alembic/versions/e0a1b2c3d4e5_disposition_to_3_traits.py

@@ -1,62 +0,0 @@
-"""disposition_to_3_traits
-
-Revision ID: e0a1b2c3d4e5
-Revises: rename_personality
-Create Date: 2024-12-08
-
-Migrate disposition traits from Big Five (openness, conscientiousness, extraversion,
-agreeableness, neuroticism, bias_strength with 0-1 float values) to the new 3-trait
-system (skepticism, literalism, empathy with 1-5 integer values).
-"""
-from typing import Sequence, Union
-
-from alembic import op
-import sqlalchemy as sa
-
-
-# revision identifiers, used by Alembic.
-revision: str = 'e0a1b2c3d4e5'
-down_revision: Union[str, Sequence[str], None] = 'rename_personality'
-branch_labels: Union[str, Sequence[str], None] = None
-depends_on: Union[str, Sequence[str], None] = None
-
-
-def upgrade() -> None:
-    """Convert Big Five disposition to 3-trait disposition."""
-    conn = op.get_bind()
-
-    # Update all existing banks to use the new disposition format
-    # Convert from old format to new format with reasonable mappings:
-    # - skepticism: derived from inverse of agreeableness (skeptical people are less agreeable)
-    # - literalism: derived from conscientiousness (detail-oriented people are more literal)
-    # - empathy: derived from agreeableness + inverse of neuroticism
-    # Default all to 3 (neutral) for simplicity
-    conn.execute(sa.text("""
-        UPDATE banks
-        SET disposition = '{"skepticism": 3, "literalism": 3, "empathy": 3}'::jsonb
-        WHERE disposition IS NOT NULL
-    """))
-
-    # Update the default for new banks
-    conn.execute(sa.text("""
-        ALTER TABLE banks
-        ALTER COLUMN disposition SET DEFAULT '{"skepticism": 3, "literalism": 3, "empathy": 3}'::jsonb
-    """))
-
-
-def downgrade() -> None:
-    """Convert back to Big Five disposition."""
-    conn = op.get_bind()
-
-    # Revert to Big Five format with default values
-    conn.execute(sa.text("""
-        UPDATE banks
-        SET disposition = '{"openness": 0.5, "conscientiousness": 0.5, "extraversion": 0.5, "agreeableness": 0.5, "neuroticism": 0.5, "bias_strength": 0.5}'::jsonb
-        WHERE disposition IS NOT NULL
-    """))
-
-    # Update the default for new banks
-    conn.execute(sa.text("""
-        ALTER TABLE banks
-        ALTER COLUMN disposition SET DEFAULT '{"openness": 0.5, "conscientiousness": 0.5, "extraversion": 0.5, "agreeableness": 0.5, "neuroticism": 0.5, "bias_strength": 0.5}'::jsonb
-    """))

hindsight-api/alembic/versions/rename_personality_to_disposition.py

@@ -1,65 +0,0 @@
-"""rename_personality_to_disposition
-
-Revision ID: rename_personality
-Revises: d9f6a3b4c5e2
-Create Date: 2024-12-04
-
-"""
-from typing import Sequence, Union
-
-from alembic import op
-import sqlalchemy as sa
-from sqlalchemy.dialects import postgresql
-
-
-# revision identifiers, used by Alembic.
-revision: str = 'rename_personality'
-down_revision: Union[str, Sequence[str], None] = 'd9f6a3b4c5e2'
-branch_labels: Union[str, Sequence[str], None] = None
-depends_on: Union[str, Sequence[str], None] = None
-
-
-def upgrade() -> None:
-    """Rename personality column to disposition in banks table (if it exists)."""
-    conn = op.get_bind()
-
-    # Check if 'personality' column exists (old database)
-    result = conn.execute(sa.text("""
-        SELECT column_name
-        FROM information_schema.columns
-        WHERE table_name = 'banks' AND column_name = 'personality'
-    """))
-    has_personality = result.fetchone() is not None
-
-    # Check if 'disposition' column exists (new database)
-    result = conn.execute(sa.text("""
-        SELECT column_name
-        FROM information_schema.columns
-        WHERE table_name = 'banks' AND column_name = 'disposition'
-    """))
-    has_disposition = result.fetchone() is not None
-
-    if has_personality and not has_disposition:
-        # Old database: rename personality -> disposition
-        op.alter_column('banks', 'personality', new_column_name='disposition')
-    elif not has_personality and not has_disposition:
-        # Neither exists (shouldn't happen, but be safe): add disposition column
-        op.add_column('banks', sa.Column(
-            'disposition',
-            postgresql.JSONB(astext_type=sa.Text()),
-            server_default=sa.text("'{}'::jsonb"),
-            nullable=False
-        ))
-    # else: disposition already exists, nothing to do
-
-
-def downgrade() -> None:
-    """Revert disposition column back to personality."""
-    conn = op.get_bind()
-    result = conn.execute(sa.text("""
-        SELECT column_name
-        FROM information_schema.columns
-        WHERE table_name = 'banks' AND column_name = 'disposition'
-    """))
-    if result.fetchone():
-        op.alter_column('banks', 'disposition', new_column_name='personality')

hindsight-api/hindsight_api/__init__.py

@@ -3,24 +3,31 @@ Memory System for AI Agents.
 
 Temporal + Semantic Memory Architecture using PostgreSQL with pgvector.
 """
+
+from .config import HindsightConfig, get_config
+from .engine.cross_encoder import CrossEncoderModel, LocalSTCrossEncoder, RemoteTEICrossEncoder
+from .engine.embeddings import Embeddings, LocalSTEmbeddings, RemoteTEIEmbeddings
+from .engine.llm_wrapper import LLMConfig
 from .engine.memory_engine import MemoryEngine
 from .engine.search.trace import (
-    SearchTrace,
-    QueryInfo,
     EntryPoint,
-    NodeVisit,
-    WeightComponents,
     LinkInfo,
+    NodeVisit,
     PruningDecision,
-    SearchSummary,
+    QueryInfo,
     SearchPhaseMetrics,
+    SearchSummary,
+    SearchTrace,
+    WeightComponents,
 )
 from .engine.search.tracer import SearchTracer
-from .engine.embeddings import Embeddings, SentenceTransformersEmbeddings
-from .engine.llm_wrapper import LLMConfig
+from .models import RequestContext
 
 __all__ = [
     "MemoryEngine",
+    "RequestContext",
+    "HindsightConfig",
+    "get_config",
     "SearchTrace",
     "SearchTracer",
     "QueryInfo",
@@ -32,7 +39,11 @@ __all__ = [
     "SearchSummary",
     "SearchPhaseMetrics",
     "Embeddings",
-    "SentenceTransformersEmbeddings",
+    "LocalSTEmbeddings",
+    "RemoteTEIEmbeddings",
+    "CrossEncoderModel",
+    "LocalSTCrossEncoder",
+    "RemoteTEICrossEncoder",
     "LLMConfig",
 ]
-__version__ = "0.1.0"
+__version__ = "0.4.11"

hindsight-api/hindsight_api/admin/__init__.py

@@ -0,0 +1 @@
+# Admin CLI for Hindsight

hindsight-api/hindsight_api/admin/cli.py

@@ -0,0 +1,311 @@
+"""
+Hindsight Admin CLI - backup and restore operations.
+"""
+
+import asyncio
+import io
+import json
+import logging
+import zipfile
+from datetime import datetime, timezone
+from pathlib import Path
+from typing import Any
+
+import asyncpg
+import typer
+
+from ..config import HindsightConfig
+from ..pg0 import parse_pg0_url, resolve_database_url
+
+
+def _fq_table(table: str, schema: str) -> str:
+    """Get fully-qualified table name with schema prefix."""
+    return f"{schema}.{table}"
+
+
+# Setup logging
+logging.basicConfig(
+    level=logging.INFO,
+    format="%(message)s",
+)
+logger = logging.getLogger(__name__)
+
+app = typer.Typer(name="hindsight-admin", help="Hindsight administrative commands")
+
+# Tables to backup/restore in dependency order
+# Import must happen in this order due to foreign key constraints
+BACKUP_TABLES = [
+    "banks",
+    "documents",
+    "entities",
+    "chunks",
+    "memory_units",
+    "unit_entities",
+    "entity_cooccurrences",
+    "memory_links",
+]
+
+MANIFEST_VERSION = "1"
+
+
+async def _backup(database_url: str, output_path: Path, schema: str = "public") -> dict[str, Any]:
+    """Backup all tables to a zip file using binary COPY protocol."""
+    conn = await asyncpg.connect(database_url)
+    try:
+        tables: dict[str, Any] = {}
+        manifest: dict[str, Any] = {
+            "version": MANIFEST_VERSION,
+            "created_at": datetime.now(timezone.utc).isoformat(),
+            "schema": schema,
+            "tables": tables,
+        }
+
+        # Use a transaction with REPEATABLE READ isolation to get a consistent
+        # snapshot across all tables. This prevents race conditions where
+        # entity_cooccurrences could reference entities created after the
+        # entities table was backed up.
+        async with conn.transaction(isolation="repeatable_read"):
+            with zipfile.ZipFile(output_path, "w", zipfile.ZIP_DEFLATED) as zf:
+                for i, table in enumerate(BACKUP_TABLES, 1):
+                    typer.echo(f"  [{i}/{len(BACKUP_TABLES)}] Backing up {table}...", nl=False)
+
+                    buffer = io.BytesIO()
+
+                    # Use binary COPY for exact type preservation
+                    # asyncpg requires schema_name as separate parameter
+                    await conn.copy_from_table(table, schema_name=schema, output=buffer, format="binary")
+
+                    data = buffer.getvalue()
+                    zf.writestr(f"{table}.bin", data)
+
+                    # Get row count for manifest
+                    qualified_table = _fq_table(table, schema)
+                    row_count = await conn.fetchval(f"SELECT COUNT(*) FROM {qualified_table}")
+                    tables[table] = {
+                        "rows": row_count,
+                        "size_bytes": len(data),
+                    }
+
+                    typer.echo(f" {row_count} rows")
+
+                zf.writestr("manifest.json", json.dumps(manifest, indent=2))
+
+        return manifest
+    finally:
+        await conn.close()
+
+
+async def _restore(database_url: str, input_path: Path, schema: str = "public") -> dict[str, Any]:
+    """Restore all tables from a zip file using binary COPY protocol."""
+    conn = await asyncpg.connect(database_url)
+    try:
+        with zipfile.ZipFile(input_path, "r") as zf:
+            # Read and validate manifest
+            manifest: dict[str, Any] = json.loads(zf.read("manifest.json"))
+            if manifest.get("version") != MANIFEST_VERSION:
+                raise ValueError(f"Unsupported backup version: {manifest.get('version')}")
+
+            # Use a transaction for atomic restore - either all tables are
+            # restored or none are, preventing partial/inconsistent state.
+            async with conn.transaction():
+                typer.echo("  Clearing existing data...")
+                # Truncate tables in reverse order (respects FK constraints)
+                for table in reversed(BACKUP_TABLES):
+                    qualified_table = _fq_table(table, schema)
+                    await conn.execute(f"TRUNCATE TABLE {qualified_table} CASCADE")
+
+                # Restore tables in forward order
+                for i, table in enumerate(BACKUP_TABLES, 1):
+                    filename = f"{table}.bin"
+                    if filename not in zf.namelist():
+                        typer.echo(f"  [{i}/{len(BACKUP_TABLES)}] {table}: skipped (not in backup)")
+                        continue
+
+                    expected_rows = manifest["tables"].get(table, {}).get("rows", "?")
+                    typer.echo(f"  [{i}/{len(BACKUP_TABLES)}] Restoring {table}... {expected_rows} rows")
+
+                    data = zf.read(filename)
+                    buffer = io.BytesIO(data)
+                    # asyncpg requires schema_name as separate parameter
+                    await conn.copy_to_table(table, schema_name=schema, source=buffer, format="binary")
+
+                # Refresh materialized view
+                typer.echo("  Refreshing materialized views...")
+                await conn.execute(f"REFRESH MATERIALIZED VIEW {_fq_table('memory_units_bm25', schema)}")
+
+        return manifest
+    finally:
+        await conn.close()
+
+
+async def _run_backup(db_url: str, output: Path, schema: str = "public") -> dict[str, Any]:
+    """Resolve database URL and run backup."""
+    is_pg0, instance_name, _ = parse_pg0_url(db_url)
+    if is_pg0:
+        typer.echo(f"Starting embedded PostgreSQL (instance: {instance_name})...")
+    resolved_url = await resolve_database_url(db_url)
+    return await _backup(resolved_url, output, schema)
+
+
+async def _run_restore(db_url: str, input_file: Path, schema: str = "public") -> dict[str, Any]:
+    """Resolve database URL and run restore."""
+    is_pg0, instance_name, _ = parse_pg0_url(db_url)
+    if is_pg0:
+        typer.echo(f"Starting embedded PostgreSQL (instance: {instance_name})...")
+    resolved_url = await resolve_database_url(db_url)
+    return await _restore(resolved_url, input_file, schema)
+
+
+@app.command()
+def backup(
+    output: Path = typer.Argument(..., help="Output file path (.zip)"),
+    schema: str = typer.Option("public", "--schema", "-s", help="Database schema to backup"),
+):
+    """Backup the Hindsight database to a zip file."""
+    config = HindsightConfig.from_env()
+
+    if not config.database_url:
+        typer.echo("Error: Database URL not configured.", err=True)
+        typer.echo("Set HINDSIGHT_API_DATABASE_URL environment variable.", err=True)
+        raise typer.Exit(1)
+
+    if output.suffix != ".zip":
+        output = output.with_suffix(".zip")
+
+    typer.echo(f"Backing up database (schema: {schema}) to {output}...")
+
+    manifest = asyncio.run(_run_backup(config.database_url, output, schema))
+
+    total_rows = sum(t["rows"] for t in manifest["tables"].values())
+    typer.echo(f"Backed up {total_rows} rows across {len(BACKUP_TABLES)} tables")
+    typer.echo(f"Backup saved to {output}")
+
+
+@app.command()
+def restore(
+    input_file: Path = typer.Argument(..., help="Input backup file (.zip)"),
+    schema: str = typer.Option("public", "--schema", "-s", help="Database schema to restore to"),
+    yes: bool = typer.Option(False, "--yes", "-y", help="Skip confirmation prompt"),
+):
+    """Restore the database from a backup file. WARNING: This deletes all existing data."""
+    config = HindsightConfig.from_env()
+
+    if not config.database_url:
+        typer.echo("Error: Database URL not configured.", err=True)
+        typer.echo("Set HINDSIGHT_API_DATABASE_URL environment variable.", err=True)
+        raise typer.Exit(1)
+
+    if not input_file.exists():
+        typer.echo(f"Error: File not found: {input_file}", err=True)
+        raise typer.Exit(1)
+
+    if not yes:
+        typer.confirm(
+            "This will DELETE all existing data and replace it with the backup. Continue?",
+            abort=True,
+        )
+
+    typer.echo(f"Restoring database (schema: {schema}) from {input_file}...")
+
+    manifest = asyncio.run(_run_restore(config.database_url, input_file, schema))
+
+    total_rows = sum(t["rows"] for t in manifest["tables"].values())
+    typer.echo(f"Restored {total_rows} rows across {len(BACKUP_TABLES)} tables")
+    typer.echo("Restore complete")
+
+
+async def _run_migration(db_url: str, schema: str = "public") -> None:
+    """Resolve database URL and run migrations."""
+    from ..migrations import run_migrations
+
+    is_pg0, instance_name, _ = parse_pg0_url(db_url)
+    if is_pg0:
+        typer.echo(f"Starting embedded PostgreSQL (instance: {instance_name})...")
+    resolved_url = await resolve_database_url(db_url)
+    run_migrations(resolved_url, schema=schema)
+
+
+@app.command(name="run-db-migration")
+def run_db_migration(
+    schema: str = typer.Option("public", "--schema", "-s", help="Database schema to run migrations on"),
+):
+    """Run database migrations to the latest version."""
+    config = HindsightConfig.from_env()
+
+    if not config.database_url:
+        typer.echo("Error: Database URL not configured.", err=True)
+        typer.echo("Set HINDSIGHT_API_DATABASE_URL environment variable.", err=True)
+        raise typer.Exit(1)
+
+    typer.echo(f"Running database migrations (schema: {schema})...")
+
+    asyncio.run(_run_migration(config.database_url, schema))
+
+    typer.echo("Database migrations completed successfully")
+
+
+async def _decommission_worker(db_url: str, worker_id: str, schema: str = "public") -> int:
+    """Release all tasks owned by a worker, setting them back to pending status."""
+    is_pg0, instance_name, _ = parse_pg0_url(db_url)
+    if is_pg0:
+        typer.echo(f"Starting embedded PostgreSQL (instance: {instance_name})...")
+    resolved_url = await resolve_database_url(db_url)
+
+    conn = await asyncpg.connect(resolved_url)
+    try:
+        table = _fq_table("async_operations", schema)
+        result = await conn.fetch(
+            f"""
+            UPDATE {table}
+            SET status = 'pending', worker_id = NULL, claimed_at = NULL, updated_at = now()
+            WHERE worker_id = $1 AND status = 'processing'
+            RETURNING operation_id
+            """,
+            worker_id,
+        )
+        return len(result)
+    finally:
+        await conn.close()
+
+
+@app.command(name="decommission-worker")
+def decommission_worker(
+    worker_id: str = typer.Argument(..., help="Worker ID to decommission"),
+    schema: str = typer.Option("public", "--schema", "-s", help="Database schema"),
+    yes: bool = typer.Option(False, "--yes", "-y", help="Skip confirmation prompt"),
+):
+    """Release all tasks owned by a worker (sets status back to pending).
+
+    Use this command when a worker has crashed or been removed without graceful shutdown.
+    All tasks that were being processed by the worker will be released back to the queue
+    so other workers can pick them up.
+    """
+    config = HindsightConfig.from_env()
+
+    if not config.database_url:
+        typer.echo("Error: Database URL not configured.", err=True)
+        typer.echo("Set HINDSIGHT_API_DATABASE_URL environment variable.", err=True)
+        raise typer.Exit(1)
+
+    if not yes:
+        typer.confirm(
+            f"This will release all tasks owned by worker '{worker_id}' back to pending. Continue?",
+            abort=True,
+        )
+
+    typer.echo(f"Decommissioning worker '{worker_id}' (schema: {schema})...")
+
+    count = asyncio.run(_decommission_worker(config.database_url, worker_id, schema))
+
+    if count > 0:
+        typer.echo(f"Released {count} task(s) from worker '{worker_id}'")
+    else:
+        typer.echo(f"No tasks found for worker '{worker_id}'")
+
+
+def main():
+    app()
+
+
+if __name__ == "__main__":
+    main()

hindsight-api/hindsight_api/alembic/env.py

@@ -2,20 +2,19 @@
 Alembic environment configuration for SQLAlchemy with pgvector.
 Uses synchronous psycopg2 driver for migrations to avoid pgbouncer issues.
 """
+
 import logging
 import os
-import sys
 from pathlib import Path
 
-from sqlalchemy import pool, engine_from_config
-from sqlalchemy.engine import Connection
-
 from alembic import context
 from dotenv import load_dotenv
+from sqlalchemy import engine_from_config, pool
 
 # Import your models here
 from hindsight_api.models import Base
 
+
 # Load environment variables based on HINDSIGHT_API_DATABASE_URL env var or default to local
 def load_env():
     """Load environment variables from .env"""
@@ -30,6 +29,7 @@ def load_env():
     if env_file.exists():
         load_dotenv(env_file)
 
+
 load_env()
 
 # this is the Alembic Config object, which provides
@@ -109,6 +109,9 @@ def run_migrations_online() -> None:
 
     get_database_url()  # Process and set the database URL in config
 
+    # Check if we're targeting a specific schema (for multi-tenant isolation)
+    target_schema = config.get_main_option("target_schema")
+
     connectable = engine_from_config(
         config.get_section(config.config_ini_section, {}),
         prefix="sqlalchemy.",
@@ -121,17 +124,34 @@ def run_migrations_online() -> None:
     def set_read_write_mode(dbapi_connection, connection_record):
         cursor = dbapi_connection.cursor()
         cursor.execute("SET SESSION CHARACTERISTICS AS TRANSACTION READ WRITE")
+        # If targeting a specific schema, set search_path
+        # Include public in search_path for access to shared extensions (pgvector)
+        if target_schema:
+            cursor.execute(f'CREATE SCHEMA IF NOT EXISTS "{target_schema}"')
+            cursor.execute(f'SET search_path TO "{target_schema}", public')
         cursor.close()
 
     with connectable.connect() as connection:
         # Also explicitly set read-write mode on this connection
         connection.execute(text("SET SESSION CHARACTERISTICS AS TRANSACTION READ WRITE"))
+
+        # If targeting a specific schema, set search_path
+        # Include public in search_path for access to shared extensions (pgvector)
+        if target_schema:
+            connection.execute(text(f'CREATE SCHEMA IF NOT EXISTS "{target_schema}"'))
+            connection.execute(text(f'SET search_path TO "{target_schema}", public'))
+
         connection.commit()  # Commit the SET command
 
-        context.configure(
-            connection=connection,
-            target_metadata=target_metadata
-        )
+        # Configure context with version_table_schema if using a specific schema
+        context_opts = {
+            "connection": connection,
+            "target_metadata": target_metadata,
+        }
+        if target_schema:
+            context_opts["version_table_schema"] = target_schema
+
+        context.configure(**context_opts)
 
         with context.begin_transaction():
             context.run_migrations()

hindsight-api/hindsight_api/alembic/versions/5a366d414dce_initial_schema.py

@@ -0,0 +1,488 @@
+"""initial_schema
+
+Revision ID: 5a366d414dce
+Revises:
+Create Date: 2025-11-27 11:54:19.228030
+
+"""
+
+import os
+from collections.abc import Sequence
+
+import sqlalchemy as sa
+from alembic import op
+from pgvector.sqlalchemy import Vector
+from sqlalchemy import text
+from sqlalchemy.dialects import postgresql
+
+# revision identifiers, used by Alembic.
+revision: str = "5a366d414dce"
+down_revision: str | Sequence[str] | None = None
+branch_labels: str | Sequence[str] | None = None
+depends_on: str | Sequence[str] | None = None
+
+
+def _detect_vector_extension() -> str:
+    """
+    Detect or validate vector extension: 'vchord' or 'pgvector'.
+    Respects HINDSIGHT_API_VECTOR_EXTENSION env var if set.
+    """
+    conn = op.get_bind()
+    vector_extension = os.getenv("HINDSIGHT_API_VECTOR_EXTENSION", "pgvector").lower()
+
+    # Validate configured extension is installed
+    if vector_extension == "vchord":
+        vchord_check = conn.execute(text("SELECT 1 FROM pg_extension WHERE extname = 'vchord'")).scalar()
+        if not vchord_check:
+            raise RuntimeError(
+                "Configured vector extension 'vchord' not found. Install it with: CREATE EXTENSION vchord CASCADE;"
+            )
+        return "vchord"
+    elif vector_extension == "pgvector":
+        pgvector_check = conn.execute(text("SELECT 1 FROM pg_extension WHERE extname = 'vector'")).scalar()
+        if not pgvector_check:
+            raise RuntimeError(
+                "Configured vector extension 'pgvector' not found. Install it with: CREATE EXTENSION vector;"
+            )
+        return "pgvector"
+    else:
+        raise ValueError(f"Invalid HINDSIGHT_API_VECTOR_EXTENSION: {vector_extension}. Must be 'pgvector' or 'vchord'")
+
+
+def _detect_text_search_extension() -> str:
+    """
+    Detect or validate text search extension: 'native', 'vchord', or 'pg_textsearch'.
+    Respects HINDSIGHT_API_TEXT_SEARCH_EXTENSION env var.
+    Creates the extension if needed.
+    """
+    text_search_extension = os.getenv("HINDSIGHT_API_TEXT_SEARCH_EXTENSION", "native").lower()
+
+    if text_search_extension == "vchord":
+        # Create vchord_bm25 extension if not exists
+        try:
+            op.execute("CREATE EXTENSION IF NOT EXISTS vchord_bm25 CASCADE")
+        except Exception:
+            # Extension might already exist or user lacks permissions - verify it exists
+            conn = op.get_bind()
+            result = conn.execute(text("SELECT 1 FROM pg_extension WHERE extname = 'vchord_bm25'")).fetchone()
+            if not result:
+                # Extension truly doesn't exist - re-raise the error
+                raise
+        return "vchord"
+    elif text_search_extension == "pg_textsearch":
+        # Create pg_textsearch extension if not exists
+        try:
+            op.execute("CREATE EXTENSION IF NOT EXISTS pg_textsearch CASCADE")
+        except Exception:
+            # Extension might already exist or user lacks permissions - verify it exists
+            conn = op.get_bind()
+            result = conn.execute(text("SELECT 1 FROM pg_extension WHERE extname = 'pg_textsearch'")).fetchone()
+            if not result:
+                # Extension truly doesn't exist - re-raise the error
+                raise
+        return "pg_textsearch"
+    elif text_search_extension == "native":
+        return "native"
+    else:
+        raise ValueError(
+            f"Invalid HINDSIGHT_API_TEXT_SEARCH_EXTENSION: {text_search_extension}. Must be 'native', 'vchord', or 'pg_textsearch'"
+        )
+
+
+def upgrade() -> None:
+    """Upgrade schema - create all tables from scratch."""
+
+    # Note: pgvector extension is installed globally BEFORE migrations run
+    # See migrations.py:run_migrations() - this ensures the extension is available
+    # to all schemas, not just the one being migrated
+
+    # We keep this here as a fallback for backwards compatibility
+    # This may fail if user lacks permissions, which is fine if extension already exists
+    try:
+        op.execute("CREATE EXTENSION IF NOT EXISTS vector")
+    except Exception:
+        # Extension might already exist or user lacks permissions - verify it exists
+        conn = op.get_bind()
+        result = conn.execute(text("SELECT 1 FROM pg_extension WHERE extname = 'vector'")).fetchone()
+        if not result:
+            # Extension truly doesn't exist - re-raise the error
+            raise
+
+    # Create banks table
+    op.create_table(
+        "banks",
+        sa.Column("bank_id", sa.Text(), nullable=False),
+        sa.Column("name", sa.Text(), nullable=True),
+        sa.Column(
+            "personality",
+            postgresql.JSONB(astext_type=sa.Text()),
+            server_default=sa.text("'{}'::jsonb"),
+            nullable=False,
+        ),
+        sa.Column("background", sa.Text(), nullable=True),
+        sa.Column("created_at", postgresql.TIMESTAMP(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.Column("updated_at", postgresql.TIMESTAMP(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.PrimaryKeyConstraint("bank_id", name=op.f("pk_banks")),
+    )
+
+    # Create documents table
+    op.create_table(
+        "documents",
+        sa.Column("id", sa.Text(), nullable=False),
+        sa.Column("bank_id", sa.Text(), nullable=False),
+        sa.Column("original_text", sa.Text(), nullable=True),
+        sa.Column("content_hash", sa.Text(), nullable=True),
+        sa.Column(
+            "metadata", postgresql.JSONB(astext_type=sa.Text()), server_default=sa.text("'{}'::jsonb"), nullable=False
+        ),
+        sa.Column("created_at", postgresql.TIMESTAMP(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.Column("updated_at", postgresql.TIMESTAMP(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.PrimaryKeyConstraint("id", "bank_id", name=op.f("pk_documents")),
+    )
+    op.create_index("idx_documents_bank_id", "documents", ["bank_id"])
+    op.create_index("idx_documents_content_hash", "documents", ["content_hash"])
+
+    # Create async_operations table
+    op.create_table(
+        "async_operations",
+        sa.Column(
+            "operation_id", postgresql.UUID(as_uuid=True), server_default=sa.text("gen_random_uuid()"), nullable=False
+        ),
+        sa.Column("bank_id", sa.Text(), nullable=False),
+        sa.Column("operation_type", sa.Text(), nullable=False),
+        sa.Column("status", sa.Text(), server_default="pending", nullable=False),
+        sa.Column("created_at", postgresql.TIMESTAMP(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.Column("updated_at", postgresql.TIMESTAMP(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.Column("completed_at", postgresql.TIMESTAMP(timezone=True), nullable=True),
+        sa.Column("error_message", sa.Text(), nullable=True),
+        sa.Column(
+            "result_metadata",
+            postgresql.JSONB(astext_type=sa.Text()),
+            server_default=sa.text("'{}'::jsonb"),
+            nullable=False,
+        ),
+        sa.PrimaryKeyConstraint("operation_id", name=op.f("pk_async_operations")),
+        sa.CheckConstraint(
+            "status IN ('pending', 'processing', 'completed', 'failed')", name="async_operations_status_check"
+        ),
+    )
+    op.create_index("idx_async_operations_bank_id", "async_operations", ["bank_id"])
+    op.create_index("idx_async_operations_status", "async_operations", ["status"])
+    op.create_index("idx_async_operations_bank_status", "async_operations", ["bank_id", "status"])
+
+    # Create entities table
+    op.create_table(
+        "entities",
+        sa.Column("id", postgresql.UUID(as_uuid=True), server_default=sa.text("gen_random_uuid()"), nullable=False),
+        sa.Column("canonical_name", sa.Text(), nullable=False),
+        sa.Column("bank_id", sa.Text(), nullable=False),
+        sa.Column(
+            "metadata", postgresql.JSONB(astext_type=sa.Text()), server_default=sa.text("'{}'::jsonb"), nullable=False
+        ),
+        sa.Column("first_seen", postgresql.TIMESTAMP(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.Column("last_seen", postgresql.TIMESTAMP(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.Column("mention_count", sa.Integer(), server_default="1", nullable=False),
+        sa.PrimaryKeyConstraint("id", name=op.f("pk_entities")),
+    )
+    op.create_index("idx_entities_bank_id", "entities", ["bank_id"])
+    op.create_index("idx_entities_canonical_name", "entities", ["canonical_name"])
+    op.create_index("idx_entities_bank_name", "entities", ["bank_id", "canonical_name"])
+    # Create unique index on (bank_id, LOWER(canonical_name)) for entity resolution
+    op.execute("CREATE UNIQUE INDEX idx_entities_bank_lower_name ON entities (bank_id, LOWER(canonical_name))")
+
+    # Create memory_units table
+    op.create_table(
+        "memory_units",
+        sa.Column("id", postgresql.UUID(as_uuid=True), server_default=sa.text("gen_random_uuid()"), nullable=False),
+        sa.Column("bank_id", sa.Text(), nullable=False),
+        sa.Column("document_id", sa.Text(), nullable=True),
+        sa.Column("text", sa.Text(), nullable=False),
+        sa.Column("embedding", Vector(384), nullable=True),
+        sa.Column("context", sa.Text(), nullable=True),
+        sa.Column("event_date", postgresql.TIMESTAMP(timezone=True), nullable=False),
+        sa.Column("occurred_start", postgresql.TIMESTAMP(timezone=True), nullable=True),
+        sa.Column("occurred_end", postgresql.TIMESTAMP(timezone=True), nullable=True),
+        sa.Column("mentioned_at", postgresql.TIMESTAMP(timezone=True), nullable=True),
+        sa.Column("fact_type", sa.Text(), server_default="world", nullable=False),
+        sa.Column("confidence_score", sa.Float(), nullable=True),
+        sa.Column("access_count", sa.Integer(), server_default="0", nullable=False),
+        sa.Column(
+            "metadata", postgresql.JSONB(astext_type=sa.Text()), server_default=sa.text("'{}'::jsonb"), nullable=False
+        ),
+        sa.Column("created_at", postgresql.TIMESTAMP(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.Column("updated_at", postgresql.TIMESTAMP(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.ForeignKeyConstraint(
+            ["document_id", "bank_id"],
+            ["documents.id", "documents.bank_id"],
+            name="memory_units_document_fkey",
+            ondelete="CASCADE",
+        ),
+        sa.PrimaryKeyConstraint("id", name=op.f("pk_memory_units")),
+        sa.CheckConstraint(
+            "fact_type IN ('world', 'bank', 'opinion', 'observation')", name="memory_units_fact_type_check"
+        ),
+        sa.CheckConstraint(
+            "confidence_score IS NULL OR (confidence_score >= 0.0 AND confidence_score <= 1.0)",
+            name="memory_units_confidence_range_check",
+        ),
+        sa.CheckConstraint(
+            "(fact_type = 'opinion' AND confidence_score IS NOT NULL) OR "
+            "(fact_type = 'observation') OR "
+            "(fact_type NOT IN ('opinion', 'observation') AND confidence_score IS NULL)",
+            name="confidence_score_fact_type_check",
+        ),
+    )
+
+    # Add search_vector column for full-text search
+    # Type depends on configured text search backend
+    text_search_ext = _detect_text_search_extension()
+
+    if text_search_ext == "vchord":
+        # VectorChord BM25: bm25vector type (no GENERATED - tokenization happens on INSERT)
+        # Note: vchord_bm25 extension creates types in bm25_catalog schema
+        op.execute("""
+            ALTER TABLE memory_units
+            ADD COLUMN search_vector bm25_catalog.bm25vector
+        """)
+    elif text_search_ext == "pg_textsearch":
+        # Timescale pg_textsearch: dummy TEXT column for consistency (indexes operate on base columns directly)
+        op.execute("""
+            ALTER TABLE memory_units
+            ADD COLUMN search_vector TEXT
+        """)
+    else:  # native
+        # Native PostgreSQL: tsvector with automatic generation
+        op.execute("""
+            ALTER TABLE memory_units
+            ADD COLUMN search_vector tsvector
+            GENERATED ALWAYS AS (to_tsvector('english', COALESCE(text, '') || ' ' || COALESCE(context, ''))) STORED
+        """)
+
+    op.create_index("idx_memory_units_bank_id", "memory_units", ["bank_id"])
+    op.create_index("idx_memory_units_document_id", "memory_units", ["document_id"])
+    op.create_index("idx_memory_units_event_date", "memory_units", [sa.text("event_date DESC")])
+    op.create_index("idx_memory_units_bank_date", "memory_units", ["bank_id", sa.text("event_date DESC")])
+    op.create_index("idx_memory_units_access_count", "memory_units", [sa.text("access_count DESC")])
+    op.create_index("idx_memory_units_fact_type", "memory_units", ["fact_type"])
+    op.create_index("idx_memory_units_bank_fact_type", "memory_units", ["bank_id", "fact_type"])
+    op.create_index(
+        "idx_memory_units_bank_type_date", "memory_units", ["bank_id", "fact_type", sa.text("event_date DESC")]
+    )
+    op.create_index(
+        "idx_memory_units_opinion_confidence",
+        "memory_units",
+        ["bank_id", sa.text("confidence_score DESC")],
+        postgresql_where=sa.text("fact_type = 'opinion'"),
+    )
+    op.create_index(
+        "idx_memory_units_opinion_date",
+        "memory_units",
+        ["bank_id", sa.text("event_date DESC")],
+        postgresql_where=sa.text("fact_type = 'opinion'"),
+    )
+    op.create_index(
+        "idx_memory_units_observation_date",
+        "memory_units",
+        ["bank_id", sa.text("event_date DESC")],
+        postgresql_where=sa.text("fact_type = 'observation'"),
+    )
+    # Create vector index - conditional based on available extension
+    vector_ext = _detect_vector_extension()
+
+    if vector_ext == "vchord":
+        # Use vchordrq index for vchord (supports high-dimensional embeddings)
+        op.execute("""
+            CREATE INDEX idx_memory_units_embedding ON memory_units
+            USING vchordrq (embedding vector_l2_ops)
+        """)
+    else:  # pgvector
+        # Use HNSW index for pgvector
+        op.create_index(
+            "idx_memory_units_embedding",
+            "memory_units",
+            ["embedding"],
+            postgresql_using="hnsw",
+            postgresql_ops={"embedding": "vector_cosine_ops"},
+        )
+
+    # Create full-text search index on search_vector
+    # Index type depends on text search backend
+    if text_search_ext == "vchord":
+        # VectorChord BM25 index
+        op.execute("""
+            CREATE INDEX idx_memory_units_text_search ON memory_units
+            USING bm25 (search_vector bm25_catalog.bm25_ops)
+        """)
+    elif text_search_ext == "pg_textsearch":
+        # Timescale pg_textsearch BM25 index on text column
+        # Note: pg_textsearch doesn't support expressions, so we index the main text column
+        op.execute("""
+            CREATE INDEX idx_memory_units_text_search ON memory_units
+            USING bm25(text)
+            WITH (text_config='english')
+        """)
+    else:  # native
+        # Native PostgreSQL GIN index
+        op.execute("""
+            CREATE INDEX idx_memory_units_text_search ON memory_units
+            USING gin(search_vector)
+        """)
+
+    op.execute("""
+        CREATE MATERIALIZED VIEW memory_units_bm25 AS
+        SELECT
+            id,
+            bank_id,
+            text,
+            to_tsvector('english', text) AS text_vector,
+            log(1.0 + length(text)::float / (SELECT avg(length(text)) FROM memory_units)) AS doc_length_factor
+        FROM memory_units
+    """)
+
+    op.create_index("idx_memory_units_bm25_bank", "memory_units_bm25", ["bank_id"])
+    op.create_index("idx_memory_units_bm25_text_vector", "memory_units_bm25", ["text_vector"], postgresql_using="gin")
+
+    # Create entity_cooccurrences table
+    op.create_table(
+        "entity_cooccurrences",
+        sa.Column("entity_id_1", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("entity_id_2", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("cooccurrence_count", sa.Integer(), server_default="1", nullable=False),
+        sa.Column(
+            "last_cooccurred", postgresql.TIMESTAMP(timezone=True), server_default=sa.text("now()"), nullable=False
+        ),
+        sa.ForeignKeyConstraint(
+            ["entity_id_1"],
+            ["entities.id"],
+            name=op.f("fk_entity_cooccurrences_entity_id_1_entities"),
+            ondelete="CASCADE",
+        ),
+        sa.ForeignKeyConstraint(
+            ["entity_id_2"],
+            ["entities.id"],
+            name=op.f("fk_entity_cooccurrences_entity_id_2_entities"),
+            ondelete="CASCADE",
+        ),
+        sa.PrimaryKeyConstraint("entity_id_1", "entity_id_2", name=op.f("pk_entity_cooccurrences")),
+        sa.CheckConstraint("entity_id_1 < entity_id_2", name="entity_cooccurrence_order_check"),
+    )
+    op.create_index("idx_entity_cooccurrences_entity1", "entity_cooccurrences", ["entity_id_1"])
+    op.create_index("idx_entity_cooccurrences_entity2", "entity_cooccurrences", ["entity_id_2"])
+    op.create_index("idx_entity_cooccurrences_count", "entity_cooccurrences", [sa.text("cooccurrence_count DESC")])
+
+    # Create memory_links table
+    op.create_table(
+        "memory_links",
+        sa.Column("from_unit_id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("to_unit_id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("link_type", sa.Text(), nullable=False),
+        sa.Column("entity_id", postgresql.UUID(as_uuid=True), nullable=True),
+        sa.Column("weight", sa.Float(), server_default="1.0", nullable=False),
+        sa.Column("created_at", postgresql.TIMESTAMP(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.ForeignKeyConstraint(
+            ["entity_id"], ["entities.id"], name=op.f("fk_memory_links_entity_id_entities"), ondelete="CASCADE"
+        ),
+        sa.ForeignKeyConstraint(
+            ["from_unit_id"],
+            ["memory_units.id"],
+            name=op.f("fk_memory_links_from_unit_id_memory_units"),
+            ondelete="CASCADE",
+        ),
+        sa.ForeignKeyConstraint(
+            ["to_unit_id"],
+            ["memory_units.id"],
+            name=op.f("fk_memory_links_to_unit_id_memory_units"),
+            ondelete="CASCADE",
+        ),
+        sa.CheckConstraint(
+            "link_type IN ('temporal', 'semantic', 'entity', 'causes', 'caused_by', 'enables', 'prevents')",
+            name="memory_links_link_type_check",
+        ),
+        sa.CheckConstraint("weight >= 0.0 AND weight <= 1.0", name="memory_links_weight_check"),
+    )
+    # Create unique constraint using COALESCE for nullable entity_id
+    op.execute(
+        "CREATE UNIQUE INDEX idx_memory_links_unique ON memory_links (from_unit_id, to_unit_id, link_type, COALESCE(entity_id, '00000000-0000-0000-0000-000000000000'::uuid))"
+    )
+    op.create_index("idx_memory_links_from_unit", "memory_links", ["from_unit_id"])
+    op.create_index("idx_memory_links_to_unit", "memory_links", ["to_unit_id"])
+    op.create_index("idx_memory_links_entity", "memory_links", ["entity_id"])
+    op.create_index("idx_memory_links_link_type", "memory_links", ["link_type"])
+
+    # Create unit_entities table
+    op.create_table(
+        "unit_entities",
+        sa.Column("unit_id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("entity_id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.ForeignKeyConstraint(
+            ["entity_id"], ["entities.id"], name=op.f("fk_unit_entities_entity_id_entities"), ondelete="CASCADE"
+        ),
+        sa.ForeignKeyConstraint(
+            ["unit_id"], ["memory_units.id"], name=op.f("fk_unit_entities_unit_id_memory_units"), ondelete="CASCADE"
+        ),
+        sa.PrimaryKeyConstraint("unit_id", "entity_id", name=op.f("pk_unit_entities")),
+    )
+    op.create_index("idx_unit_entities_unit", "unit_entities", ["unit_id"])
+    op.create_index("idx_unit_entities_entity", "unit_entities", ["entity_id"])
+
+
+def downgrade() -> None:
+    """Downgrade schema - drop all tables."""
+
+    # Drop tables in reverse dependency order
+    op.drop_index("idx_unit_entities_entity", table_name="unit_entities")
+    op.drop_index("idx_unit_entities_unit", table_name="unit_entities")
+    op.drop_table("unit_entities")
+
+    op.drop_index("idx_memory_links_link_type", table_name="memory_links")
+    op.drop_index("idx_memory_links_entity", table_name="memory_links")
+    op.drop_index("idx_memory_links_to_unit", table_name="memory_links")
+    op.drop_index("idx_memory_links_from_unit", table_name="memory_links")
+    op.execute("DROP INDEX IF EXISTS idx_memory_links_unique")
+    op.drop_table("memory_links")
+
+    op.drop_index("idx_entity_cooccurrences_count", table_name="entity_cooccurrences")
+    op.drop_index("idx_entity_cooccurrences_entity2", table_name="entity_cooccurrences")
+    op.drop_index("idx_entity_cooccurrences_entity1", table_name="entity_cooccurrences")
+    op.drop_table("entity_cooccurrences")
+
+    # Drop BM25 materialized view and index
+    op.drop_index("idx_memory_units_bm25_text_vector", table_name="memory_units_bm25")
+    op.drop_index("idx_memory_units_bm25_bank", table_name="memory_units_bm25")
+    op.execute("DROP MATERIALIZED VIEW IF EXISTS memory_units_bm25")
+
+    op.drop_index("idx_memory_units_embedding", table_name="memory_units")
+    op.drop_index("idx_memory_units_observation_date", table_name="memory_units")
+    op.drop_index("idx_memory_units_opinion_date", table_name="memory_units")
+    op.drop_index("idx_memory_units_opinion_confidence", table_name="memory_units")
+    op.drop_index("idx_memory_units_bank_type_date", table_name="memory_units")
+    op.drop_index("idx_memory_units_bank_fact_type", table_name="memory_units")
+    op.drop_index("idx_memory_units_fact_type", table_name="memory_units")
+    op.drop_index("idx_memory_units_access_count", table_name="memory_units")
+    op.drop_index("idx_memory_units_bank_date", table_name="memory_units")
+    op.drop_index("idx_memory_units_event_date", table_name="memory_units")
+    op.drop_index("idx_memory_units_document_id", table_name="memory_units")
+    op.drop_index("idx_memory_units_bank_id", table_name="memory_units")
+    op.execute("DROP INDEX IF EXISTS idx_memory_units_text_search")
+    op.drop_table("memory_units")
+
+    op.execute("DROP INDEX IF EXISTS idx_entities_bank_lower_name")
+    op.drop_index("idx_entities_bank_name", table_name="entities")
+    op.drop_index("idx_entities_canonical_name", table_name="entities")
+    op.drop_index("idx_entities_bank_id", table_name="entities")
+    op.drop_table("entities")
+
+    op.drop_index("idx_async_operations_bank_status", table_name="async_operations")
+    op.drop_index("idx_async_operations_status", table_name="async_operations")
+    op.drop_index("idx_async_operations_bank_id", table_name="async_operations")
+    op.drop_table("async_operations")
+
+    op.drop_index("idx_documents_content_hash", table_name="documents")
+    op.drop_index("idx_documents_bank_id", table_name="documents")
+    op.drop_table("documents")
+
+    op.drop_table("banks")
+
+    # Drop extensions (optional - comment out if you want to keep them)
+    # op.execute('DROP EXTENSION IF EXISTS vector')
+    # op.execute('DROP EXTENSION IF EXISTS "uuid-ossp"')

hindsight-api/hindsight_api/alembic/versions/b7c4d8e9f1a2_add_chunks_table.py

@@ -0,0 +1,70 @@
+"""add_chunks_table
+
+Revision ID: b7c4d8e9f1a2
+Revises: 5a366d414dce
+Create Date: 2025-11-28 00:00:00.000000
+
+"""
+
+from collections.abc import Sequence
+
+import sqlalchemy as sa
+from alembic import op
+from sqlalchemy.dialects import postgresql
+
+# revision identifiers, used by Alembic.
+revision: str = "b7c4d8e9f1a2"
+down_revision: str | Sequence[str] | None = "5a366d414dce"
+branch_labels: str | Sequence[str] | None = None
+depends_on: str | Sequence[str] | None = None
+
+
+def upgrade() -> None:
+    """Add chunks table and link memory_units to chunks."""
+
+    # Create chunks table with single text PK (bank_id_document_id_chunk_index)
+    op.create_table(
+        "chunks",
+        sa.Column("chunk_id", sa.Text(), nullable=False),
+        sa.Column("document_id", sa.Text(), nullable=False),
+        sa.Column("bank_id", sa.Text(), nullable=False),
+        sa.Column("chunk_index", sa.Integer(), nullable=False),
+        sa.Column("chunk_text", sa.Text(), nullable=False),
+        sa.Column("created_at", postgresql.TIMESTAMP(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.ForeignKeyConstraint(
+            ["document_id", "bank_id"],
+            ["documents.id", "documents.bank_id"],
+            name="chunks_document_fkey",
+            ondelete="CASCADE",
+        ),
+        sa.PrimaryKeyConstraint("chunk_id", name=op.f("pk_chunks")),
+    )
+
+    # Add indexes for efficient queries
+    op.create_index("idx_chunks_document_id", "chunks", ["document_id"])
+    op.create_index("idx_chunks_bank_id", "chunks", ["bank_id"])
+
+    # Add chunk_id column to memory_units (nullable, as existing records won't have chunks)
+    op.add_column("memory_units", sa.Column("chunk_id", sa.Text(), nullable=True))
+
+    # Add foreign key constraint to chunks table
+    op.create_foreign_key(
+        "memory_units_chunk_fkey", "memory_units", "chunks", ["chunk_id"], ["chunk_id"], ondelete="SET NULL"
+    )
+
+    # Add index on chunk_id for efficient lookups
+    op.create_index("idx_memory_units_chunk_id", "memory_units", ["chunk_id"])
+
+
+def downgrade() -> None:
+    """Remove chunks table and chunk_id from memory_units."""
+
+    # Drop index and foreign key from memory_units
+    op.drop_index("idx_memory_units_chunk_id", table_name="memory_units")
+    op.drop_constraint("memory_units_chunk_fkey", "memory_units", type_="foreignkey")
+    op.drop_column("memory_units", "chunk_id")
+
+    # Drop chunks table indexes and table
+    op.drop_index("idx_chunks_bank_id", table_name="chunks")
+    op.drop_index("idx_chunks_document_id", table_name="chunks")
+    op.drop_table("chunks")

hindsight-api/hindsight_api/alembic/versions/c8e5f2a3b4d1_add_retain_params_to_documents.py

@@ -5,35 +5,35 @@ Revises: b7c4d8e9f1a2
 Create Date: 2025-12-02 00:00:00.000000
 
 """
-from typing import Sequence, Union
 
-from alembic import op
+from collections.abc import Sequence
+
 import sqlalchemy as sa
+from alembic import op
 from sqlalchemy.dialects import postgresql
 
-
 # revision identifiers, used by Alembic.
-revision: str = 'c8e5f2a3b4d1'
-down_revision: Union[str, Sequence[str], None] = 'b7c4d8e9f1a2'
-branch_labels: Union[str, Sequence[str], None] = None
-depends_on: Union[str, Sequence[str], None] = None
+revision: str = "c8e5f2a3b4d1"
+down_revision: str | Sequence[str] | None = "b7c4d8e9f1a2"
+branch_labels: str | Sequence[str] | None = None
+depends_on: str | Sequence[str] | None = None
 
 
 def upgrade() -> None:
     """Add retain_params JSONB column to documents table."""
 
     # Add retain_params column to store parameters passed during retain
-    op.add_column('documents', sa.Column('retain_params', postgresql.JSONB(), nullable=True))
+    op.add_column("documents", sa.Column("retain_params", postgresql.JSONB(), nullable=True))
 
     # Add index for efficient queries on retain_params
-    op.create_index('idx_documents_retain_params', 'documents', ['retain_params'], postgresql_using='gin')
+    op.create_index("idx_documents_retain_params", "documents", ["retain_params"], postgresql_using="gin")
 
 
 def downgrade() -> None:
     """Remove retain_params column from documents table."""
 
     # Drop index
-    op.drop_index('idx_documents_retain_params', table_name='documents')
+    op.drop_index("idx_documents_retain_params", table_name="documents")
 
     # Drop column
-    op.drop_column('documents', 'retain_params')
+    op.drop_column("documents", "retain_params")

hindsight-api/hindsight_api/alembic/versions/d9f6a3b4c5e2_rename_bank_to_interactions.py

@@ -0,0 +1,53 @@
+"""Rename fact_type 'bank' to 'experience'
+
+Revision ID: d9f6a3b4c5e2
+Revises: c8e5f2a3b4d1
+Create Date: 2024-12-04 15:00:00.000000
+
+"""
+
+from alembic import context, op
+
+# revision identifiers, used by Alembic.
+revision = "d9f6a3b4c5e2"
+down_revision = "c8e5f2a3b4d1"
+branch_labels = None
+depends_on = None
+
+
+def _get_schema_prefix() -> str:
+    """Get schema prefix for table names (e.g., 'tenant_x.' or '' for public)."""
+    schema = context.config.get_main_option("target_schema")
+    return f'"{schema}".' if schema else ""
+
+
+def upgrade():
+    schema = _get_schema_prefix()
+
+    # Drop old check constraint FIRST (before updating data)
+    op.drop_constraint("memory_units_fact_type_check", "memory_units", type_="check")
+
+    # Update existing 'bank' values to 'experience'
+    op.execute(f"UPDATE {schema}memory_units SET fact_type = 'experience' WHERE fact_type = 'bank'")
+    # Also update any 'interactions' values (in case of partial migration)
+    op.execute(f"UPDATE {schema}memory_units SET fact_type = 'experience' WHERE fact_type = 'interactions'")
+
+    # Create new check constraint with 'experience' instead of 'bank'
+    op.create_check_constraint(
+        "memory_units_fact_type_check", "memory_units", "fact_type IN ('world', 'experience', 'opinion', 'observation')"
+    )
+
+
+def downgrade():
+    schema = _get_schema_prefix()
+
+    # Drop new check constraint FIRST
+    op.drop_constraint("memory_units_fact_type_check", "memory_units", type_="check")
+
+    # Update 'experience' back to 'bank'
+    op.execute(f"UPDATE {schema}memory_units SET fact_type = 'bank' WHERE fact_type = 'experience'")
+
+    # Recreate old check constraint
+    op.create_check_constraint(
+        "memory_units_fact_type_check", "memory_units", "fact_type IN ('world', 'bank', 'opinion', 'observation')"
+    )

hindsight-api/hindsight_api/alembic/versions/e0a1b2c3d4e5_disposition_to_3_traits.py

@@ -0,0 +1,111 @@
+"""disposition_to_3_traits
+
+Revision ID: e0a1b2c3d4e5
+Revises: rename_personality
+Create Date: 2024-12-08
+
+Migrate disposition traits from Big Five (openness, conscientiousness, extraversion,
+agreeableness, neuroticism, bias_strength with 0-1 float values) to the new 3-trait
+system (skepticism, literalism, empathy with 1-5 integer values).
+"""
+
+from collections.abc import Sequence
+
+import sqlalchemy as sa
+from alembic import context, op
+
+# revision identifiers, used by Alembic.
+revision: str = "e0a1b2c3d4e5"
+down_revision: str | Sequence[str] | None = "rename_personality"
+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 (e.g., 'tenant_x.' or '' for public)."""
+    schema = context.config.get_main_option("target_schema")
+    return f'"{schema}".' if schema else ""
+
+
+def _get_target_schema() -> str:
+    """Get the target schema name (tenant schema or 'public')."""
+    schema = context.config.get_main_option("target_schema")
+    return schema if schema else "public"
+
+
+def upgrade() -> None:
+    """Convert Big Five disposition to 3-trait disposition."""
+    conn = op.get_bind()
+    schema = _get_schema_prefix()
+    target_schema = _get_target_schema()
+
+    # Check if disposition column exists (should have been created by previous migration)
+    result = conn.execute(
+        sa.text("""
+        SELECT column_name
+        FROM information_schema.columns
+        WHERE table_schema = :schema AND table_name = 'banks' AND column_name = 'disposition'
+    """),
+        {"schema": target_schema},
+    )
+    if not result.fetchone():
+        # Column doesn't exist yet (shouldn't happen but be safe)
+        return
+
+    # Update all existing banks to use the new disposition format
+    # Convert from old format to new format with reasonable mappings:
+    # - skepticism: derived from inverse of agreeableness (skeptical people are less agreeable)
+    # - literalism: derived from conscientiousness (detail-oriented people are more literal)
+    # - empathy: derived from agreeableness + inverse of neuroticism
+    # Default all to 3 (neutral) for simplicity
+    conn.execute(
+        sa.text(f"""
+        UPDATE {schema}banks
+        SET disposition = '{{"skepticism": 3, "literalism": 3, "empathy": 3}}'::jsonb
+        WHERE disposition IS NOT NULL
+    """)
+    )
+
+    # Update the default for new banks
+    conn.execute(
+        sa.text(f"""
+        ALTER TABLE {schema}banks
+        ALTER COLUMN disposition SET DEFAULT '{{"skepticism": 3, "literalism": 3, "empathy": 3}}'::jsonb
+    """)
+    )
+
+
+def downgrade() -> None:
+    """Convert back to Big Five disposition."""
+    conn = op.get_bind()
+    schema = _get_schema_prefix()
+    target_schema = _get_target_schema()
+
+    # Check if disposition column exists
+    result = conn.execute(
+        sa.text("""
+        SELECT column_name
+        FROM information_schema.columns
+        WHERE table_schema = :schema AND table_name = 'banks' AND column_name = 'disposition'
+    """),
+        {"schema": target_schema},
+    )
+    if not result.fetchone():
+        return
+
+    # Revert to Big Five format with default values
+    conn.execute(
+        sa.text(f"""
+        UPDATE {schema}banks
+        SET disposition = '{{"openness": 0.5, "conscientiousness": 0.5, "extraversion": 0.5, "agreeableness": 0.5, "neuroticism": 0.5, "bias_strength": 0.5}}'::jsonb
+        WHERE disposition IS NOT NULL
+    """)
+    )
+
+    # Update the default for new banks
+    conn.execute(
+        sa.text(f"""
+        ALTER TABLE {schema}banks
+        ALTER COLUMN disposition SET DEFAULT '{{"openness": 0.5, "conscientiousness": 0.5, "extraversion": 0.5, "agreeableness": 0.5, "neuroticism": 0.5, "bias_strength": 0.5}}'::jsonb
+    """)
+    )

hindsight-api/hindsight_api/alembic/versions/f1a2b3c4d5e6_add_memory_links_composite_index.py

@@ -0,0 +1,44 @@
+"""add_memory_links_from_type_weight_index
+
+Revision ID: f1a2b3c4d5e6
+Revises: e0a1b2c3d4e5
+Create Date: 2025-01-12
+
+Add composite index on memory_links (from_unit_id, link_type, weight DESC)
+to optimize MPFP graph traversal queries that need top-k edges per type.
+"""
+
+from collections.abc import Sequence
+
+from alembic import context, op
+
+# revision identifiers, used by Alembic.
+revision: str = "f1a2b3c4d5e6"
+down_revision: str | Sequence[str] | None = "e0a1b2c3d4e5"
+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 (e.g., 'tenant_x.' or '' for public)."""
+    schema = context.config.get_main_option("target_schema")
+    return f'"{schema}".' if schema else ""
+
+
+def upgrade() -> None:
+    """Add composite index for efficient MPFP edge loading."""
+    schema = _get_schema_prefix()
+    # Create composite index for efficient top-k per (from_node, link_type) queries
+    # This enables LATERAL joins to use index-only scans with early termination
+    # Note: Not using CONCURRENTLY here as it requires running outside a transaction
+    # For production with large tables, consider running this manually with CONCURRENTLY
+    op.execute(
+        f"CREATE INDEX IF NOT EXISTS idx_memory_links_from_type_weight "
+        f"ON {schema}memory_links(from_unit_id, link_type, weight DESC)"
+    )
+
+
+def downgrade() -> None:
+    """Remove the composite index."""
+    schema = _get_schema_prefix()
+    op.execute(f"DROP INDEX IF EXISTS {schema}idx_memory_links_from_type_weight")

hindsight-api/hindsight_api/alembic/versions/g2a3b4c5d6e7_add_tags_column.py

@@ -0,0 +1,48 @@
+"""add_tags_column
+
+Revision ID: g2a3b4c5d6e7
+Revises: f1a2b3c4d5e6
+Create Date: 2025-01-13
+
+Add tags column to memory_units and documents tables for visibility scoping.
+Tags enable filtering memories by scope (e.g., user IDs, session IDs) during recall/reflect.
+"""
+
+from collections.abc import Sequence
+
+from alembic import context, op
+
+# revision identifiers, used by Alembic.
+revision: str = "g2a3b4c5d6e7"
+down_revision: str | Sequence[str] | None = "f1a2b3c4d5e6"
+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 (e.g., 'tenant_x.' or '' for public)."""
+    schema = context.config.get_main_option("target_schema")
+    return f'"{schema}".' if schema else ""
+
+
+def upgrade() -> None:
+    """Add tags column to memory_units and documents tables."""
+    schema = _get_schema_prefix()
+
+    # Add tags column to memory_units table
+    op.execute(f"ALTER TABLE {schema}memory_units ADD COLUMN IF NOT EXISTS tags VARCHAR[] NOT NULL DEFAULT '{{}}'")
+
+    # Create GIN index for efficient array containment queries (tags && ARRAY['x'])
+    op.execute(f"CREATE INDEX IF NOT EXISTS idx_memory_units_tags ON {schema}memory_units USING GIN (tags)")
+
+    # Add tags column to documents table for document-level tags
+    op.execute(f"ALTER TABLE {schema}documents ADD COLUMN IF NOT EXISTS tags VARCHAR[] NOT NULL DEFAULT '{{}}'")
+
+
+def downgrade() -> None:
+    """Remove tags columns and index."""
+    schema = _get_schema_prefix()
+
+    op.execute(f"DROP INDEX IF EXISTS {schema}idx_memory_units_tags")
+    op.execute(f"ALTER TABLE {schema}memory_units DROP COLUMN IF EXISTS tags")
+    op.execute(f"ALTER TABLE {schema}documents DROP COLUMN IF EXISTS tags")

hindsight-api/hindsight_api/alembic/versions/h3c4d5e6f7g8_mental_models_v4.py

@@ -0,0 +1,112 @@
+"""mental_models_v4
+
+Revision ID: h3c4d5e6f7g8
+Revises: g2a3b4c5d6e7
+Create Date: 2026-01-08 00:00:00.000000
+
+This migration implements the v4 mental models system:
+1. Deletes existing observation memory_units (observations now in mental models)
+2. Adds mission column to banks (replacing background)
+3. Creates mental_models table with final schema
+
+Mental models can reference entities when an entity is "promoted" to a mental model.
+Summary content is stored as JSONB observations with per-observation fact attribution.
+"""
+
+from collections.abc import Sequence
+
+from alembic import context, op
+
+# revision identifiers, used by Alembic.
+revision: str = "h3c4d5e6f7g8"
+down_revision: str | Sequence[str] | None = "g2a3b4c5d6e7"
+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:
+    """Apply mental models v4 changes."""
+    schema = _get_schema_prefix()
+
+    # Step 1: Delete observation memory_units (cascades to unit_entities links)
+    # Observations are now handled through mental models, not memory_units
+    op.execute(f"DELETE FROM {schema}memory_units WHERE fact_type = 'observation'")
+
+    # Step 2: Drop observation-specific index (if it exists)
+    op.execute(f"DROP INDEX IF EXISTS {schema}idx_memory_units_observation_date")
+
+    # Step 3: Add mission column to banks (replacing background)
+    op.execute(f"ALTER TABLE {schema}banks ADD COLUMN IF NOT EXISTS mission TEXT")
+
+    # Migrate: copy background to mission if background column exists
+    # Use DO block to check column existence first (idempotent for re-runs)
+    schema_name = context.config.get_main_option("target_schema") or "public"
+    op.execute(f"""
+        DO $$
+        BEGIN
+            IF EXISTS (
+                SELECT 1 FROM information_schema.columns
+                WHERE table_schema = '{schema_name}' AND table_name = 'banks' AND column_name = 'background'
+            ) THEN
+                UPDATE {schema}banks
+                SET mission = background
+                WHERE mission IS NULL;
+            END IF;
+        END $$;
+    """)
+
+    # Remove background column (replaced by mission)
+    op.execute(f"ALTER TABLE {schema}banks DROP COLUMN IF EXISTS background")
+
+    # Step 4: Create mental_models table with final v4 schema (if not exists)
+    op.execute(f"""
+        CREATE TABLE IF NOT EXISTS {schema}mental_models (
+            id VARCHAR(64) NOT NULL,
+            bank_id VARCHAR(64) NOT NULL,
+            subtype VARCHAR(32) NOT NULL,
+            name VARCHAR(256) NOT NULL,
+            description TEXT NOT NULL,
+            entity_id UUID,
+            observations JSONB DEFAULT '{{"observations": []}}'::jsonb,
+            links VARCHAR[],
+            tags VARCHAR[] DEFAULT '{{}}',
+            last_updated TIMESTAMP WITH TIME ZONE,
+            created_at TIMESTAMP WITH TIME ZONE NOT NULL DEFAULT now(),
+            PRIMARY KEY (id, bank_id),
+            FOREIGN KEY (bank_id) REFERENCES {schema}banks(bank_id) ON DELETE CASCADE,
+            FOREIGN KEY (entity_id) REFERENCES {schema}entities(id) ON DELETE SET NULL,
+            CONSTRAINT ck_mental_models_subtype CHECK (subtype IN ('structural', 'emergent', 'pinned', 'learned'))
+        )
+    """)
+
+    # Step 5: Create indexes for efficient queries (if not exist)
+    op.execute(f"CREATE INDEX IF NOT EXISTS idx_mental_models_bank_id ON {schema}mental_models(bank_id)")
+    op.execute(f"CREATE INDEX IF NOT EXISTS idx_mental_models_subtype ON {schema}mental_models(bank_id, subtype)")
+    op.execute(f"CREATE INDEX IF NOT EXISTS idx_mental_models_entity_id ON {schema}mental_models(entity_id)")
+    # GIN index for efficient tags array filtering
+    op.execute(f"CREATE INDEX IF NOT EXISTS idx_mental_models_tags ON {schema}mental_models USING GIN(tags)")
+
+
+def downgrade() -> None:
+    """Revert mental models v4 changes."""
+    schema = _get_schema_prefix()
+
+    # Drop mental_models table (cascades to indexes)
+    op.execute(f"DROP TABLE IF EXISTS {schema}mental_models CASCADE")
+
+    # Add back background column to banks
+    op.execute(f"ALTER TABLE {schema}banks ADD COLUMN IF NOT EXISTS background TEXT")
+
+    # Migrate mission back to background
+    op.execute(f"UPDATE {schema}banks SET background = mission WHERE background IS NULL")
+
+    # Remove mission column
+    op.execute(f"ALTER TABLE {schema}banks DROP COLUMN IF EXISTS mission")
+
+    # Note: Cannot restore deleted observations - they are lost on downgrade

hindsight-api/hindsight_api/alembic/versions/i4d5e6f7g8h9_delete_opinions.py

@@ -0,0 +1,41 @@
+"""delete_opinions
+
+Revision ID: i4d5e6f7g8h9
+Revises: h3c4d5e6f7g8
+Create Date: 2026-01-15 00:00:00.000000
+
+This migration removes opinion facts from memory_units.
+Opinions are no longer a separate fact type - they are now represented
+through mental model observations with confidence scores.
+"""
+
+from collections.abc import Sequence
+
+from alembic import context, op
+
+# revision identifiers, used by Alembic.
+revision: str = "i4d5e6f7g8h9"
+down_revision: str | Sequence[str] | None = "h3c4d5e6f7g8"
+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:
+    """Delete opinion memory_units."""
+    schema = _get_schema_prefix()
+
+    # Delete opinion memory_units (cascades to unit_entities links)
+    # Opinions are now handled through mental model observations
+    op.execute(f"DELETE FROM {schema}memory_units WHERE fact_type = 'opinion'")
+
+
+def downgrade() -> None:
+    """Cannot restore deleted opinions."""
+    # Note: Cannot restore deleted opinions - they are lost on downgrade
+    pass

hindsight-api/hindsight_api/alembic/versions/j5e6f7g8h9i0_mental_model_versions.py

@@ -0,0 +1,95 @@
+"""mental_model_versions
+
+Revision ID: j5e6f7g8h9i0
+Revises: i4d5e6f7g8h9
+Create Date: 2026-01-16 00:00:00.000000
+
+This migration adds versioning support for mental models:
+1. Creates mental_model_versions table to store observation snapshots
+2. Adds version column to mental_models for tracking current version
+
+This enables changelog/diff functionality for mental model observations.
+"""
+
+from collections.abc import Sequence
+
+from alembic import context, op
+
+# revision identifiers, used by Alembic.
+revision: str = "j5e6f7g8h9i0"
+down_revision: str | Sequence[str] | None = "i4d5e6f7g8h9"
+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 mental_model_versions table and add version tracking."""
+    schema = _get_schema_prefix()
+
+    # Create mental_model_versions table for storing observation snapshots
+    op.execute(f"""
+        CREATE TABLE {schema}mental_model_versions (
+            id SERIAL PRIMARY KEY,
+            mental_model_id VARCHAR(64) NOT NULL,
+            bank_id VARCHAR(64) NOT NULL,
+            version INT NOT NULL,
+            observations JSONB NOT NULL DEFAULT '{{"observations": []}}'::jsonb,
+            created_at TIMESTAMP WITH TIME ZONE NOT NULL DEFAULT now(),
+            FOREIGN KEY (mental_model_id, bank_id)
+                REFERENCES {schema}mental_models(id, bank_id) ON DELETE CASCADE,
+            UNIQUE (mental_model_id, bank_id, version)
+        )
+    """)
+
+    # Index for efficient version queries (get latest, list versions)
+    op.execute(f"""
+        CREATE INDEX idx_mental_model_versions_lookup
+        ON {schema}mental_model_versions(mental_model_id, bank_id, version DESC)
+    """)
+
+    # Add version column to mental_models to track current version
+    op.execute(f"""
+        ALTER TABLE {schema}mental_models
+        ADD COLUMN IF NOT EXISTS version INT NOT NULL DEFAULT 0
+    """)
+
+    # Migrate existing mental models: create version 1 for any that have observations
+    op.execute(f"""
+        INSERT INTO {schema}mental_model_versions (mental_model_id, bank_id, version, observations, created_at)
+        SELECT id, bank_id, 1, observations, COALESCE(last_updated, created_at)
+        FROM {schema}mental_models
+        WHERE observations IS NOT NULL
+          AND observations != '{{"observations": []}}'::jsonb
+          AND (observations->'observations') IS NOT NULL
+          AND jsonb_array_length(observations->'observations') > 0
+    """)
+
+    # Update version to 1 for migrated mental models
+    op.execute(f"""
+        UPDATE {schema}mental_models
+        SET version = 1
+        WHERE observations IS NOT NULL
+          AND observations != '{{"observations": []}}'::jsonb
+          AND (observations->'observations') IS NOT NULL
+          AND jsonb_array_length(observations->'observations') > 0
+    """)
+
+
+def downgrade() -> None:
+    """Remove mental_model_versions table and version column."""
+    schema = _get_schema_prefix()
+
+    # Drop index
+    op.execute(f"DROP INDEX IF EXISTS {schema}idx_mental_model_versions_lookup")
+
+    # Drop versions table
+    op.execute(f"DROP TABLE IF EXISTS {schema}mental_model_versions")
+
+    # Remove version column from mental_models
+    op.execute(f"ALTER TABLE {schema}mental_models DROP COLUMN IF EXISTS version")

hindsight-api/hindsight_api/alembic/versions/k6f7g8h9i0j1_add_directive_subtype.py

@@ -0,0 +1,58 @@
+"""add_directive_subtype
+
+Revision ID: k6f7g8h9i0j1
+Revises: j5e6f7g8h9i0
+Create Date: 2026-01-16 00:00:00.000000
+
+This migration adds 'directive' to the mental_models subtype constraint.
+Directives are hard rules with user-provided observations that the reflect agent must follow.
+"""
+
+from collections.abc import Sequence
+
+from alembic import context, op
+
+# revision identifiers, used by Alembic.
+revision: str = "k6f7g8h9i0j1"
+down_revision: str | Sequence[str] | None = "j5e6f7g8h9i0"
+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 'directive' to mental_models subtype constraint."""
+    schema = _get_schema_prefix()
+
+    # Drop existing constraint
+    op.execute(f"ALTER TABLE {schema}mental_models DROP CONSTRAINT IF EXISTS ck_mental_models_subtype")
+
+    # Create new constraint with 'directive' added
+    op.execute(f"""
+        ALTER TABLE {schema}mental_models
+        ADD CONSTRAINT ck_mental_models_subtype
+        CHECK (subtype IN ('structural', 'emergent', 'pinned', 'learned', 'directive'))
+    """)
+
+
+def downgrade() -> None:
+    """Remove 'directive' from mental_models subtype constraint."""
+    schema = _get_schema_prefix()
+
+    # First delete any directives (cannot downgrade if they exist)
+    op.execute(f"DELETE FROM {schema}mental_models WHERE subtype = 'directive'")
+
+    # Drop constraint with directive
+    op.execute(f"ALTER TABLE {schema}mental_models DROP CONSTRAINT IF EXISTS ck_mental_models_subtype")
+
+    # Recreate original constraint without directive
+    op.execute(f"""
+        ALTER TABLE {schema}mental_models
+        ADD CONSTRAINT ck_mental_models_subtype
+        CHECK (subtype IN ('structural', 'emergent', 'pinned', 'learned'))
+    """)

hindsight-api/hindsight_api/alembic/versions/l7g8h9i0j1k2_add_worker_columns.py

@@ -0,0 +1,109 @@
+"""add_worker_columns
+
+Revision ID: l7g8h9i0j1k2
+Revises: k6f7g8h9i0j1
+Create Date: 2026-01-19 00:00:00.000000
+
+This migration adds columns to async_operations for distributed worker support:
+- worker_id: ID of the worker that claimed the task
+- claimed_at: When the task was claimed
+- retry_count: Number of retry attempts
+- task_payload: The serialized task dictionary
+"""
+
+from collections.abc import Sequence
+
+import sqlalchemy as sa
+from alembic import context, op
+from sqlalchemy.dialects import postgresql
+
+# revision identifiers, used by Alembic.
+revision: str = "l7g8h9i0j1k2"
+down_revision: str | Sequence[str] | None = "k6f7g8h9i0j1"
+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 worker columns to async_operations."""
+    schema = _get_schema_prefix()
+
+    # Add worker_id column (ID of worker that claimed the task)
+    op.add_column(
+        "async_operations",
+        sa.Column("worker_id", sa.Text(), nullable=True),
+        schema=context.config.get_main_option("target_schema") or None,
+    )
+
+    # Add claimed_at column (when task was claimed by worker)
+    op.add_column(
+        "async_operations",
+        sa.Column("claimed_at", postgresql.TIMESTAMP(timezone=True), nullable=True),
+        schema=context.config.get_main_option("target_schema") or None,
+    )
+
+    # Add retry_count column (number of retry attempts)
+    op.add_column(
+        "async_operations",
+        sa.Column("retry_count", sa.Integer(), server_default="0", nullable=False),
+        schema=context.config.get_main_option("target_schema") or None,
+    )
+
+    # Add task_payload column (serialized task dictionary)
+    op.add_column(
+        "async_operations",
+        sa.Column(
+            "task_payload",
+            postgresql.JSONB(astext_type=sa.Text()),
+            nullable=True,
+        ),
+        schema=context.config.get_main_option("target_schema") or None,
+    )
+
+    # Add index for efficient worker polling (pending tasks ordered by creation time)
+    op.execute(
+        f"CREATE INDEX idx_async_operations_pending_claim ON {schema}async_operations (status, created_at) "
+        f"WHERE status = 'pending' AND task_payload IS NOT NULL"
+    )
+
+    # Add index for finding tasks by worker_id (for decommissioning)
+    op.execute(
+        f"CREATE INDEX idx_async_operations_worker_id ON {schema}async_operations (worker_id) WHERE worker_id IS NOT NULL"
+    )
+
+
+def downgrade() -> None:
+    """Remove worker columns from async_operations."""
+    schema = _get_schema_prefix()
+
+    # Drop indexes
+    op.execute(f"DROP INDEX IF EXISTS {schema}idx_async_operations_pending_claim")
+    op.execute(f"DROP INDEX IF EXISTS {schema}idx_async_operations_worker_id")
+
+    # Drop columns
+    op.drop_column(
+        "async_operations",
+        "task_payload",
+        schema=context.config.get_main_option("target_schema") or None,
+    )
+    op.drop_column(
+        "async_operations",
+        "retry_count",
+        schema=context.config.get_main_option("target_schema") or None,
+    )
+    op.drop_column(
+        "async_operations",
+        "claimed_at",
+        schema=context.config.get_main_option("target_schema") or None,
+    )
+    op.drop_column(
+        "async_operations",
+        "worker_id",
+        schema=context.config.get_main_option("target_schema") or None,
+    )

hindsight-api/hindsight_api/alembic/versions/m8h9i0j1k2l3_mental_model_id_to_text.py

@@ -0,0 +1,41 @@
+"""mental_model_id_to_text
+
+Revision ID: m8h9i0j1k2l3
+Revises: l7g8h9i0j1k2
+Create Date: 2026-01-19 00:00:00.000000
+
+This migration changes the mental_models.id column from VARCHAR(64) to TEXT
+to support longer model IDs (e.g., entity names that exceed 64 characters).
+"""
+
+from collections.abc import Sequence
+
+from alembic import context, op
+
+# revision identifiers, used by Alembic.
+revision: str = "m8h9i0j1k2l3"
+down_revision: str | Sequence[str] | None = "l7g8h9i0j1k2"
+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:
+    """Change mental_models.id from VARCHAR(64) to TEXT."""
+    schema = _get_schema_prefix()
+
+    # Alter the id column type from VARCHAR(64) to TEXT
+    op.execute(f"ALTER TABLE {schema}mental_models ALTER COLUMN id TYPE TEXT")
+
+
+def downgrade() -> None:
+    """Revert mental_models.id from TEXT to VARCHAR(64)."""
+    schema = _get_schema_prefix()
+
+    # Note: This may fail if any id values exceed 64 characters
+    op.execute(f"ALTER TABLE {schema}mental_models ALTER COLUMN id TYPE VARCHAR(64)")

hindsight-api/hindsight_api/alembic/versions/n9i0j1k2l3m4_learnings_and_pinned_reflections.py

@@ -0,0 +1,270 @@
+"""learnings_and_pinned_reflections
+
+Revision ID: n9i0j1k2l3m4
+Revises: m8h9i0j1k2l3
+Create Date: 2026-01-21 00:00:00.000000
+
+This migration:
+1. Creates the 'learnings' table for automatic bottom-up consolidation
+2. Creates the 'pinned_reflections' table for user-curated living documents
+3. Adds consolidation tracking columns to the 'banks' table
+"""
+
+import os
+from collections.abc import Sequence
+
+from alembic import context, op
+from sqlalchemy import text
+
+# revision identifiers, used by Alembic.
+revision: str = "n9i0j1k2l3m4"
+down_revision: str | Sequence[str] | None = "m8h9i0j1k2l3"
+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 _detect_vector_extension() -> str:
+    """
+    Detect or validate vector extension: 'vchord' or 'pgvector'.
+    Respects HINDSIGHT_API_VECTOR_EXTENSION env var if set.
+    """
+    conn = op.get_bind()
+    vector_extension = os.getenv("HINDSIGHT_API_VECTOR_EXTENSION", "pgvector").lower()
+
+    # Validate configured extension is installed
+    if vector_extension == "vchord":
+        vchord_check = conn.execute(text("SELECT 1 FROM pg_extension WHERE extname = 'vchord'")).scalar()
+        if not vchord_check:
+            raise RuntimeError(
+                "Configured vector extension 'vchord' not found. Install it with: CREATE EXTENSION vchord CASCADE;"
+            )
+        return "vchord"
+    elif vector_extension == "pgvector":
+        pgvector_check = conn.execute(text("SELECT 1 FROM pg_extension WHERE extname = 'vector'")).scalar()
+        if not pgvector_check:
+            raise RuntimeError(
+                "Configured vector extension 'pgvector' not found. Install it with: CREATE EXTENSION vector;"
+            )
+        return "pgvector"
+    else:
+        raise ValueError(f"Invalid HINDSIGHT_API_VECTOR_EXTENSION: {vector_extension}. Must be 'pgvector' or 'vchord'")
+
+
+def _detect_text_search_extension() -> str:
+    """
+    Detect or validate text search extension: 'native', 'vchord', or 'pg_textsearch'.
+    Respects HINDSIGHT_API_TEXT_SEARCH_EXTENSION env var.
+    Creates the extension if needed.
+    """
+    text_search_extension = os.getenv("HINDSIGHT_API_TEXT_SEARCH_EXTENSION", "native").lower()
+
+    if text_search_extension == "vchord":
+        # Create vchord_bm25 extension if not exists
+        try:
+            op.execute("CREATE EXTENSION IF NOT EXISTS vchord_bm25 CASCADE")
+        except Exception:
+            # Extension might already exist or user lacks permissions - verify it exists
+            conn = op.get_bind()
+            result = conn.execute(text("SELECT 1 FROM pg_extension WHERE extname = 'vchord_bm25'")).fetchone()
+            if not result:
+                # Extension truly doesn't exist - re-raise the error
+                raise
+        return "vchord"
+    elif text_search_extension == "pg_textsearch":
+        # Create pg_textsearch extension if not exists
+        try:
+            op.execute("CREATE EXTENSION IF NOT EXISTS pg_textsearch CASCADE")
+        except Exception:
+            # Extension might already exist or user lacks permissions - verify it exists
+            conn = op.get_bind()
+            result = conn.execute(text("SELECT 1 FROM pg_extension WHERE extname = 'pg_textsearch'")).fetchone()
+            if not result:
+                # Extension truly doesn't exist - re-raise the error
+                raise
+        return "pg_textsearch"
+    elif text_search_extension == "native":
+        return "native"
+    else:
+        raise ValueError(
+            f"Invalid HINDSIGHT_API_TEXT_SEARCH_EXTENSION: {text_search_extension}. Must be 'native', 'vchord', or 'pg_textsearch'"
+        )
+
+
+def upgrade() -> None:
+    """Create learnings and pinned_reflections tables."""
+    schema = _get_schema_prefix()
+
+    # Detect which vector extension is available
+    vector_ext = _detect_vector_extension()
+
+    # Detect which text search extension to use
+    text_search_ext = _detect_text_search_extension()
+
+    # 1. Create learnings table
+    op.execute(f"""
+        CREATE TABLE {schema}learnings (
+            id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
+            bank_id VARCHAR(64) NOT NULL,
+            text TEXT NOT NULL,
+            proof_count INT NOT NULL DEFAULT 1,
+            history JSONB DEFAULT '[]'::jsonb,
+            mission_context VARCHAR(64),
+            pre_mission_change BOOLEAN DEFAULT FALSE,
+            embedding vector(384),
+            tags VARCHAR[] DEFAULT ARRAY[]::VARCHAR[],
+            created_at TIMESTAMP WITH TIME ZONE NOT NULL DEFAULT now(),
+            updated_at TIMESTAMP WITH TIME ZONE NOT NULL DEFAULT now()
+        )
+    """)
+
+    # Add foreign key constraint
+    op.execute(f"""
+        ALTER TABLE {schema}learnings
+        ADD CONSTRAINT fk_learnings_bank_id
+        FOREIGN KEY (bank_id) REFERENCES {schema}banks(bank_id) ON DELETE CASCADE
+    """)
+
+    # Indexes for learnings
+    op.execute(f"CREATE INDEX idx_learnings_bank_id ON {schema}learnings(bank_id)")
+
+    # Create vector index based on detected extension
+    if vector_ext == "vchord":
+        op.execute(f"""
+            CREATE INDEX idx_learnings_embedding ON {schema}learnings
+            USING vchordrq (embedding vector_l2_ops)
+        """)
+    else:  # pgvector
+        op.execute(f"""
+            CREATE INDEX idx_learnings_embedding ON {schema}learnings
+            USING hnsw (embedding vector_cosine_ops)
+        """)
+
+    op.execute(f"CREATE INDEX idx_learnings_tags ON {schema}learnings USING GIN(tags)")
+
+    # Full-text search for learnings
+    if text_search_ext == "vchord":
+        # VectorChord BM25: bm25vector type (no GENERATED - tokenization happens on INSERT)
+        # Note: vchord_bm25 extension creates types in bm25_catalog schema
+        op.execute(f"""
+            ALTER TABLE {schema}learnings ADD COLUMN search_vector bm25_catalog.bm25vector
+        """)
+        op.execute(f"""
+            CREATE INDEX idx_learnings_text_search ON {schema}learnings
+            USING bm25 (search_vector bm25_catalog.bm25_ops)
+        """)
+    elif text_search_ext == "pg_textsearch":
+        # Timescale pg_textsearch: dummy TEXT column for consistency (indexes operate on base columns directly)
+        op.execute(f"""
+            ALTER TABLE {schema}learnings ADD COLUMN search_vector TEXT
+        """)
+        op.execute(f"""
+            CREATE INDEX idx_learnings_text_search ON {schema}learnings
+            USING bm25(text) WITH (text_config='english')
+        """)
+    else:  # native
+        # Native PostgreSQL: tsvector with automatic generation
+        op.execute(f"""
+            ALTER TABLE {schema}learnings ADD COLUMN search_vector tsvector
+            GENERATED ALWAYS AS (to_tsvector('english', text)) STORED
+        """)
+        op.execute(f"CREATE INDEX idx_learnings_text_search ON {schema}learnings USING gin(search_vector)")
+
+    # 2. Create pinned_reflections table
+    op.execute(f"""
+        CREATE TABLE {schema}pinned_reflections (
+            id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
+            bank_id VARCHAR(64) NOT NULL,
+            name VARCHAR(256) NOT NULL,
+            source_query TEXT NOT NULL,
+            content TEXT NOT NULL,
+            embedding vector(384),
+            tags VARCHAR[] DEFAULT ARRAY[]::VARCHAR[],
+            last_refreshed_at TIMESTAMP WITH TIME ZONE NOT NULL DEFAULT now(),
+            created_at TIMESTAMP WITH TIME ZONE NOT NULL DEFAULT now()
+        )
+    """)
+
+    # Add foreign key constraint
+    op.execute(f"""
+        ALTER TABLE {schema}pinned_reflections
+        ADD CONSTRAINT fk_pinned_reflections_bank_id
+        FOREIGN KEY (bank_id) REFERENCES {schema}banks(bank_id) ON DELETE CASCADE
+    """)
+
+    # Indexes for pinned_reflections
+    op.execute(f"CREATE INDEX idx_pinned_reflections_bank_id ON {schema}pinned_reflections(bank_id)")
+
+    # Create vector index based on detected extension
+    if vector_ext == "vchord":
+        op.execute(f"""
+            CREATE INDEX idx_pinned_reflections_embedding ON {schema}pinned_reflections
+            USING vchordrq (embedding vector_l2_ops)
+        """)
+    else:  # pgvector
+        op.execute(f"""
+            CREATE INDEX idx_pinned_reflections_embedding ON {schema}pinned_reflections
+            USING hnsw (embedding vector_cosine_ops)
+        """)
+
+    op.execute(f"CREATE INDEX idx_pinned_reflections_tags ON {schema}pinned_reflections USING GIN(tags)")
+
+    # Full-text search for pinned_reflections
+    if text_search_ext == "vchord":
+        # VectorChord BM25: bm25vector type (no GENERATED - tokenization happens on INSERT/UPDATE)
+        # Note: vchord_bm25 extension creates types in bm25_catalog schema
+        op.execute(f"""
+            ALTER TABLE {schema}pinned_reflections ADD COLUMN search_vector bm25_catalog.bm25vector
+        """)
+        op.execute(f"""
+            CREATE INDEX idx_pinned_reflections_text_search ON {schema}pinned_reflections
+            USING bm25 (search_vector bm25_catalog.bm25_ops)
+        """)
+    elif text_search_ext == "pg_textsearch":
+        # Timescale pg_textsearch: dummy TEXT column for consistency (indexes operate on base columns directly)
+        op.execute(f"""
+            ALTER TABLE {schema}pinned_reflections ADD COLUMN search_vector TEXT
+        """)
+        op.execute(f"""
+            CREATE INDEX idx_pinned_reflections_text_search ON {schema}pinned_reflections
+            USING bm25(content)
+            WITH (text_config='english')
+        """)
+    else:  # native
+        # Native PostgreSQL: tsvector with automatic generation
+        op.execute(f"""
+            ALTER TABLE {schema}pinned_reflections ADD COLUMN search_vector tsvector
+            GENERATED ALWAYS AS (to_tsvector('english', COALESCE(name, '') || ' ' || content)) STORED
+        """)
+        op.execute(f"""
+            CREATE INDEX idx_pinned_reflections_text_search ON {schema}pinned_reflections
+            USING gin(search_vector)
+        """)
+
+    # 3. Add consolidation tracking columns to banks table
+    op.execute(f"""
+        ALTER TABLE {schema}banks
+        ADD COLUMN IF NOT EXISTS last_consolidated_at TIMESTAMP WITH TIME ZONE
+    """)
+    op.execute(f"""
+        ALTER TABLE {schema}banks
+        ADD COLUMN IF NOT EXISTS mission_changed_at TIMESTAMP WITH TIME ZONE
+    """)
+
+
+def downgrade() -> None:
+    """Drop learnings and pinned_reflections tables."""
+    schema = _get_schema_prefix()
+
+    # Drop tables
+    op.execute(f"DROP TABLE IF EXISTS {schema}learnings CASCADE")
+    op.execute(f"DROP TABLE IF EXISTS {schema}pinned_reflections CASCADE")
+
+    # Remove columns from banks
+    op.execute(f"ALTER TABLE {schema}banks DROP COLUMN IF EXISTS last_consolidated_at")
+    op.execute(f"ALTER TABLE {schema}banks DROP COLUMN IF EXISTS mission_changed_at")

hindsight-api/hindsight_api/alembic/versions/o0j1k2l3m4n5_migrate_mental_models_data.py

@@ -0,0 +1,113 @@
+"""migrate_mental_models_data
+
+Revision ID: o0j1k2l3m4n5
+Revises: n9i0j1k2l3m4
+Create Date: 2026-01-21 00:00:00.000000
+
+This migration:
+1. Migrates existing 'pinned' mental models to the new 'pinned_reflections' table
+2. Migrates existing 'learned' mental models to the new 'learnings' table
+3. Deletes non-directive mental models (structural, emergent, pinned, learned)
+4. Drops the mental_model_versions table (no longer used)
+5. Adds a CHECK constraint that only 'directive' subtype is allowed
+"""
+
+from collections.abc import Sequence
+
+from alembic import context, op
+
+# revision identifiers, used by Alembic.
+revision: str = "o0j1k2l3m4n5"
+down_revision: str | Sequence[str] | None = "n9i0j1k2l3m4"
+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:
+    """Migrate data and clean up old mental models."""
+    schema = _get_schema_prefix()
+
+    # 1. Migrate 'pinned' mental models to pinned_reflections
+    # For pinned models, the first observation's content becomes the pinned reflection content
+    op.execute(f"""
+        INSERT INTO {schema}pinned_reflections (bank_id, name, source_query, content, tags, created_at)
+        SELECT
+            bank_id,
+            name,
+            description AS source_query,
+            COALESCE(
+                observations->'observations'->0->>'content',
+                description,
+                ''
+            ) AS content,
+            tags,
+            created_at
+        FROM {schema}mental_models
+        WHERE subtype = 'pinned'
+        ON CONFLICT DO NOTHING
+    """)
+
+    # 2. Migrate 'learned' mental models to learnings
+    # Each observation in a learned model becomes a separate learning
+    op.execute(f"""
+        INSERT INTO {schema}learnings (bank_id, text, proof_count, tags, created_at)
+        SELECT
+            mm.bank_id,
+            obs->>'content' AS text,
+            GREATEST(1, COALESCE(jsonb_array_length(obs->'evidence'), 1)) AS proof_count,
+            mm.tags,
+            mm.created_at
+        FROM {schema}mental_models mm,
+        LATERAL jsonb_array_elements(mm.observations->'observations') AS obs
+        WHERE mm.subtype = 'learned'
+          AND obs->>'content' IS NOT NULL
+          AND obs->>'content' != ''
+        ON CONFLICT DO NOTHING
+    """)
+
+    # 3. Delete all non-directive mental models (they've been migrated or are obsolete)
+    op.execute(f"""
+        DELETE FROM {schema}mental_models
+        WHERE subtype != 'directive'
+    """)
+
+    # 4. Drop the mental_model_versions table (no longer used)
+    op.execute(f"DROP TABLE IF EXISTS {schema}mental_model_versions CASCADE")
+
+    # 5. Drop old constraints and add new one that only allows 'directive'
+    op.execute(f"ALTER TABLE {schema}mental_models DROP CONSTRAINT IF EXISTS ck_mental_models_subtype")
+    op.execute(f"""
+        ALTER TABLE {schema}mental_models
+        ADD CONSTRAINT ck_mental_models_subtype CHECK (subtype = 'directive')
+    """)
+
+
+def downgrade() -> None:
+    """Reverse the migration (data migration is one-way, so this just removes constraints)."""
+    schema = _get_schema_prefix()
+
+    # Remove the directive-only constraint
+    op.execute(f"ALTER TABLE {schema}mental_models DROP CONSTRAINT IF EXISTS ck_mental_models_subtype")
+
+    # Re-create mental_model_versions table
+    op.execute(f"""
+        CREATE TABLE IF NOT EXISTS {schema}mental_model_versions (
+            id SERIAL PRIMARY KEY,
+            bank_id VARCHAR(64) NOT NULL,
+            model_id VARCHAR(128) NOT NULL,
+            version INT NOT NULL,
+            observations JSONB NOT NULL,
+            created_at TIMESTAMP WITH TIME ZONE NOT NULL DEFAULT now()
+        )
+    """)
+    op.execute(
+        f"CREATE INDEX IF NOT EXISTS idx_mm_versions_lookup ON {schema}mental_model_versions(bank_id, model_id, version DESC)"
+    )
+
+    # Note: Data migration cannot be reversed - pinned_reflections and learnings data remains

hindsight-api/hindsight_api/alembic/versions/p1k2l3m4n5o6_new_knowledge_architecture.py

@@ -0,0 +1,194 @@
+"""new_knowledge_architecture
+
+Revision ID: p1k2l3m4n5o6
+Revises: o0j1k2l3m4n5
+Create Date: 2026-01-21 00:00:00.000000
+
+This migration implements the new knowledge architecture:
+1. Drops the 'learnings' table (mental models are now in memory_units)
+2. Renames 'pinned_reflections' to 'reflections'
+3. Drops the 'mental_models' table completely
+4. Creates 'directives' table for hard rules
+5. Adds mental model support columns to 'memory_units' (proof_count, source_memory_ids, history)
+
+The new architecture:
+- Directives: Hard rules in their own table
+- Mental Models: Stored in memory_units with fact_type='mental_model'
+- Reflections: User-curated documents (renamed from pinned_reflections)
+"""
+
+from collections.abc import Sequence
+
+from alembic import context, op
+
+# revision identifiers, used by Alembic.
+revision: str = "p1k2l3m4n5o6"
+down_revision: str | Sequence[str] | None = "o0j1k2l3m4n5"
+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:
+    """Implement new knowledge architecture."""
+    schema = _get_schema_prefix()
+
+    # 1. Drop the learnings table (mental models will be in memory_units)
+    op.execute(f"DROP TABLE IF EXISTS {schema}learnings CASCADE")
+
+    # 2. Rename pinned_reflections to reflections
+    op.execute(f"ALTER TABLE IF EXISTS {schema}pinned_reflections RENAME TO reflections")
+
+    # Rename indexes for reflections
+    op.execute(f"ALTER INDEX IF EXISTS {schema}idx_pinned_reflections_bank_id RENAME TO idx_reflections_bank_id")
+    op.execute(f"ALTER INDEX IF EXISTS {schema}idx_pinned_reflections_embedding RENAME TO idx_reflections_embedding")
+    op.execute(f"ALTER INDEX IF EXISTS {schema}idx_pinned_reflections_tags RENAME TO idx_reflections_tags")
+    op.execute(
+        f"ALTER INDEX IF EXISTS {schema}idx_pinned_reflections_text_search RENAME TO idx_reflections_text_search"
+    )
+
+    # Rename foreign key constraint
+    op.execute(f"""
+        ALTER TABLE {schema}reflections
+        DROP CONSTRAINT IF EXISTS fk_pinned_reflections_bank_id
+    """)
+    op.execute(f"""
+        ALTER TABLE {schema}reflections
+        ADD CONSTRAINT fk_reflections_bank_id
+        FOREIGN KEY (bank_id) REFERENCES {schema}banks(bank_id) ON DELETE CASCADE
+    """)
+
+    # 3. Drop the mental_models table completely
+    op.execute(f"DROP TABLE IF EXISTS {schema}mental_models CASCADE")
+
+    # 4. Create directives table
+    op.execute(f"""
+        CREATE TABLE {schema}directives (
+            id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
+            bank_id VARCHAR(64) NOT NULL,
+            name VARCHAR(256) NOT NULL,
+            content TEXT NOT NULL,
+            priority INT NOT NULL DEFAULT 0,
+            is_active BOOLEAN NOT NULL DEFAULT TRUE,
+            tags VARCHAR[] DEFAULT ARRAY[]::VARCHAR[],
+            created_at TIMESTAMP WITH TIME ZONE NOT NULL DEFAULT now(),
+            updated_at TIMESTAMP WITH TIME ZONE NOT NULL DEFAULT now()
+        )
+    """)
+
+    # Add foreign key and indexes for directives
+    op.execute(f"""
+        ALTER TABLE {schema}directives
+        ADD CONSTRAINT fk_directives_bank_id
+        FOREIGN KEY (bank_id) REFERENCES {schema}banks(bank_id) ON DELETE CASCADE
+    """)
+    op.execute(f"CREATE INDEX idx_directives_bank_id ON {schema}directives(bank_id)")
+    op.execute(f"CREATE INDEX idx_directives_bank_active ON {schema}directives(bank_id, is_active)")
+    op.execute(f"CREATE INDEX idx_directives_tags ON {schema}directives USING GIN(tags)")
+
+    # 5. Add mental model support columns to memory_units
+    # proof_count: Number of memories that support this mental model
+    op.execute(f"""
+        ALTER TABLE {schema}memory_units
+        ADD COLUMN IF NOT EXISTS proof_count INT DEFAULT 1
+    """)
+
+    # source_memory_ids: Array of memory IDs that consolidated into this mental model
+    op.execute(f"""
+        ALTER TABLE {schema}memory_units
+        ADD COLUMN IF NOT EXISTS source_memory_ids UUID[] DEFAULT ARRAY[]::UUID[]
+    """)
+
+    # history: JSONB array tracking changes to mental models
+    op.execute(f"""
+        ALTER TABLE {schema}memory_units
+        ADD COLUMN IF NOT EXISTS history JSONB DEFAULT '[]'::jsonb
+    """)
+
+    # Add index for finding mental models
+    op.execute(f"""
+        CREATE INDEX IF NOT EXISTS idx_memory_units_mental_models
+        ON {schema}memory_units(bank_id, fact_type)
+        WHERE fact_type = 'mental_model'
+    """)
+
+    # 6. Update fact_type check constraint to include 'mental_model'
+    op.execute(f"ALTER TABLE {schema}memory_units DROP CONSTRAINT IF EXISTS memory_units_fact_type_check")
+    op.execute(f"""
+        ALTER TABLE {schema}memory_units
+        ADD CONSTRAINT memory_units_fact_type_check
+        CHECK (fact_type IN ('world', 'experience', 'opinion', 'observation', 'mental_model'))
+    """)
+
+
+def downgrade() -> None:
+    """Reverse the migration."""
+    schema = _get_schema_prefix()
+
+    # Restore original fact_type check constraint (without 'mental_model')
+    op.execute(f"ALTER TABLE {schema}memory_units DROP CONSTRAINT IF EXISTS memory_units_fact_type_check")
+    op.execute(f"""
+        ALTER TABLE {schema}memory_units
+        ADD CONSTRAINT memory_units_fact_type_check
+        CHECK (fact_type IN ('world', 'experience', 'opinion', 'observation'))
+    """)
+
+    # Drop mental model columns from memory_units
+    op.execute(f"ALTER TABLE {schema}memory_units DROP COLUMN IF EXISTS proof_count")
+    op.execute(f"ALTER TABLE {schema}memory_units DROP COLUMN IF EXISTS source_memory_ids")
+    op.execute(f"ALTER TABLE {schema}memory_units DROP COLUMN IF EXISTS history")
+    op.execute(f"DROP INDEX IF EXISTS {schema}idx_memory_units_mental_models")
+
+    # Drop directives table
+    op.execute(f"DROP TABLE IF EXISTS {schema}directives CASCADE")
+
+    # Rename reflections back to pinned_reflections
+    op.execute(f"ALTER TABLE IF EXISTS {schema}reflections RENAME TO pinned_reflections")
+
+    # Restore indexes
+    op.execute(f"ALTER INDEX IF EXISTS {schema}idx_reflections_bank_id RENAME TO idx_pinned_reflections_bank_id")
+    op.execute(f"ALTER INDEX IF EXISTS {schema}idx_reflections_embedding RENAME TO idx_pinned_reflections_embedding")
+    op.execute(f"ALTER INDEX IF EXISTS {schema}idx_reflections_tags RENAME TO idx_pinned_reflections_tags")
+    op.execute(
+        f"ALTER INDEX IF EXISTS {schema}idx_reflections_text_search RENAME TO idx_pinned_reflections_text_search"
+    )
+
+    # Restore foreign key
+    op.execute(f"""
+        ALTER TABLE {schema}pinned_reflections
+        DROP CONSTRAINT IF EXISTS fk_reflections_bank_id
+    """)
+    op.execute(f"""
+        ALTER TABLE {schema}pinned_reflections
+        ADD CONSTRAINT fk_pinned_reflections_bank_id
+        FOREIGN KEY (bank_id) REFERENCES {schema}banks(bank_id) ON DELETE CASCADE
+    """)
+
+    # Re-create learnings table
+    op.execute(f"""
+        CREATE TABLE IF NOT EXISTS {schema}learnings (
+            id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
+            bank_id VARCHAR(64) NOT NULL,
+            text TEXT NOT NULL,
+            proof_count INT NOT NULL DEFAULT 1,
+            history JSONB DEFAULT '[]'::jsonb,
+            mission_context VARCHAR(64),
+            pre_mission_change BOOLEAN DEFAULT FALSE,
+            embedding vector(384),
+            tags VARCHAR[] DEFAULT ARRAY[]::VARCHAR[],
+            created_at TIMESTAMP WITH TIME ZONE NOT NULL DEFAULT now(),
+            updated_at TIMESTAMP WITH TIME ZONE NOT NULL DEFAULT now()
+        )
+    """)
+    op.execute(f"""
+        ALTER TABLE {schema}learnings
+        ADD CONSTRAINT fk_learnings_bank_id
+        FOREIGN KEY (bank_id) REFERENCES {schema}banks(bank_id) ON DELETE CASCADE
+    """)
+
+    # Note: mental_models table recreation is complex and would need separate handling

hindsight-api/hindsight_api/alembic/versions/q2l3m4n5o6p7_fix_mental_model_fact_type.py

@@ -0,0 +1,50 @@
+"""fix_mental_model_fact_type
+
+Revision ID: q2l3m4n5o6p7
+Revises: p1k2l3m4n5o6
+Create Date: 2026-01-21 13:30:00.000000
+
+Fix the fact_type check constraint to include 'mental_model'.
+This is a fix for p1k2l3m4n5o6 which should have included this change.
+"""
+
+from collections.abc import Sequence
+
+from alembic import context, op
+
+# revision identifiers, used by Alembic.
+revision: str = "q2l3m4n5o6p7"
+down_revision: str | Sequence[str] | None = "p1k2l3m4n5o6"
+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 'mental_model' to the fact_type check constraint."""
+    schema = _get_schema_prefix()
+
+    # Drop the old constraint and add the new one with mental_model included
+    op.execute(f"ALTER TABLE {schema}memory_units DROP CONSTRAINT IF EXISTS memory_units_fact_type_check")
+    op.execute(f"""
+        ALTER TABLE {schema}memory_units
+        ADD CONSTRAINT memory_units_fact_type_check
+        CHECK (fact_type IN ('world', 'experience', 'opinion', 'observation', 'mental_model'))
+    """)
+
+
+def downgrade() -> None:
+    """Remove 'mental_model' from the fact_type check constraint."""
+    schema = _get_schema_prefix()
+
+    op.execute(f"ALTER TABLE {schema}memory_units DROP CONSTRAINT IF EXISTS memory_units_fact_type_check")
+    op.execute(f"""
+        ALTER TABLE {schema}memory_units
+        ADD CONSTRAINT memory_units_fact_type_check
+        CHECK (fact_type IN ('world', 'experience', 'opinion', 'observation'))
+    """)

hindsight-api/hindsight_api/alembic/versions/r3m4n5o6p7q8_add_reflect_response_to_reflections.py

@@ -0,0 +1,47 @@
+"""Add reflect_response JSONB column to reflections
+
+Revision ID: r3m4n5o6p7q8
+Revises: q2l3m4n5o6p7
+Create Date: 2026-01-21
+
+This migration adds a reflect_response JSONB column to store the full
+reflect API response payload, including based_on facts and trace data.
+
+Note: Table was renamed from pinned_reflections to reflections in p1k2l3m4n5o6.
+"""
+
+from collections.abc import Sequence
+
+from alembic import context, op
+
+revision: str = "r3m4n5o6p7q8"
+down_revision: str | Sequence[str] | None = "q2l3m4n5o6p7"
+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 reflect_response JSONB column to reflections."""
+    schema = _get_schema_prefix()
+
+    # Add reflect_response column to store the full reflect API response
+    op.execute(f"""
+        ALTER TABLE {schema}reflections
+        ADD COLUMN IF NOT EXISTS reflect_response JSONB
+    """)
+
+
+def downgrade() -> None:
+    """Remove reflect_response column from reflections."""
+    schema = _get_schema_prefix()
+
+    op.execute(f"""
+        ALTER TABLE {schema}reflections
+        DROP COLUMN IF EXISTS reflect_response
+    """)

hindsight-api/hindsight_api/alembic/versions/rename_personality_to_disposition.py

@@ -0,0 +1,85 @@
+"""rename_personality_to_disposition
+
+Revision ID: rename_personality
+Revises: d9f6a3b4c5e2
+Create Date: 2024-12-04
+
+"""
+
+from collections.abc import Sequence
+
+import sqlalchemy as sa
+from alembic import context, op
+from sqlalchemy.dialects import postgresql
+
+# revision identifiers, used by Alembic.
+revision: str = "rename_personality"
+down_revision: str | Sequence[str] | None = "d9f6a3b4c5e2"
+branch_labels: str | Sequence[str] | None = None
+depends_on: str | Sequence[str] | None = None
+
+
+def _get_target_schema() -> str:
+    """Get the target schema name (tenant schema or 'public')."""
+    schema = context.config.get_main_option("target_schema")
+    return schema if schema else "public"
+
+
+def upgrade() -> None:
+    """Rename personality column to disposition in banks table (if it exists)."""
+    conn = op.get_bind()
+    target_schema = _get_target_schema()
+
+    # Check if 'personality' column exists (old database)
+    result = conn.execute(
+        sa.text("""
+        SELECT column_name
+        FROM information_schema.columns
+        WHERE table_schema = :schema AND table_name = 'banks' AND column_name = 'personality'
+    """),
+        {"schema": target_schema},
+    )
+    has_personality = result.fetchone() is not None
+
+    # Check if 'disposition' column exists (new database)
+    result = conn.execute(
+        sa.text("""
+        SELECT column_name
+        FROM information_schema.columns
+        WHERE table_schema = :schema AND table_name = 'banks' AND column_name = 'disposition'
+    """),
+        {"schema": target_schema},
+    )
+    has_disposition = result.fetchone() is not None
+
+    if has_personality and not has_disposition:
+        # Old database: rename personality -> disposition
+        op.alter_column("banks", "personality", new_column_name="disposition")
+    elif not has_personality and not has_disposition:
+        # Neither exists (shouldn't happen, but be safe): add disposition column
+        op.add_column(
+            "banks",
+            sa.Column(
+                "disposition",
+                postgresql.JSONB(astext_type=sa.Text()),
+                server_default=sa.text("'{}'::jsonb"),
+                nullable=False,
+            ),
+        )
+    # else: disposition already exists, nothing to do
+
+
+def downgrade() -> None:
+    """Revert disposition column back to personality."""
+    conn = op.get_bind()
+    target_schema = _get_target_schema()
+    result = conn.execute(
+        sa.text("""
+        SELECT column_name
+        FROM information_schema.columns
+        WHERE table_schema = :schema AND table_name = 'banks' AND column_name = 'disposition'
+    """),
+        {"schema": target_schema},
+    )
+    if result.fetchone():
+        op.alter_column("banks", "disposition", new_column_name="personality")

hindsight-api/hindsight_api/alembic/versions/s4n5o6p7q8r9_add_consolidated_at_to_memory_units.py

@@ -0,0 +1,53 @@
+"""Add consolidated_at column to memory_units for incremental consolidation tracking.
+
+This allows consolidation to track progress at the memory level rather than
+using a bank-level watermark. If consolidation crashes, already-processed
+memories won't be reprocessed.
+
+Revision ID: s4n5o6p7q8r9
+Revises: r3m4n5o6p7q8
+Create Date: 2025-01-22
+"""
+
+from collections.abc import Sequence
+
+from alembic import context, op
+
+revision: str = "s4n5o6p7q8r9"
+down_revision: str | Sequence[str] | None = "r3m4n5o6p7q8"
+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()
+
+    # Add consolidated_at column to memory_units
+    op.execute(
+        f"""
+        ALTER TABLE {schema}memory_units
+        ADD COLUMN IF NOT EXISTS consolidated_at TIMESTAMPTZ DEFAULT NULL
+        """
+    )
+
+    # Create index for efficient querying of unconsolidated memories
+    op.execute(
+        f"""
+        CREATE INDEX IF NOT EXISTS idx_memory_units_unconsolidated
+        ON {schema}memory_units (bank_id, created_at)
+        WHERE consolidated_at IS 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_unconsolidated")
+    op.execute(f"ALTER TABLE {schema}memory_units DROP COLUMN IF EXISTS consolidated_at")

hindsight-api/hindsight_api/alembic/versions/t5o6p7q8r9s0_rename_mental_models_to_observations.py

@@ -0,0 +1,134 @@
+"""Rename mental_model fact_type to observation and reflections table to mental_models
+
+Revision ID: t5o6p7q8r9s0
+Revises: s4n5o6p7q8r9
+Create Date: 2026-01-26
+
+This migration implements the terminology rename:
+1. mental_model (fact_type in memory_units) -> observation
+2. reflections table -> mental_models table
+
+The new terminology:
+- Observations: Consolidated knowledge synthesized from facts (was mental_model)
+- Mental Models: Stored reflect responses (was reflections)
+"""
+
+from collections.abc import Sequence
+
+from alembic import context, op
+
+revision: str = "t5o6p7q8r9s0"
+down_revision: str | Sequence[str] | None = "s4n5o6p7q8r9"
+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:
+    """Rename mental_model -> observation and reflections -> mental_models."""
+    schema = _get_schema_prefix()
+
+    # 1. Update fact_type values: mental_model -> observation
+    op.execute(f"""
+        UPDATE {schema}memory_units
+        SET fact_type = 'observation'
+        WHERE fact_type = 'mental_model'
+    """)
+
+    # 2. Update the CHECK constraint - remove mental_model, keep observation
+    op.execute(f"ALTER TABLE {schema}memory_units DROP CONSTRAINT IF EXISTS memory_units_fact_type_check")
+    op.execute(f"""
+        ALTER TABLE {schema}memory_units
+        ADD CONSTRAINT memory_units_fact_type_check
+        CHECK (fact_type IN ('world', 'experience', 'opinion', 'observation'))
+    """)
+
+    # 3. Rename the index for observations (was for mental_models)
+    op.execute(f"DROP INDEX IF EXISTS {schema}idx_memory_units_mental_models")
+    op.execute(f"""
+        CREATE INDEX IF NOT EXISTS idx_memory_units_observations
+        ON {schema}memory_units(bank_id, fact_type)
+        WHERE fact_type = 'observation'
+    """)
+
+    # 4. Update the unconsolidated index to not filter by fact_type since observations
+    # are now the consolidated type
+    op.execute(f"DROP INDEX IF EXISTS {schema}idx_memory_units_unconsolidated")
+    op.execute(f"""
+        CREATE INDEX IF NOT EXISTS idx_memory_units_unconsolidated
+        ON {schema}memory_units (bank_id, created_at)
+        WHERE consolidated_at IS NULL AND fact_type IN ('experience', 'world')
+    """)
+
+    # 5. Rename reflections table to mental_models
+    op.execute(f"ALTER TABLE IF EXISTS {schema}reflections RENAME TO mental_models")
+
+    # 6. Rename indexes for mental_models (was reflections)
+    op.execute(f"ALTER INDEX IF EXISTS {schema}idx_reflections_bank_id RENAME TO idx_mental_models_bank_id")
+    op.execute(f"ALTER INDEX IF EXISTS {schema}idx_reflections_embedding RENAME TO idx_mental_models_embedding")
+    op.execute(f"ALTER INDEX IF EXISTS {schema}idx_reflections_tags RENAME TO idx_mental_models_tags")
+    op.execute(f"ALTER INDEX IF EXISTS {schema}idx_reflections_text_search RENAME TO idx_mental_models_text_search")
+
+    # 7. Rename foreign key constraint
+    op.execute(f"""
+        ALTER TABLE {schema}mental_models
+        DROP CONSTRAINT IF EXISTS fk_reflections_bank_id
+    """)
+    op.execute(f"""
+        ALTER TABLE {schema}mental_models
+        ADD CONSTRAINT fk_mental_models_bank_id
+        FOREIGN KEY (bank_id) REFERENCES {schema}banks(bank_id) ON DELETE CASCADE
+    """)
+
+
+def downgrade() -> None:
+    """Reverse: observation -> mental_model and mental_models -> reflections."""
+    schema = _get_schema_prefix()
+
+    # 1. Rename mental_models table back to reflections
+    op.execute(f"ALTER TABLE IF EXISTS {schema}mental_models RENAME TO reflections")
+
+    # 2. Rename indexes back
+    op.execute(f"ALTER INDEX IF EXISTS {schema}idx_mental_models_bank_id RENAME TO idx_reflections_bank_id")
+    op.execute(f"ALTER INDEX IF EXISTS {schema}idx_mental_models_embedding RENAME TO idx_reflections_embedding")
+    op.execute(f"ALTER INDEX IF EXISTS {schema}idx_mental_models_tags RENAME TO idx_reflections_tags")
+    op.execute(f"ALTER INDEX IF EXISTS {schema}idx_mental_models_text_search RENAME TO idx_reflections_text_search")
+
+    # 3. Rename foreign key back
+    op.execute(f"""
+        ALTER TABLE {schema}reflections
+        DROP CONSTRAINT IF EXISTS fk_mental_models_bank_id
+    """)
+    op.execute(f"""
+        ALTER TABLE {schema}reflections
+        ADD CONSTRAINT fk_reflections_bank_id
+        FOREIGN KEY (bank_id) REFERENCES {schema}banks(bank_id) ON DELETE CASCADE
+    """)
+
+    # 4. Update fact_type values: observation -> mental_model
+    op.execute(f"""
+        UPDATE {schema}memory_units
+        SET fact_type = 'mental_model'
+        WHERE fact_type = 'observation'
+    """)
+
+    # 5. Update the CHECK constraint back
+    op.execute(f"ALTER TABLE {schema}memory_units DROP CONSTRAINT IF EXISTS memory_units_fact_type_check")
+    op.execute(f"""
+        ALTER TABLE {schema}memory_units
+        ADD CONSTRAINT memory_units_fact_type_check
+        CHECK (fact_type IN ('world', 'experience', 'opinion', 'observation', 'mental_model'))
+    """)
+
+    # 6. Rename index back
+    op.execute(f"DROP INDEX IF EXISTS {schema}idx_memory_units_observations")
+    op.execute(f"""
+        CREATE INDEX IF NOT EXISTS idx_memory_units_mental_models
+        ON {schema}memory_units(bank_id, fact_type)
+        WHERE fact_type = 'mental_model'
+    """)

hindsight-api/hindsight_api/alembic/versions/u6p7q8r9s0t1_mental_models_text_id.py

@@ -0,0 +1,41 @@
+"""Change mental_models.id from UUID to TEXT
+
+Revision ID: u6p7q8r9s0t1
+Revises: t5o6p7q8r9s0
+Create Date: 2026-01-27
+
+This migration changes the mental_models.id column from UUID to TEXT
+to support user-defined text identifiers like 'team-communication' instead of UUIDs.
+"""
+
+from collections.abc import Sequence
+
+from alembic import context, op
+
+revision: str = "u6p7q8r9s0t1"
+down_revision: str | Sequence[str] | None = "t5o6p7q8r9s0"
+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:
+    """Change mental_models.id from UUID to TEXT."""
+    schema = _get_schema_prefix()
+
+    # Change the id column type from UUID to TEXT
+    # Existing UUIDs will be converted to their string representation
+    op.execute(f"ALTER TABLE {schema}mental_models ALTER COLUMN id TYPE TEXT USING id::TEXT")
+
+
+def downgrade() -> None:
+    """Revert mental_models.id from TEXT to UUID."""
+    schema = _get_schema_prefix()
+
+    # Note: This will fail if any id values are not valid UUIDs
+    op.execute(f"ALTER TABLE {schema}mental_models ALTER COLUMN id TYPE UUID USING id::UUID")

hindsight-api/hindsight_api/alembic/versions/v7q8r9s0t1u2_add_max_tokens_to_mental_models.py

@@ -0,0 +1,50 @@
+"""Add max_tokens and trigger columns to mental_models
+
+Revision ID: v7q8r9s0t1u2
+Revises: u6p7q8r9s0t1
+Create Date: 2026-01-27
+
+This migration adds:
+- max_tokens column: token limit for content generation during refresh
+- trigger column: JSONB for trigger settings (e.g., refresh_after_consolidation)
+"""
+
+from collections.abc import Sequence
+
+from alembic import context, op
+
+revision: str = "v7q8r9s0t1u2"
+down_revision: str | Sequence[str] | None = "u6p7q8r9s0t1"
+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 max_tokens and trigger columns to mental_models."""
+    schema = _get_schema_prefix()
+
+    op.execute(f"""
+        ALTER TABLE {schema}mental_models
+        ADD COLUMN IF NOT EXISTS max_tokens INT NOT NULL DEFAULT 2048
+    """)
+
+    # trigger column stores trigger settings as JSONB
+    # Default: refresh_after_consolidation = false (not "real time")
+    op.execute(f"""
+        ALTER TABLE {schema}mental_models
+        ADD COLUMN IF NOT EXISTS trigger JSONB NOT NULL DEFAULT '{{"refresh_after_consolidation": false}}'::jsonb
+    """)
+
+
+def downgrade() -> None:
+    """Remove max_tokens and trigger columns from mental_models."""
+    schema = _get_schema_prefix()
+
+    op.execute(f"ALTER TABLE {schema}mental_models DROP COLUMN IF EXISTS max_tokens")
+    op.execute(f"ALTER TABLE {schema}mental_models DROP COLUMN IF EXISTS trigger")

hindsight-api/hindsight_api/alembic/versions/w8r9s0t1u2v3_fix_mental_models_pk_isolation.py

@@ -0,0 +1,60 @@
+"""Fix mental_models primary key to be scoped per bank
+
+Revision ID: w8r9s0t1u2v3
+Revises: v7q8r9s0t1u2
+Create Date: 2026-02-05
+
+This migration fixes a critical bank isolation bug where mental_models.id was
+globally unique across all banks instead of being scoped per bank. This caused
+conflicts when different banks tried to use the same custom ID.
+
+CRITICAL FIX: Changes primary key from (id) to (bank_id, id) to ensure proper isolation.
+"""
+
+from collections.abc import Sequence
+
+from alembic import context, op
+
+revision: str = "w8r9s0t1u2v3"
+down_revision: str | Sequence[str] | None = "v7q8r9s0t1u2"
+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:
+    """Change mental_models primary key from (id) to (bank_id, id) for proper bank isolation."""
+    schema = _get_schema_prefix()
+
+    # Drop the old primary key constraint (just id)
+    # Note: The constraint might be named differently on different DBs
+    # Try both old names (pinned_reflections_pkey from original, mental_models_pkey from rename)
+    op.execute(f"ALTER TABLE {schema}mental_models DROP CONSTRAINT IF EXISTS pinned_reflections_pkey")
+    op.execute(f"ALTER TABLE {schema}mental_models DROP CONSTRAINT IF EXISTS mental_models_pkey")
+
+    # Create the new composite primary key (bank_id, id)
+    # This ensures IDs are scoped per bank, not globally
+    op.execute(f"""
+        ALTER TABLE {schema}mental_models
+        ADD CONSTRAINT mental_models_pkey PRIMARY KEY (bank_id, id)
+    """)
+
+
+def downgrade() -> None:
+    """Revert mental_models primary key from (bank_id, id) to (id)."""
+    schema = _get_schema_prefix()
+
+    # Drop the composite primary key
+    op.execute(f"ALTER TABLE {schema}mental_models DROP CONSTRAINT IF EXISTS mental_models_pkey")
+
+    # Restore the old primary key (just id)
+    # WARNING: This downgrade will fail if there are duplicate IDs across banks
+    op.execute(f"""
+        ALTER TABLE {schema}mental_models
+        ADD CONSTRAINT mental_models_pkey PRIMARY KEY (id)
+    """)

hindsight-api/hindsight_api/alembic/versions/x9s0t1u2v3w4_add_bank_config_column.py

@@ -0,0 +1,64 @@
+"""Add config JSONB column to banks table for hierarchical configuration
+
+Revision ID: x9s0t1u2v3w4
+Revises: w8r9s0t1u2v3
+Create Date: 2026-02-09
+
+This migration adds a `config` JSONB column to the banks table to support
+per-bank configuration overrides. This enables hierarchical configuration where:
+- Global config is loaded from environment variables
+- Tenant config is provided via TenantExtension
+- Bank config overrides are stored in banks.config JSONB column
+
+The config column stores overrides for hierarchical fields (LLM settings,
+retention parameters, retrieval settings, etc.) in Python field name format.
+"""
+
+from collections.abc import Sequence
+
+import sqlalchemy as sa
+from alembic import context, op
+from sqlalchemy.dialects.postgresql import JSONB
+
+revision: str = "x9s0t1u2v3w4"
+down_revision: str | Sequence[str] | None = "w8r9s0t1u2v3"
+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 config JSONB column to banks table with GIN index."""
+    schema = _get_schema_prefix()
+
+    # Add config column to banks table
+    op.execute(f"""
+        ALTER TABLE {schema}banks
+        ADD COLUMN config JSONB NOT NULL DEFAULT '{{}}'::jsonb
+    """)
+
+    # Add GIN index for efficient JSONB queries
+    op.execute(f"""
+        CREATE INDEX idx_banks_config
+        ON {schema}banks
+        USING gin(config)
+    """)
+
+
+def downgrade() -> None:
+    """Remove config column and index from banks table."""
+    schema = _get_schema_prefix()
+
+    # Drop index first
+    op.execute(f"DROP INDEX IF EXISTS {schema}idx_banks_config")
+
+    # Drop column
+    op.execute(f"""
+        ALTER TABLE {schema}banks
+        DROP COLUMN IF EXISTS config
+    """)

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

@@ -0,0 +1,49 @@
+"""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 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/hindsight_api/api/__init__.py

@@ -3,8 +3,10 @@ Unified API module for Hindsight.
 
 Provides both HTTP REST API and MCP (Model Context Protocol) server.
 """
+
 import logging
-from typing import Optional
+from contextlib import asynccontextmanager
+
 from fastapi import FastAPI
 
 from hindsight_api import MemoryEngine
@@ -17,18 +19,17 @@ def create_app(
     http_api_enabled: bool = True,
     mcp_api_enabled: bool = False,
     mcp_mount_path: str = "/mcp",
-    run_migrations: bool = True,
-    initialize_memory: bool = True
+    initialize_memory: bool = True,
 ) -> FastAPI:
     """
     Create and configure the unified Hindsight API application.
 
     Args:
-        memory: MemoryEngine instance (already initialized with required parameters)
+        memory: MemoryEngine instance (already initialized with required parameters).
+                Migrations are controlled by the MemoryEngine's run_migrations parameter.
         http_api_enabled: Whether to enable HTTP REST API endpoints (default: True)
         mcp_api_enabled: Whether to enable MCP server (default: False)
         mcp_mount_path: Path to mount MCP server (default: /mcp)
-        run_migrations: Whether to run database migrations on startup (default: True)
         initialize_memory: Whether to initialize memory system on startup (default: True)
 
     Returns:
@@ -44,50 +45,81 @@ def create_app(
         # Both HTTP and MCP
         app = create_app(memory, mcp_api_enabled=True)
     """
+    mcp_servers = None
+
+    # Create MCP servers first if enabled (we need their lifespans for chaining)
+    if mcp_api_enabled:
+        try:
+            from .mcp import MCPMiddleware, create_mcp_servers
+
+            mcp_servers = create_mcp_servers(memory=memory)
+        except ImportError as e:
+            logger.error(f"MCP server requested but dependencies not available: {e}")
+            logger.error("Install with: pip install hindsight-api[mcp]")
+            raise
 
     # Import and create HTTP API if enabled
     if http_api_enabled:
         from .http import create_app as create_http_app
-        app = create_http_app(
-            memory=memory,
-            run_migrations=run_migrations,
-            initialize_memory=initialize_memory
-        )
+
+        app = create_http_app(memory=memory, initialize_memory=initialize_memory)
         logger.info("HTTP REST API enabled")
     else:
         # Create minimal FastAPI app
         app = FastAPI(title="Hindsight API", version="0.0.7")
         logger.info("HTTP REST API disabled")
 
-    # Mount MCP server if enabled
-    if mcp_api_enabled:
-        try:
-            from .mcp import create_mcp_app
+    # Add MCP middleware and chain its lifespan if enabled
+    if mcp_servers is not None:
+        multi_bank_server, single_bank_server, multi_bank_starlette_app, single_bank_starlette_app = mcp_servers
 
-            # Create MCP app with dynamic bank_id support
-            # Supports: /mcp/{bank_id}/sse (bank-specific SSE endpoint)
-            mcp_app = create_mcp_app(memory=memory)
-            app.mount(mcp_mount_path, mcp_app)
-            logger.info(f"MCP server enabled at {mcp_mount_path}/{{bank_id}}/sse")
-        except ImportError as e:
-            logger.error(f"MCP server requested but dependencies not available: {e}")
-            logger.error("Install with: pip install hindsight-api[mcp]")
-            raise
+        # Store the original lifespan
+        original_lifespan = app.router.lifespan_context
+
+        @asynccontextmanager
+        async def chained_lifespan(app_instance: FastAPI):
+            """Chain both MCP lifespans with the main app lifespan."""
+            # Start both MCP lifespans (multi-bank and single-bank)
+            async with multi_bank_starlette_app.router.lifespan_context(multi_bank_starlette_app):
+                async with single_bank_starlette_app.router.lifespan_context(single_bank_starlette_app):
+                    logger.info("MCP lifespans started (multi-bank and single-bank)")
+                    # Then start the original app lifespan
+                    async with original_lifespan(app_instance):
+                        yield
+                logger.info("MCP lifespans stopped")
+
+        # Replace the app's lifespan with the chained version
+        app.router.lifespan_context = chained_lifespan
+
+        # Add MCP as a wrapping middleware — intercepts /mcp* requests directly,
+        # passes everything else through to the FastAPI app. No Starlette Mount
+        # means no 307 redirect for /mcp (no trailing slash).
+        app.add_middleware(
+            MCPMiddleware,
+            memory=memory,
+            prefix=mcp_mount_path,
+            multi_bank_app=multi_bank_starlette_app,
+            single_bank_app=single_bank_starlette_app,
+            multi_bank_server=multi_bank_server,
+            single_bank_server=single_bank_server,
+        )
+
+        logger.info(f"MCP server enabled at {mcp_mount_path}/")
 
     return app
 
 
 # Re-export commonly used items for backwards compatibility
 from .http import (
-    RecallRequest,
-    RecallResult,
-    RecallResponse,
-    MemoryItem,
-    RetainRequest,
-    ReflectRequest,
-    ReflectResponse,
     CreateBankRequest,
     DispositionTraits,
+    MemoryItem,
+    RecallRequest,
+    RecallResponse,
+    RecallResult,
+    ReflectRequest,
+    ReflectResponse,
+    RetainRequest,
 )
 
 __all__ = [

hindsight-api/hindsight_api/api/mcp.py

@@ -1,215 +1,387 @@
-"""Hindsight MCP Server implementation using FastMCP."""
+"""Hindsight MCP Server implementation using FastMCP (HTTP transport)."""
 
 import json
 import logging
 import os
 from contextvars import ContextVar
-from typing import Optional
 
 from fastmcp import FastMCP
+
 from hindsight_api import MemoryEngine
-from hindsight_api.engine.response_models import VALID_RECALL_FACT_TYPES
+from hindsight_api.engine.memory_engine import _current_schema
+from hindsight_api.extensions import MCPExtension, load_extension
+from hindsight_api.extensions.tenant import AuthenticationError
+from hindsight_api.mcp_tools import MCPToolsConfig, register_mcp_tools
+from hindsight_api.models import RequestContext
 
 # Configure logging from HINDSIGHT_API_LOG_LEVEL environment variable
 _log_level_str = os.environ.get("HINDSIGHT_API_LOG_LEVEL", "info").lower()
-_log_level_map = {"critical": logging.CRITICAL, "error": logging.ERROR, "warning": logging.WARNING,
-                  "info": logging.INFO, "debug": logging.DEBUG, "trace": logging.DEBUG}
+_log_level_map = {
+    "critical": logging.CRITICAL,
+    "error": logging.ERROR,
+    "warning": logging.WARNING,
+    "info": logging.INFO,
+    "debug": logging.DEBUG,
+    "trace": logging.DEBUG,
+}
 logging.basicConfig(
     level=_log_level_map.get(_log_level_str, logging.INFO),
-    format="%(asctime)s - %(levelname)s - %(name)s - %(message)s"
+    format="%(asctime)s - %(levelname)s - %(name)s - %(message)s",
 )
 logger = logging.getLogger(__name__)
 
-# Context variable to hold the current bank_id from the URL path
-_current_bank_id: ContextVar[Optional[str]] = ContextVar("current_bank_id", default=None)
+# Default bank_id from environment variable
+DEFAULT_BANK_ID = os.environ.get("HINDSIGHT_MCP_BANK_ID", "default")
+
+# Legacy MCP authentication token (for backwards compatibility)
+# If set, this token is checked first before TenantExtension auth
+MCP_AUTH_TOKEN = os.environ.get("HINDSIGHT_API_MCP_AUTH_TOKEN")
+
+# Context variable to hold the current bank_id
+_current_bank_id: ContextVar[str | None] = ContextVar("current_bank_id", default=None)
+
+# Context variable to hold the current API key (for tenant auth propagation)
+_current_api_key: ContextVar[str | None] = ContextVar("current_api_key", default=None)
+
+# Context variables for tenant_id and api_key_id (set by authenticate, used by usage metering)
+_current_tenant_id: ContextVar[str | None] = ContextVar("current_tenant_id", default=None)
+_current_api_key_id: ContextVar[str | None] = ContextVar("current_api_key_id", default=None)
 
 
-def get_current_bank_id() -> Optional[str]:
-    """Get the current bank_id from context (set from URL path)."""
+def get_current_bank_id() -> str | None:
+    """Get the current bank_id from context."""
     return _current_bank_id.get()
 
 
-def create_mcp_server(memory: MemoryEngine) -> FastMCP:
+def get_current_api_key() -> str | None:
+    """Get the current API key from context."""
+    return _current_api_key.get()
+
+
+def get_current_tenant_id() -> str | None:
+    """Get the current tenant_id from context."""
+    return _current_tenant_id.get()
+
+
+def get_current_api_key_id() -> str | None:
+    """Get the current api_key_id from context."""
+    return _current_api_key_id.get()
+
+
+def create_mcp_server(memory: MemoryEngine, multi_bank: bool = True) -> FastMCP:
     """
     Create and configure the Hindsight MCP server.
 
     Args:
         memory: MemoryEngine instance (required)
+        multi_bank: If True, expose all tools with bank_id parameters (default).
+                   If False, only expose bank-scoped tools without bank_id parameters.
 
     Returns:
-        Configured FastMCP server instance
+        Configured FastMCP server instance with stateless_http enabled
     """
-    mcp = FastMCP("hindsight-mcp-server")
+    # Use stateless_http=True for Claude Code compatibility
+    mcp = FastMCP("hindsight-mcp-server", stateless_http=True)
 
-    @mcp.tool()
-    async def retain(content: str, context: str = "general") -> str:
-        """
-        Store important information to long-term memory.
+    # Configure and register tools using shared module
+    config = MCPToolsConfig(
+        bank_id_resolver=get_current_bank_id,
+        api_key_resolver=get_current_api_key,  # Propagate API key for tenant auth
+        tenant_id_resolver=get_current_tenant_id,  # Propagate tenant_id for usage metering
+        api_key_id_resolver=get_current_api_key_id,  # Propagate api_key_id for usage metering
+        include_bank_id_param=multi_bank,
+        tools=None
+        if multi_bank
+        else {
+            "retain",
+            "recall",
+            "reflect",
+            "list_mental_models",
+            "get_mental_model",
+            "create_mental_model",
+            "update_mental_model",
+            "delete_mental_model",
+            "refresh_mental_model",
+        },  # Scoped tools for single-bank mode (excludes bank management: list_banks, create_bank)
+        retain_fire_and_forget=False,  # HTTP MCP supports sync/async modes
+    )
 
-        Use this tool PROACTIVELY whenever the user shares:
-        - Personal facts, preferences, or interests
-        - Important events or milestones
-        - User history, experiences, or background
-        - Decisions, opinions, or stated preferences
-        - Goals, plans, or future intentions
-        - Relationships or people mentioned
-        - Work context, projects, or responsibilities
+    register_mcp_tools(mcp, memory, config)
 
-        Args:
-            content: The fact/memory to store (be specific and include relevant details)
-            context: Category for the memory (e.g., 'preferences', 'work', 'hobbies', 'family'). Default: 'general'
-        """
-        try:
-            bank_id = get_current_bank_id()
-            await memory.put_batch_async(
-                bank_id=bank_id,
-                contents=[{"content": content, "context": context}]
-            )
-            return "Memory stored successfully"
-        except Exception as e:
-            logger.error(f"Error storing memory: {e}", exc_info=True)
-            return f"Error: {str(e)}"
+    # Load and register additional tools from MCP extension if configured
+    mcp_extension = load_extension("MCP", MCPExtension)
+    if mcp_extension:
+        logger.info(f"Loading MCP extension: {mcp_extension.__class__.__name__}")
+        mcp_extension.register_tools(mcp, memory)
 
-    @mcp.tool()
-    async def recall(query: str, max_results: int = 10) -> str:
-        """
-        Search memories to provide personalized, context-aware responses.
-
-        Use this tool PROACTIVELY to:
-        - Check user's preferences before making suggestions
-        - Recall user's history to provide continuity
-        - Remember user's goals and context
-        - Personalize responses based on past interactions
-
-        Args:
-            query: Natural language search query (e.g., "user's food preferences", "what projects is user working on")
-            max_results: Maximum number of results to return (default: 10)
-        """
-        try:
-            bank_id = get_current_bank_id()
-            from hindsight_api.engine.memory_engine import Budget
-            search_result = await memory.recall_async(
-                bank_id=bank_id,
-                query=query,
-                fact_type=list(VALID_RECALL_FACT_TYPES),
-                budget=Budget.LOW
-            )
-
-            results = [
-                {
-                    "id": fact.id,
-                    "text": fact.text,
-                    "type": fact.fact_type,
-                    "context": fact.context,
-                    "event_date": fact.event_date,
-                }
-                for fact in search_result.results[:max_results]
-            ]
-
-            return json.dumps({"results": results}, indent=2)
-        except Exception as e:
-            logger.error(f"Error searching: {e}", exc_info=True)
-            return json.dumps({"error": str(e), "results": []})
+    # Make all tools tolerant of extra arguments from LLMs (e.g., "explanation")
+    _make_tools_tolerant(mcp)
 
     return mcp
 
 
-class MCPMiddleware:
-    """ASGI middleware that extracts bank_id from path and sets context."""
+def _make_tools_tolerant(mcp: FastMCP) -> None:
+    """Wrap all tool run methods to strip unknown arguments before validation.
 
-    def __init__(self, app, memory: MemoryEngine):
+    LLMs frequently add extra fields like "explanation" or "reasoning" to tool calls.
+    FastMCP's Pydantic TypeAdapter rejects these with "Unexpected keyword argument".
+    This wraps each tool's run() to filter arguments to only known parameters.
+    """
+    try:
+        for name, tool in mcp._tool_manager._tools.items():
+            if hasattr(tool, "parameters") and tool.parameters:
+                allowed = set(tool.parameters.get("properties", {}).keys())
+                original_run = tool.run
+
+                async def _tolerant_run(arguments, _allowed=allowed, _orig=original_run):
+                    extra_keys = set(arguments.keys()) - _allowed
+                    if extra_keys:
+                        logger.debug(f"Stripping unknown arguments from tool call: {extra_keys}")
+                        arguments = {k: v for k, v in arguments.items() if k in _allowed}
+                    return await _orig(arguments)
+
+                # FunctionTool is a Pydantic model with extra='forbid', so use
+                # object.__setattr__ to bypass Pydantic's setter validation.
+                object.__setattr__(tool, "run", _tolerant_run)
+    except (AttributeError, KeyError) as e:
+        logger.warning(f"Could not make tools tolerant of extra arguments: {e}")
+
+
+class MCPMiddleware:
+    """ASGI middleware that intercepts MCP requests and routes to appropriate MCP server.
+
+    This middleware wraps the main FastAPI app and intercepts requests matching the
+    configured prefix (default: /mcp). Non-MCP requests pass through to the inner app.
+
+    Authentication:
+        1. If HINDSIGHT_API_MCP_AUTH_TOKEN is set (legacy), validates against that token
+        2. Otherwise, uses TenantExtension.authenticate_mcp() from the MemoryEngine
+           - DefaultTenantExtension: no auth required (local dev)
+           - ApiKeyTenantExtension: validates against env var
+
+    Two modes based on URL structure:
+
+    1. Multi-bank mode (for /mcp/ root endpoint):
+       - Exposes all tools: retain, recall, reflect, list_banks, create_bank
+       - All tools include optional bank_id parameter for cross-bank operations
+       - Bank ID from: X-Bank-Id header or HINDSIGHT_MCP_BANK_ID env var
+
+    2. Single-bank mode (for /mcp/{bank_id}/ endpoints):
+       - Exposes bank-scoped tools only: retain, recall, reflect
+       - No bank_id parameter (comes from URL)
+       - No bank management tools (list_banks, create_bank)
+       - Recommended for agent isolation
+
+    Bank ID resolution priority:
+        1. URL path (e.g., /mcp/{bank_id}/) → single-bank mode
+        2. X-Bank-Id header → multi-bank mode
+        3. HINDSIGHT_MCP_BANK_ID env var → multi-bank mode (default: "default")
+
+    Examples:
+        # Single-bank mode (recommended for agent isolation)
+        claude mcp add --transport http my-agent http://localhost:8888/mcp/my-agent-bank/ \\
+            --header "Authorization: Bearer <token>"
+
+        # Multi-bank mode (for cross-bank operations)
+        claude mcp add --transport http hindsight http://localhost:8888/mcp \\
+            --header "X-Bank-Id: my-bank" --header "Authorization: Bearer <token>"
+    """
+
+    def __init__(
+        self,
+        app,
+        memory: MemoryEngine,
+        prefix: str = "/mcp",
+        multi_bank_app=None,
+        single_bank_app=None,
+        multi_bank_server=None,
+        single_bank_server=None,
+    ):
         self.app = app
+        self.prefix = prefix
         self.memory = memory
-        self.mcp_server = create_mcp_server(memory)
-        # Use sse_app - http_app requires lifespan management that's complex with middleware
-        import warnings
-        with warnings.catch_warnings():
-            warnings.simplefilter("ignore", DeprecationWarning)
-            self.mcp_app = self.mcp_server.sse_app()
+        self.tenant_extension = memory._tenant_extension
+
+        if multi_bank_app and single_bank_app:
+            # Pre-created servers (used when called via add_middleware from create_app)
+            self.multi_bank_app = multi_bank_app
+            self.single_bank_app = single_bank_app
+            self.multi_bank_server = multi_bank_server
+            self.single_bank_server = single_bank_server
+        else:
+            # Create servers internally (for direct construction / tests)
+            self.multi_bank_server = create_mcp_server(memory, multi_bank=True)
+            self.multi_bank_app = self.multi_bank_server.http_app(path="/")
+            self.single_bank_server = create_mcp_server(memory, multi_bank=False)
+            self.single_bank_app = self.single_bank_server.http_app(path="/")
+
+    def _get_header(self, scope: dict, name: str) -> str | None:
+        """Extract a header value from ASGI scope."""
+        name_lower = name.lower().encode()
+        for header_name, header_value in scope.get("headers", []):
+            if header_name.lower() == name_lower:
+                return header_value.decode()
+        return None
 
     async def __call__(self, scope, receive, send):
         if scope["type"] != "http":
-            await self.mcp_app(scope, receive, send)
+            await self.app(scope, receive, send)
             return
 
         path = scope.get("path", "")
 
-        # Strip any mount prefix (e.g., /mcp) that FastAPI might not have stripped
-        root_path = scope.get("root_path", "")
-        if root_path and path.startswith(root_path):
-            path = path[len(root_path):] or "/"
-
-        # Also handle case where mount path wasn't stripped (e.g., /mcp/...)
-        if path.startswith("/mcp/"):
-            path = path[4:]  # Remove /mcp prefix
-
-        # Extract bank_id from path: /{bank_id}/ or /{bank_id}
-        # http_app expects requests at /
-        if not path.startswith("/") or len(path) <= 1:
-            # No bank_id in path - return error
-            await self._send_error(send, 400, "bank_id required in path: /mcp/{bank_id}/")
+        # Check if this is an MCP request (matches prefix)
+        if not (path == self.prefix or path.startswith(self.prefix + "/")):
+            # Not an MCP request — pass through to the inner app
+            await self.app(scope, receive, send)
             return
 
-        # Extract bank_id from first path segment
-        parts = path[1:].split("/", 1)
-        if not parts[0]:
-            await self._send_error(send, 400, "bank_id required in path: /mcp/{bank_id}/")
-            return
+        # Strip prefix from path
+        path = path[len(self.prefix) :] or "/"
 
-        bank_id = parts[0]
-        new_path = "/" + parts[1] if len(parts) > 1 else "/"
+        # Extract auth token from header (for tenant auth propagation)
+        auth_header = self._get_header(scope, "Authorization")
+        auth_token: str | None = None
+        if auth_header:
+            # Support both "Bearer <token>" and direct token
+            auth_token = auth_header[7:].strip() if auth_header.startswith("Bearer ") else auth_header.strip()
 
-        # Set bank_id context
-        token = _current_bank_id.set(bank_id)
+        # Authenticate: check legacy MCP_AUTH_TOKEN first, then TenantExtension
+        tenant_context = None
+        auth_tenant_id: str | None = None
+        auth_api_key_id: str | None = None
+        if MCP_AUTH_TOKEN:
+            # Legacy authentication mode - validate against static token
+            if not auth_token:
+                await self._send_error(send, 401, "Authorization header required")
+                return
+            if auth_token != MCP_AUTH_TOKEN:
+                await self._send_error(send, 401, "Invalid authentication token")
+                return
+            # Legacy mode doesn't use tenant schemas
+            tenant_context = None
+        else:
+            # Use TenantExtension.authenticate_mcp() for auth
+            try:
+                auth_context = RequestContext(api_key=auth_token)
+                tenant_context = await self.tenant_extension.authenticate_mcp(auth_context)
+                # Capture tenant_id and api_key_id set by authenticate() for usage metering
+                auth_tenant_id = auth_context.tenant_id
+                auth_api_key_id = auth_context.api_key_id
+            except AuthenticationError as e:
+                await self._send_error(send, 401, str(e))
+                return
+
+        # Set schema from tenant context so downstream DB queries use the correct schema
+        schema_token = (
+            _current_schema.set(tenant_context.schema_name) if tenant_context and tenant_context.schema_name else None
+        )
+
+        # Resolve bank_id: path takes priority over header.
+        # Path = user's explicit connection endpoint (e.g., /mcp/my-bank/).
+        # X-Bank-Id header = per-request override for multi-bank mode only.
+        bank_id = None
+        bank_id_from_path = False
+        new_path = path
+
+        # First, try to extract from path: /{bank_id}/...
+        if path.startswith("/") and len(path) > 1:
+            parts = path[1:].split("/", 1)
+            if parts[0]:
+                bank_id = parts[0]
+                bank_id_from_path = True
+                new_path = "/" + parts[1] if len(parts) > 1 else "/"
+
+        # If no path-based bank_id, try X-Bank-Id header (multi-bank mode)
+        if not bank_id:
+            bank_id = self._get_header(scope, "X-Bank-Id")
+
+        # Fall back to default bank_id
+        if not bank_id:
+            bank_id = DEFAULT_BANK_ID
+            logger.debug(f"Using default bank_id: {bank_id}")
+
+        # Select the appropriate MCP app based on how bank_id was provided:
+        # - Path-based bank_id → single-bank app (no bank_id param, scoped tools)
+        # - Header/env bank_id → multi-bank app (bank_id param, all tools)
+        target_app = self.single_bank_app if bank_id_from_path else self.multi_bank_app
+
+        # Set bank_id, api_key, tenant_id, and api_key_id context
+        bank_id_token = _current_bank_id.set(bank_id)
+        # Store the auth token for tenant extension to validate
+        api_key_token = _current_api_key.set(auth_token) if auth_token else None
+        # Store tenant_id and api_key_id from authentication for usage metering
+        tenant_id_token = _current_tenant_id.set(auth_tenant_id) if auth_tenant_id else None
+        api_key_id_token = _current_api_key_id.set(auth_api_key_id) if auth_api_key_id else None
         try:
             new_scope = scope.copy()
             new_scope["path"] = new_path
+            # Clear root_path since we're passing directly to the app
+            new_scope["root_path"] = ""
+
+            # Wrap send to rewrite the SSE endpoint URL to include bank_id if using path-based routing.
+            # Only rewrite SSE (text/event-stream) responses to avoid corrupting tool results
+            # that might contain the literal string "data: /messages".
+            is_sse_response = False
 
-            # Wrap send to rewrite the SSE endpoint URL to include bank_id
-            # The SSE app sends "event: endpoint\ndata: /messages\n" but we need
-            # the client to POST to /{bank_id}/messages instead
             async def send_wrapper(message):
-                if message["type"] == "http.response.body":
+                nonlocal is_sse_response
+                if message["type"] == "http.response.start":
+                    for header_name, header_value in message.get("headers", []):
+                        if header_name == b"content-type" and b"text/event-stream" in header_value:
+                            is_sse_response = True
+                            break
+                if message["type"] == "http.response.body" and bank_id_from_path and is_sse_response:
                     body = message.get("body", b"")
                     if body and b"/messages" in body:
                         # Rewrite /messages to /{bank_id}/messages in SSE endpoint event
-                        body = body.replace(
-                            b"data: /messages",
-                            f"data: /{bank_id}/messages".encode()
-                        )
+                        body = body.replace(b"data: /messages", f"data: /{bank_id}/messages".encode())
                         message = {**message, "body": body}
                 await send(message)
 
-            await self.mcp_app(new_scope, receive, send_wrapper)
+            await target_app(new_scope, receive, send_wrapper)
         finally:
-            _current_bank_id.reset(token)
+            _current_bank_id.reset(bank_id_token)
+            if api_key_token is not None:
+                _current_api_key.reset(api_key_token)
+            if tenant_id_token is not None:
+                _current_tenant_id.reset(tenant_id_token)
+            if api_key_id_token is not None:
+                _current_api_key_id.reset(api_key_id_token)
+            if schema_token is not None:
+                _current_schema.reset(schema_token)
 
     async def _send_error(self, send, status: int, message: str):
         """Send an error response."""
         body = json.dumps({"error": message}).encode()
-        await send({
-            "type": "http.response.start",
-            "status": status,
-            "headers": [(b"content-type", b"application/json")],
-        })
-        await send({
-            "type": "http.response.body",
-            "body": body,
-        })
+        await send(
+            {
+                "type": "http.response.start",
+                "status": status,
+                "headers": [(b"content-type", b"application/json")],
+            }
+        )
+        await send(
+            {
+                "type": "http.response.body",
+                "body": body,
+            }
+        )
 
 
-def create_mcp_app(memory: MemoryEngine):
-    """
-    Create an ASGI app that handles MCP requests.
+def create_mcp_servers(memory: MemoryEngine):
+    """Create multi-bank and single-bank MCP servers and their Starlette apps.
 
-    URL pattern: /mcp/{bank_id}/
-
-    The bank_id is extracted from the URL path and made available to tools.
-
-    Args:
-        memory: MemoryEngine instance
+    Returns the servers and apps separately so lifespans can be chained before
+    the middleware wraps the main app.
 
     Returns:
-        ASGI application
+        Tuple of (multi_bank_server, single_bank_server, multi_bank_app, single_bank_app)
     """
-    return MCPMiddleware(None, memory)
+    multi_bank_server = create_mcp_server(memory, multi_bank=True)
+    multi_bank_app = multi_bank_server.http_app(path="/")
+
+    single_bank_server = create_mcp_server(memory, multi_bank=False)
+    single_bank_app = single_bank_server.http_app(path="/")
+
+    return multi_bank_server, single_bank_server, multi_bank_app, single_bank_app

hindsight-api/hindsight_api/banner.py

@@ -0,0 +1,101 @@
+"""
+Banner display for Hindsight API startup.
+
+Shows the logo and tagline with gradient colors.
+"""
+
+from .utils import mask_network_location
+
+# Gradient colors: #0074d9 -> #009296
+GRADIENT_START = (0, 116, 217)  # #0074d9
+GRADIENT_END = (0, 146, 150)  # #009296
+
+# Pre-generated logo (generated by test-logo.py)
+LOGO = """\
+  \033[38;2;9;127;184m\u2584\033[0m\033[48;2;8;130;178m\033[38;2;5;133;186m\u2584\033[0m       \033[48;2;10;143;160m\033[38;2;10;143;165m\u2584\033[0m\033[38;2;7;140;156m\u2584\033[0m
+ \033[38;2;8;125;192m\u2584\033[0m \033[38;2;3;132;191m\u2580\033[0m\033[38;2;2;133;192m\u2584\033[0m \033[38;2;3;132;180m\u2584\033[0m\033[38;2;1;137;184m\u2584\033[0m\033[38;2;3;133;174m\u2584\033[0m \033[38;2;3;142;176m\u2584\033[0m\033[38;2;4;142;169m\u2580\033[0m \033[38;2;10;144;164m\u2584\033[0m
+\033[38;2;6;121;195m\u2580\033[0m\033[38;2;5;128;203m\u2580\033[0m\033[48;2;5;124;195m\033[38;2;3;125;200m\u2584\033[0m\033[38;2;2;126;196m\u2584\033[0m\033[48;2;3;128;188m\033[38;2;1;131;196m\u2584\033[0m\033[48;2;0;152;219m\033[38;2;2;131;191m\u2584\033[0m\033[38;2;1;141;196m\u2580\033[0m\033[38;2;1;135;183m\u2580\033[0m\033[38;2;1;148;198m\u2580\033[0m\033[48;2;1;156;202m\033[38;2;2;135;180m\u2584\033[0m\033[48;2;4;134;169m\033[38;2;1;137;177m\u2584\033[0m\033[38;2;3;138;173m\u2584\033[0m\033[48;2;6;137;165m\033[38;2;2;140;170m\u2584\033[0m\033[38;2;7;144;169m\u2580\033[0m\033[38;2;7;139;158m\u2580\033[0m
+   \033[48;2;2;128;202m\033[38;2;2;124;201m\u2584\033[0m\033[48;2;1;130;201m\033[38;2;0;135;212m\u2584\033[0m\033[38;2;2;128;196m\u2584\033[0m \033[48;2;2;142;204m\033[38;2;7;138;199m\u2584\033[0m \033[38;2;1;135;186m\u2584\033[0m\033[48;2;1;142;186m\033[38;2;2;144;194m\u2584\033[0m\033[48;2;3;138;176m\033[38;2;2;134;176m\u2584\033[0m
+ \033[48;2;8;118;200m\033[38;2;8;121;209m\u2584\033[0m\033[38;2;3;121;203m\u2580\033[0m \033[38;2;3;122;192m\u2580\033[0m\033[38;2;1;138;216m\u2580\033[0m\033[48;2;0;138;210m\033[38;2;3;128;198m\u2584\033[0m\033[48;2;0;126;188m\033[38;2;2;131;198m\u2584\033[0m\033[48;2;0;142;205m\033[38;2;3;132;193m\u2584\033[0m\033[38;2;1;140;196m\u2580\033[0m  \033[38;2;4;134;175m\u2580\033[0m\033[48;2;13;135;167m\033[38;2;8;136;174m\u2584\033[0m """
+
+
+def _interpolate_color(start: tuple, end: tuple, t: float) -> tuple:
+    """Interpolate between two RGB colors."""
+    return (
+        int(start[0] + (end[0] - start[0]) * t),
+        int(start[1] + (end[1] - start[1]) * t),
+        int(start[2] + (end[2] - start[2]) * t),
+    )
+
+
+def gradient_text(text: str, start: tuple = GRADIENT_START, end: tuple = GRADIENT_END) -> str:
+    """Render text with a gradient color effect."""
+    result = []
+    length = len(text)
+    for i, char in enumerate(text):
+        if char == " ":
+            result.append(" ")
+        else:
+            t = i / max(length - 1, 1)
+            r, g, b = _interpolate_color(start, end, t)
+            result.append(f"\033[38;2;{r};{g};{b}m{char}")
+    result.append("\033[0m")
+    return "".join(result)
+
+
+def print_banner():
+    """Print the Hindsight startup banner."""
+    print(LOGO)
+    tagline = gradient_text("Hindsight: Agent Memory That Works Like Human Memory")
+    print(f"\n  {tagline}\n")
+
+
+def color(text: str, t: float = 0.0) -> str:
+    """Color text using gradient position (0.0 = start, 1.0 = end)."""
+    r, g, b = _interpolate_color(GRADIENT_START, GRADIENT_END, t)
+    return f"\033[38;2;{r};{g};{b}m{text}\033[0m"
+
+
+def color_start(text: str) -> str:
+    """Color text with gradient start color (#0074d9)."""
+    return color(text, 0.0)
+
+
+def color_end(text: str) -> str:
+    """Color text with gradient end color (#009296)."""
+    return color(text, 1.0)
+
+
+def color_mid(text: str) -> str:
+    """Color text with gradient middle color."""
+    return color(text, 0.5)
+
+
+def dim(text: str) -> str:
+    """Dim/gray text."""
+    return f"\033[38;2;128;128;128m{text}\033[0m"
+
+
+def print_startup_info(
+    host: str,
+    port: int,
+    database_url: str,
+    llm_provider: str,
+    llm_model: str,
+    embeddings_provider: str,
+    reranker_provider: str,
+    mcp_enabled: bool = False,
+    version: str | None = None,
+):
+    """Print styled startup information."""
+    print(color_start("Starting Hindsight API..."))
+    if version:
+        print(f"  {dim('Version:')} {color(f'v{version}', 0.1)}")
+    print(f"  {dim('URL:')} {color(f'http://{host}:{port}', 0.2)}")
+    print(f"  {dim('Database:')} {color(mask_network_location(database_url), 0.4)}")
+    print(f"  {dim('LLM:')} {color(f'{llm_provider} / {llm_model}', 0.6)}")
+    print(f"  {dim('Embeddings:')} {color(embeddings_provider, 0.8)}")
+    print(f"  {dim('Reranker:')} {color(reranker_provider, 1.0)}")
+    if mcp_enabled:
+        print(f"  {dim('MCP:')} {color_end('enabled at /mcp')}")
+    print()

hindsight-api/hindsight_api/cli.py

@@ -1,128 +0,0 @@
-"""
-Command-line interface for Hindsight API.
-
-Run the server with:
-    hindsight-api
-
-Stop with Ctrl+C.
-"""
-import argparse
-import asyncio
-import atexit
-import os
-import signal
-import sys
-from typing import Optional
-
-import uvicorn
-
-from . import MemoryEngine
-from .api import create_app
-
-
-# Disable tokenizers parallelism to avoid warnings
-os.environ["TOKENIZERS_PARALLELISM"] = "false"
-
-# Global reference for cleanup
-_memory: Optional[MemoryEngine] = None
-
-
-def _cleanup():
-    """Synchronous cleanup function to stop resources on exit."""
-    global _memory
-    if _memory is not None and _memory._pg0 is not None:
-        try:
-            loop = asyncio.new_event_loop()
-            loop.run_until_complete(_memory._pg0.stop())
-            loop.close()
-            print("\npg0 stopped.")
-        except Exception as e:
-            print(f"\nError stopping pg0: {e}")
-
-
-def _signal_handler(signum, frame):
-    """Handle SIGINT/SIGTERM to ensure cleanup."""
-    print(f"\nReceived signal {signum}, shutting down...")
-    _cleanup()
-    sys.exit(0)
-
-
-def main():
-    """Main entry point for the CLI."""
-    global _memory
-
-    parser = argparse.ArgumentParser(
-        prog="hindsight-api",
-        description="Hindsight API Server",
-    )
-    parser.add_argument(
-        "--host", default="0.0.0.0",
-        help="Host to bind to (default: 0.0.0.0)"
-    )
-    parser.add_argument(
-        "--port", type=int, default=8888,
-        help="Port to bind to (default: 8888)"
-    )
-    parser.add_argument(
-        "--log-level", default="info",
-        choices=["critical", "error", "warning", "info", "debug", "trace"],
-        help="Log level (default: info)"
-    )
-    parser.add_argument(
-        "--access-log", action="store_true",
-        help="Enable access log"
-    )
-
-    args = parser.parse_args()
-
-    # Register cleanup handlers
-    atexit.register(_cleanup)
-    signal.signal(signal.SIGINT, _signal_handler)
-    signal.signal(signal.SIGTERM, _signal_handler)
-
-    # Get configuration from environment variables
-    db_url = os.getenv("HINDSIGHT_API_DATABASE_URL", "pg0")
-    llm_provider = os.getenv("HINDSIGHT_API_LLM_PROVIDER", "groq")
-    llm_api_key = os.getenv("HINDSIGHT_API_LLM_API_KEY", "")
-    llm_model = os.getenv("HINDSIGHT_API_LLM_MODEL", "openai/gpt-oss-20b")
-    llm_base_url = os.getenv("HINDSIGHT_API_LLM_BASE_URL") or None
-
-    # Create MemoryEngine
-    _memory = MemoryEngine(
-        db_url=db_url,
-        memory_llm_provider=llm_provider,
-        memory_llm_api_key=llm_api_key,
-        memory_llm_model=llm_model,
-        memory_llm_base_url=llm_base_url,
-    )
-
-    # Create FastAPI app
-    app = create_app(
-        memory=_memory,
-        http_api_enabled=True,
-        mcp_api_enabled=True,
-        mcp_mount_path="/mcp",
-        run_migrations=True,
-        initialize_memory=True,
-    )
-
-    # Prepare uvicorn config
-    uvicorn_config = {
-        "app": app,
-        "host": args.host,
-        "port": args.port,
-        "log_level": args.log_level,
-        "access_log": args.access_log,
-    }
-
-    print(f"\nStarting Hindsight API...")
-    print(f"  URL: http://{args.host}:{args.port}")
-    print(f"  Database: {db_url}")
-    print(f"  LLM Provider: {llm_provider}")
-    print()
-
-    uvicorn.run(**uvicorn_config)
-
-
-if __name__ == "__main__":
-    main()

hindsight-api/hindsight_api/config_resolver.py

@@ -0,0 +1,274 @@
+"""
+Configuration resolution with hierarchical overrides.
+
+Resolves config values through the hierarchy:
+  Global (env vars) → Tenant config (via extension) → Bank config (database)
+
+Config values are resolved on every request to ensure consistency across
+multiple API servers.
+"""
+
+import json
+import logging
+from dataclasses import asdict
+from typing import Any
+
+import asyncpg
+
+from hindsight_api.config import HindsightConfig, _get_raw_config, normalize_config_dict
+from hindsight_api.extensions.tenant import TenantExtension
+from hindsight_api.models import RequestContext
+
+logger = logging.getLogger(__name__)
+
+
+class ConfigResolver:
+    """Resolves hierarchical configuration with tenant/bank overrides."""
+
+    def __init__(self, pool: asyncpg.Pool, tenant_extension: TenantExtension | None = None):
+        """
+        Initialize config resolver.
+
+        Args:
+            pool: Database connection pool
+            tenant_extension: Optional tenant extension for tenant-level config and permissions
+        """
+        self.pool = pool
+        self.tenant_extension = tenant_extension
+        self._global_config = _get_raw_config()
+        self._configurable_fields = HindsightConfig.get_configurable_fields()
+        self._credential_fields = HindsightConfig.get_credential_fields()
+
+    async def resolve_full_config(self, bank_id: str, context: RequestContext | None = None) -> HindsightConfig:
+        """
+        Resolve full HindsightConfig for a bank with hierarchical overrides applied.
+
+        This is for INTERNAL USE ONLY. Returns the complete config object with all fields
+        including credentials and static fields. Use get_bank_config() for API responses.
+
+        Resolution order:
+        1. Global config (from environment variables)
+        2. Tenant config overrides (from TenantExtension.get_tenant_config())
+        3. Bank config overrides (from banks.config JSONB)
+
+        Args:
+            bank_id: Bank identifier
+            context: Request context for tenant config resolution
+
+        Returns:
+            Complete HindsightConfig with hierarchical overrides applied
+        """
+        # Start with global config (all fields)
+        config_dict = asdict(self._global_config)
+
+        # Load tenant config overrides (if tenant extension available)
+        if self.tenant_extension and context:
+            try:
+                tenant_overrides = await self.tenant_extension.get_tenant_config(context)
+                if tenant_overrides:
+                    # Normalize keys and filter to configurable fields only
+                    normalized_tenant = normalize_config_dict(tenant_overrides)
+                    configurable_tenant = {k: v for k, v in normalized_tenant.items() if k in self._configurable_fields}
+                    config_dict.update(configurable_tenant)
+                    logger.debug(
+                        f"Applied tenant config overrides for bank {bank_id}: {list(configurable_tenant.keys())}"
+                    )
+            except Exception as e:
+                logger.warning(f"Failed to load tenant config for bank {bank_id}: {e}")
+
+        # Load bank config overrides
+        bank_overrides = await self._load_bank_config(bank_id)
+        if bank_overrides:
+            config_dict.update(bank_overrides)
+            logger.debug(f"Applied bank config overrides for bank {bank_id}: {list(bank_overrides.keys())}")
+
+        # Return full config object (dataclass doesn't have __init__ that accepts kwargs, so we update the object)
+        # Create a new config instance by copying the global config and updating fields
+        resolved_config = HindsightConfig(**config_dict)
+        return resolved_config
+
+    async def get_bank_config(self, bank_id: str, context: RequestContext | None = None) -> dict[str, Any]:
+        """
+        Get fully resolved config for a bank (filtered by permissions).
+
+        Resolution order:
+        1. Global config (from environment variables)
+        2. Tenant config overrides (from TenantExtension.get_tenant_config())
+        3. Bank config overrides (from banks.config JSONB)
+
+        Note: Config is resolved on every call (not cached) to ensure consistency
+        across multiple API servers.
+
+        SECURITY:
+        - Only returns configurable fields (excludes static/infrastructure fields)
+        - Filters out ALL credential fields (API keys, base URLs, etc.)
+        - Further filtered by tenant/bank permissions if extension provides them
+
+        Args:
+            bank_id: Bank identifier
+            context: Request context for tenant config resolution and permissions
+
+        Returns:
+            Dict of allowed configurable fields only (never includes credentials or static fields)
+        """
+        # Resolve full config with all hierarchical overrides
+        resolved_config = await self.resolve_full_config(bank_id, context)
+        config_dict = asdict(resolved_config)
+
+        # SECURITY: Filter to only configurable fields (exclude static/infrastructure)
+        filtered = {k: v for k, v in config_dict.items() if k in self._configurable_fields}
+
+        # SECURITY: Remove ALL credential fields (API keys, base URLs, etc.)
+        filtered = {k: v for k, v in filtered.items() if k not in self._credential_fields}
+
+        # PERMISSIONS: Further filter based on tenant/bank permissions
+        if self.tenant_extension and context:
+            try:
+                allowed_fields = await self.tenant_extension.get_allowed_config_fields(context, bank_id)
+                if allowed_fields is not None:  # None means "allow all"
+                    filtered = {k: v for k, v in filtered.items() if k in allowed_fields}
+                    logger.debug(
+                        f"Applied permission filter for bank {bank_id}: allowed={len(allowed_fields)} fields, "
+                        f"returned={len(filtered)} fields"
+                    )
+            except Exception as e:
+                logger.warning(f"Failed to load permissions for bank {bank_id}: {e}")
+
+        return filtered
+
+    async def _load_bank_config(self, bank_id: str) -> dict[str, Any]:
+        """
+        Load bank config overrides from banks.config JSONB column.
+
+        Args:
+            bank_id: Bank identifier
+
+        Returns:
+            Dict of config overrides (only configurable fields, normalized keys)
+        """
+        try:
+            async with self.pool.acquire() as conn:
+                row = await conn.fetchrow(
+                    """
+                    SELECT config FROM banks WHERE bank_id = $1
+                    """,
+                    bank_id,
+                )
+
+                if row and row["config"]:
+                    config_data = row["config"]
+
+                    # Handle case where JSONB is returned as JSON string
+                    if isinstance(config_data, str):
+                        config_data = json.loads(config_data)
+
+                    # Normalize keys (handle both env var format and Python field format)
+                    normalized = normalize_config_dict(config_data)
+
+                    # Only return overrides for configurable fields
+                    return {k: v for k, v in normalized.items() if k in self._configurable_fields}
+        except Exception as e:
+            logger.error(f"Failed to load bank config for {bank_id}: {e}")
+
+        return {}
+
+    async def update_bank_config(
+        self, bank_id: str, updates: dict[str, Any], context: RequestContext | None = None
+    ) -> None:
+        """
+        Update bank configuration overrides (with permission checking).
+
+        Args:
+            bank_id: Bank identifier
+            updates: Dict of config field names to new values.
+                    Keys can be in env var format (HINDSIGHT_API_LLM_PROVIDER)
+                    or Python field format (llm_provider).
+                    Only configurable fields are allowed.
+            context: Request context for permission checking
+
+        Raises:
+            ValueError: If attempting to override invalid/disallowed fields
+        """
+        # Normalize keys
+        normalized_updates = normalize_config_dict(updates)
+
+        # SECURITY: Reject credential fields explicitly
+        credential_attempts = set(normalized_updates.keys()) & self._credential_fields
+        if credential_attempts:
+            raise ValueError(
+                f"Cannot set credential fields via API: {sorted(credential_attempts)}. "
+                f"Credentials (API keys, base URLs) must be set at server level only."
+            )
+
+        # Validate all fields are configurable
+        invalid_fields = set(normalized_updates.keys()) - self._configurable_fields
+        if invalid_fields:
+            static_fields = HindsightConfig.get_static_fields()
+            invalid_static = invalid_fields & static_fields
+            if invalid_static:
+                raise ValueError(
+                    f"Cannot override static (server-level) fields: {sorted(invalid_static)}. "
+                    f"Only configurable fields can be overridden per-bank. "
+                    f"Configurable fields include: {sorted(list(self._configurable_fields)[:10])}... "
+                    f"(total: {len(self._configurable_fields)} fields)"
+                )
+            else:
+                raise ValueError(
+                    f"Unknown configuration fields: {sorted(invalid_fields)}. "
+                    f"Valid configurable fields: {sorted(list(self._configurable_fields)[:10])}..."
+                )
+
+        # PERMISSIONS: Check tenant/bank permissions
+        if self.tenant_extension and context:
+            try:
+                allowed_fields = await self.tenant_extension.get_allowed_config_fields(context, bank_id)
+                if allowed_fields is not None:  # None means "allow all"
+                    disallowed = set(normalized_updates.keys()) - allowed_fields
+                    if disallowed:
+                        raise ValueError(
+                            f"Not allowed to modify fields: {sorted(disallowed)}. "
+                            f"Your permissions allow: {sorted(list(allowed_fields)[:10])}..."
+                            if allowed_fields
+                            else "Not allowed to modify fields: {sorted(disallowed)}. "
+                            "Your permissions do not allow any config modifications."
+                        )
+            except ValueError:
+                raise  # Re-raise permission errors
+            except Exception as e:
+                logger.warning(f"Failed to check permissions for bank {bank_id}: {e}")
+                # Continue without permission check (fail open for backward compatibility)
+
+        # Merge with existing config (JSONB || operator)
+        async with self.pool.acquire() as conn:
+            await conn.execute(
+                """
+                UPDATE banks
+                SET config = config || $1::jsonb,
+                    updated_at = now()
+                WHERE bank_id = $2
+                """,
+                json.dumps(normalized_updates),
+                bank_id,
+            )
+
+        logger.info(f"Updated bank config for {bank_id}: {list(normalized_updates.keys())}")
+
+    async def reset_bank_config(self, bank_id: str) -> None:
+        """
+        Reset bank configuration to defaults (remove all overrides).
+
+        Args:
+            bank_id: Bank identifier
+        """
+        async with self.pool.acquire() as conn:
+            await conn.execute(
+                """
+                UPDATE banks
+                SET config = '{}'::jsonb,
+                    updated_at = now()
+                WHERE bank_id = $1
+                """,
+                bank_id,
+            )
+
+        logger.info(f"Reset bank config for {bank_id} to defaults")

hindsight-api/hindsight_api/daemon.py

@@ -0,0 +1,113 @@
+"""
+Daemon mode support for Hindsight API.
+
+Provides idle timeout for running as a background daemon.
+"""
+
+import asyncio
+import logging
+import os
+import sys
+import time
+from pathlib import Path
+
+logger = logging.getLogger(__name__)
+
+# Default daemon configuration
+DEFAULT_DAEMON_PORT = 8888
+DEFAULT_IDLE_TIMEOUT = 0  # 0 = no auto-exit (hindsight-embed passes its own timeout)
+
+# Allow override via environment variable for profile-specific logs
+DAEMON_LOG_PATH = Path(os.getenv("HINDSIGHT_API_DAEMON_LOG", str(Path.home() / ".hindsight" / "daemon.log")))
+
+
+class IdleTimeoutMiddleware:
+    """ASGI middleware that tracks activity and exits after idle timeout."""
+
+    def __init__(self, app, idle_timeout: int = DEFAULT_IDLE_TIMEOUT):
+        self.app = app
+        self.idle_timeout = idle_timeout
+        self.last_activity = time.time()
+        self._checker_task = None
+
+    async def __call__(self, scope, receive, send):
+        # Update activity timestamp on each request
+        self.last_activity = time.time()
+        await self.app(scope, receive, send)
+
+    def start_idle_checker(self):
+        """Start the background task that checks for idle timeout."""
+        self._checker_task = asyncio.create_task(self._check_idle())
+
+    async def _check_idle(self):
+        """Background task that exits the process after idle timeout."""
+        # If idle_timeout is 0, don't auto-exit
+        if self.idle_timeout <= 0:
+            return
+
+        while True:
+            await asyncio.sleep(30)  # Check every 30 seconds
+            idle_time = time.time() - self.last_activity
+            if idle_time > self.idle_timeout:
+                logger.info(f"Idle timeout reached ({self.idle_timeout}s), shutting down daemon")
+                # Give a moment for any in-flight requests
+                await asyncio.sleep(1)
+                # Send SIGTERM to ourselves to trigger graceful shutdown
+                import signal
+
+                os.kill(os.getpid(), signal.SIGTERM)
+
+
+def daemonize():
+    """
+    Fork the current process into a background daemon.
+
+    Uses double-fork technique to properly detach from terminal.
+    """
+    # First fork - detach from parent
+    try:
+        pid = os.fork()
+        if pid > 0:
+            sys.exit(0)
+    except OSError as e:
+        sys.stderr.write(f"fork #1 failed: {e}\n")
+        sys.exit(1)
+
+    # Decouple from parent environment
+    os.chdir("/")
+    os.setsid()
+    os.umask(0)
+
+    # Second fork - prevent zombie
+    pid = os.fork()
+    if pid > 0:
+        sys.exit(0)
+
+    # Redirect standard file descriptors to log file
+    DAEMON_LOG_PATH.parent.mkdir(parents=True, exist_ok=True)
+
+    sys.stdout.flush()
+    sys.stderr.flush()
+
+    # Redirect stdin to /dev/null
+    with open("/dev/null", "r") as devnull:
+        os.dup2(devnull.fileno(), sys.stdin.fileno())
+
+    # Redirect stdout/stderr to log file
+    log_fd = open(DAEMON_LOG_PATH, "a")
+    os.dup2(log_fd.fileno(), sys.stdout.fileno())
+    os.dup2(log_fd.fileno(), sys.stderr.fileno())
+
+
+def check_daemon_running(port: int = DEFAULT_DAEMON_PORT) -> bool:
+    """Check if a daemon is running and responsive on the given port."""
+    import socket
+
+    try:
+        sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
+        sock.settimeout(1)
+        result = sock.connect_ex(("127.0.0.1", port))
+        sock.close()
+        return result == 0
+    except Exception:
+        return False

hindsight-api/hindsight_api/engine/__init__.py

@@ -7,29 +7,40 @@ This package contains all the implementation details of the memory engine:
 - Supporting modules: embeddings, cross_encoder, entity_resolver, etc.
 """
 
-from .memory_engine import MemoryEngine
+from .cross_encoder import CrossEncoderModel, LocalSTCrossEncoder, RemoteTEICrossEncoder
 from .db_utils import acquire_with_retry
-from .embeddings import Embeddings, SentenceTransformersEmbeddings
+from .embeddings import Embeddings, LocalSTEmbeddings, RemoteTEIEmbeddings
+from .llm_wrapper import LLMConfig
+from .memory_engine import (
+    MemoryEngine,
+    UnqualifiedTableError,
+    fq_table,
+    get_current_schema,
+    validate_sql_schema,
+)
+from .response_models import MemoryFact, RecallResult, ReflectResult
 from .search.trace import (
-    SearchTrace,
-    QueryInfo,
     EntryPoint,
-    NodeVisit,
-    WeightComponents,
     LinkInfo,
+    NodeVisit,
     PruningDecision,
-    SearchSummary,
+    QueryInfo,
     SearchPhaseMetrics,
+    SearchSummary,
+    SearchTrace,
+    WeightComponents,
 )
 from .search.tracer import SearchTracer
-from .llm_wrapper import LLMConfig
-from .response_models import RecallResult, ReflectResult, MemoryFact
 
 __all__ = [
     "MemoryEngine",
     "acquire_with_retry",
     "Embeddings",
-    "SentenceTransformersEmbeddings",
+    "LocalSTEmbeddings",
+    "RemoteTEIEmbeddings",
+    "CrossEncoderModel",
+    "LocalSTCrossEncoder",
+    "RemoteTEICrossEncoder",
     "SearchTrace",
     "SearchTracer",
     "QueryInfo",
@@ -44,4 +55,9 @@ __all__ = [
     "RecallResult",
     "ReflectResult",
     "MemoryFact",
+    # Schema safety utilities
+    "fq_table",
+    "get_current_schema",
+    "validate_sql_schema",
+    "UnqualifiedTableError",
 ]

hindsight-api/hindsight_api/engine/consolidation/__init__.py

@@ -0,0 +1,5 @@
+"""Consolidation engine for automatic learning creation from memories."""
+
+from .consolidator import run_consolidation_job
+
+__all__ = ["run_consolidation_job"]

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

@@ -0,0 +1,85 @@
+"""Prompts for the consolidation engine."""
+
+CONSOLIDATION_SYSTEM_PROMPT = """You are a memory consolidation system. Your job is to convert facts into durable knowledge (observations) and merge with existing knowledge when appropriate.
+
+You must output ONLY valid JSON with no markdown code blocks or additional text. However, the "text" field within each observation should use markdown formatting (headers, lists, bold, etc.) for clarity and readability.
+
+## EXTRACT DURABLE KNOWLEDGE, NOT EPHEMERAL STATE
+Facts often describe events or actions. Extract the DURABLE KNOWLEDGE implied by the fact, not the transient state.
+
+Examples of extracting durable knowledge:
+- "User moved to Room 203" -> "Room 203 exists" (location exists, not where user is now)
+- "User visited Acme Corp at Room 105" -> "Acme Corp is located in Room 105"
+- "User took the elevator to floor 3" -> "Floor 3 is accessible by elevator"
+- "User met Sarah at the lobby" -> "Sarah can be found at the lobby"
+
+DO NOT track current user position/state as knowledge - that changes constantly.
+DO track permanent facts learned from the user's actions.
+
+## PRESERVE SPECIFIC DETAILS
+Keep names, locations, numbers, and other specifics. Do NOT:
+- Abstract into general principles
+- Generate business insights
+- Make knowledge generic
+
+GOOD examples:
+- Fact: "John likes pizza" -> "John likes pizza"
+- Fact: "Alice works at Google" -> "Alice works at Google"
+
+BAD examples:
+- "John likes pizza" -> "Understanding dietary preferences helps..." (TOO ABSTRACT)
+- "User is at Room 203" -> "User is currently at Room 203" (EPHEMERAL STATE)
+
+## MERGE RULES (when comparing to existing observations):
+1. REDUNDANT: Same information worded differently → update existing
+2. CONTRADICTION: Opposite information about same topic → update with temporal markers showing change
+   Example: "Alex used to love pizza but now hates it" OR "Alex's pizza preference changed from love to hate"
+3. UPDATE: New state replacing old state → update showing the transition with "used to", "now", "changed from X to Y"
+
+## CRITICAL RULES:
+- NEVER merge facts about DIFFERENT people
+- NEVER merge unrelated topics (food preferences vs work vs hobbies)
+- When merging contradictions, the "text" field MUST capture BOTH states with temporal markers:
+  * Use "used to X, now Y" OR "changed from X to Y" OR "X but now Y"
+  * DO NOT just state the new fact - you MUST show the change
+- Keep observations focused on ONE specific topic per person
+- The "text" field MUST contain durable knowledge, not ephemeral state
+- Do NOT include "tags" in output - tags are handled automatically"""
+
+CONSOLIDATION_USER_PROMPT = """Analyze this new fact and consolidate into knowledge.
+{mission_section}
+NEW FACT: {fact_text}
+
+EXISTING OBSERVATIONS (JSON array with source memories and dates):
+{observations_text}
+
+Each observation includes:
+- id: unique identifier for updating
+- text: the observation content
+- proof_count: number of supporting memories
+- tags: visibility scope (handled automatically)
+- created_at/updated_at: when observation was created/modified
+- occurred_start/occurred_end: temporal range of source facts
+- source_memories: array of supporting facts with their text and dates
+
+Instructions:
+1. Extract DURABLE KNOWLEDGE from the new fact (not ephemeral state)
+2. Review source_memories in existing observations to understand evidence
+3. Check dates to detect contradictions or updates
+4. Compare with observations:
+   - Same topic → UPDATE with learning_id
+   - New topic → CREATE new observation
+   - Purely ephemeral → return []
+
+Output JSON array of actions (the "text" field should use markdown formatting for structure):
+[
+  {{"action": "update", "learning_id": "uuid-from-observations", "text": "## Updated Knowledge\n\n**Key point**: details here\n\n- Supporting detail 1\n- Supporting detail 2", "reason": "..."}},
+  {{"action": "create", "text": "## New Durable Knowledge\n\nDescription with **emphasis** and proper structure", "reason": "..."}}
+]
+
+Return [] if fact contains no durable knowledge.
+
+IMPORTANT: Format the "text" field with markdown for better readability:
+- Use headers, lists, bold/italic, tables where appropriate
+- CRITICAL: Add blank lines before and after block elements (tables, code blocks, lists)
+- Ensure proper spacing for markdown to render correctly"""

hindsight-api/hindsight_api/engine/db_budget.py

@@ -0,0 +1,284 @@
+"""
+Database connection budget management.
+
+Limits concurrent database connections per operation to prevent
+a single operation (e.g., recall with parallel queries) from
+exhausting the connection pool.
+"""
+
+import asyncio
+import logging
+import uuid
+from contextlib import asynccontextmanager
+from dataclasses import dataclass, field
+from typing import TYPE_CHECKING, AsyncIterator
+
+if TYPE_CHECKING:
+    import asyncpg
+
+logger = logging.getLogger(__name__)
+
+
+@dataclass
+class OperationBudget:
+    """
+    Tracks connection budget for a single operation.
+
+    Each operation gets a semaphore limiting its concurrent connections.
+    """
+
+    operation_id: str
+    max_connections: int
+    semaphore: asyncio.Semaphore = field(init=False)
+    active_count: int = field(default=0, init=False)
+
+    def __post_init__(self):
+        self.semaphore = asyncio.Semaphore(self.max_connections)
+
+
+class ConnectionBudgetManager:
+    """
+    Manages per-operation connection budgets.
+
+    Usage:
+        manager = ConnectionBudgetManager(default_budget=4)
+
+        # Start an operation
+        async with manager.operation(max_connections=2) as op:
+            # Acquire connections within the budget
+            async with op.acquire(pool) as conn:
+                await conn.fetch(...)
+
+            # Multiple connections respect the budget
+            async with op.acquire(pool) as conn1, op.acquire(pool) as conn2:
+                # At most 2 concurrent connections for this operation
+                ...
+    """
+
+    def __init__(self, default_budget: int = 4):
+        """
+        Initialize the budget manager.
+
+        Args:
+            default_budget: Default max connections per operation
+        """
+        self.default_budget = default_budget
+        self._operations: dict[str, OperationBudget] = {}
+        self._lock = asyncio.Lock()
+
+    @asynccontextmanager
+    async def operation(
+        self,
+        max_connections: int | None = None,
+        operation_id: str | None = None,
+    ) -> AsyncIterator["BudgetedOperation"]:
+        """
+        Create a budgeted operation context.
+
+        Args:
+            max_connections: Max concurrent connections for this operation.
+                           Defaults to manager's default_budget.
+            operation_id: Optional custom operation ID. Auto-generated if not provided.
+
+        Yields:
+            BudgetedOperation context for acquiring connections
+        """
+        op_id = operation_id or f"op-{uuid.uuid4().hex[:12]}"
+        budget = max_connections or self.default_budget
+
+        async with self._lock:
+            if op_id in self._operations:
+                raise ValueError(f"Operation {op_id} already exists")
+            self._operations[op_id] = OperationBudget(op_id, budget)
+
+        try:
+            yield BudgetedOperation(self, op_id)
+        finally:
+            async with self._lock:
+                self._operations.pop(op_id, None)
+
+    def _get_budget(self, operation_id: str) -> OperationBudget:
+        """Get budget for an operation (internal use)."""
+        budget = self._operations.get(operation_id)
+        if not budget:
+            raise ValueError(f"Operation {operation_id} not found")
+        return budget
+
+
+class BudgetedOperation:
+    """
+    A single operation with connection budget.
+
+    Provides methods to acquire connections within the budget.
+    """
+
+    def __init__(self, manager: ConnectionBudgetManager, operation_id: str):
+        self._manager = manager
+        self.operation_id = operation_id
+
+    @property
+    def budget(self) -> OperationBudget:
+        """Get the budget for this operation."""
+        return self._manager._get_budget(self.operation_id)
+
+    @asynccontextmanager
+    async def acquire(self, pool: "asyncpg.Pool") -> AsyncIterator["asyncpg.Connection"]:
+        """
+        Acquire a connection within the operation's budget.
+
+        Blocks if the operation has reached its connection limit.
+
+        Args:
+            pool: asyncpg connection pool
+
+        Yields:
+            Database connection
+        """
+        budget = self.budget
+        async with budget.semaphore:
+            budget.active_count += 1
+            conn = await pool.acquire()
+            try:
+                yield conn
+            finally:
+                budget.active_count -= 1
+                await pool.release(conn)
+
+    def wrap_pool(self, pool: "asyncpg.Pool") -> "BudgetedPool":
+        """
+        Wrap a pool with this operation's budget.
+
+        The returned BudgetedPool can be passed to functions expecting a pool,
+        and all acquire() calls will be limited by this operation's budget.
+
+        Args:
+            pool: asyncpg connection pool to wrap
+
+        Returns:
+            BudgetedPool that limits connections to this operation's budget
+        """
+        return BudgetedPool(pool, self)
+
+    async def acquire_many(
+        self,
+        pool: "asyncpg.Pool",
+        count: int,
+    ) -> AsyncIterator[list["asyncpg.Connection"]]:
+        """
+        Acquire multiple connections within the budget.
+
+        Note: This acquires connections sequentially to respect the budget.
+        For parallel acquisition, use multiple acquire() calls with asyncio.gather().
+
+        Args:
+            pool: asyncpg connection pool
+            count: Number of connections to acquire
+
+        Yields:
+            List of database connections
+        """
+        connections = []
+        try:
+            for _ in range(count):
+                conn = await pool.acquire()
+                connections.append(conn)
+            yield connections
+        finally:
+            for conn in connections:
+                await pool.release(conn)
+
+
+# Global default manager instance
+_default_manager: ConnectionBudgetManager | None = None
+
+
+def get_budget_manager(default_budget: int = 4) -> ConnectionBudgetManager:
+    """
+    Get or create the global budget manager.
+
+    Args:
+        default_budget: Default max connections per operation
+
+    Returns:
+        Global ConnectionBudgetManager instance
+    """
+    global _default_manager
+    if _default_manager is None:
+        _default_manager = ConnectionBudgetManager(default_budget=default_budget)
+    return _default_manager
+
+
+@asynccontextmanager
+async def budgeted_operation(
+    max_connections: int | None = None,
+    operation_id: str | None = None,
+    default_budget: int = 4,
+) -> AsyncIterator[BudgetedOperation]:
+    """
+    Convenience function to create a budgeted operation.
+
+    Args:
+        max_connections: Max concurrent connections for this operation
+        operation_id: Optional custom operation ID
+        default_budget: Default budget if manager not yet created
+
+    Yields:
+        BudgetedOperation context
+
+    Example:
+        async with budgeted_operation(max_connections=2) as op:
+            async with op.acquire(pool) as conn:
+                await conn.fetch(...)
+    """
+    manager = get_budget_manager(default_budget)
+    async with manager.operation(max_connections, operation_id) as op:
+        yield op
+
+
+class BudgetedPool:
+    """
+    A pool wrapper that limits concurrent connection acquisitions.
+
+    This can be passed to functions expecting a pool, and acquire()
+    calls will be limited by the budget semaphore.
+
+    Usage:
+        async with budgeted_operation(max_connections=4) as op:
+            budgeted_pool = op.wrap_pool(pool)
+            # Pass budgeted_pool to functions that expect a pool
+            await some_function(budgeted_pool, ...)
+    """
+
+    def __init__(self, pool: "asyncpg.Pool", operation: BudgetedOperation):
+        self._pool = pool
+        self._operation = operation
+
+    async def acquire(self) -> "asyncpg.Connection":
+        """
+        Acquire a connection within the budget.
+
+        Note: Caller must release the connection when done.
+        Prefer using as context manager via acquire_with_retry or op.acquire().
+        """
+        budget = self._operation.budget
+        await budget.semaphore.acquire()
+        budget.active_count += 1
+        try:
+            return await self._pool.acquire()
+        except Exception:
+            budget.active_count -= 1
+            budget.semaphore.release()
+            raise
+
+    async def release(self, conn: "asyncpg.Connection") -> None:
+        """Release a connection back to the pool."""
+        budget = self._operation.budget
+        try:
+            await self._pool.release(conn)
+        finally:
+            budget.active_count -= 1
+            budget.semaphore.release()
+
+    def __getattr__(self, name):
+        """Proxy other attributes to the underlying pool."""
+        return getattr(self._pool, name)

hindsight-api/hindsight_api/engine/db_utils.py

@@ -1,9 +1,11 @@
 """
 Database utility functions for connection management with retry logic.
 """
+
 import asyncio
 import logging
 from contextlib import asynccontextmanager
+
 import asyncpg
 
 logger = logging.getLogger(__name__)
@@ -54,16 +56,14 @@ async def retry_with_backoff(
         except retryable_exceptions as e:
             last_exception = e
             if attempt < max_retries:
-                delay = min(base_delay * (2 ** attempt), max_delay)
+                delay = min(base_delay * (2**attempt), max_delay)
                 logger.warning(
                     f"Database operation failed (attempt {attempt + 1}/{max_retries + 1}): {e}. "
                     f"Retrying in {delay:.1f}s..."
                 )
                 await asyncio.sleep(delay)
             else:
-                logger.error(
-                    f"Database operation failed after {max_retries + 1} attempts: {e}"
-                )
+                logger.error(f"Database operation failed after {max_retries + 1} attempts: {e}")
     raise last_exception
 
 
@@ -83,10 +83,22 @@ async def acquire_with_retry(pool: asyncpg.Pool, max_retries: int = DEFAULT_MAX_
     Yields:
         An asyncpg connection
     """
+    import time
+
+    start = time.time()
+
     async def acquire():
         return await pool.acquire()
 
     conn = await retry_with_backoff(acquire, max_retries=max_retries)
+    acquire_time = time.time() - start
+
+    # Log slow connection acquisitions (indicates pool contention)
+    if acquire_time > 0.05:  # 50ms threshold
+        pool_size = pool.get_size()
+        pool_free = pool.get_idle_size()
+        logger.warning(f"[DB POOL] Slow acquire: {acquire_time:.3f}s | size={pool_size}, idle={pool_free}")
+
     try:
         yield conn
     finally:

hindsight-api/hindsight_api/engine/directives/__init__.py

@@ -0,0 +1,5 @@
+"""Directives module for hard rules injected into prompts."""
+
+from .models import Directive
+
+__all__ = ["Directive"]

hindsight-api/hindsight_api/engine/directives/models.py

@@ -0,0 +1,37 @@
+"""Pydantic models for directives."""
+
+from datetime import datetime, timezone
+from uuid import UUID
+
+from pydantic import BaseModel, Field
+
+
+class Directive(BaseModel):
+    """A directive is a hard rule injected into prompts.
+
+    Directives are user-defined rules that guide agent behavior. Unlike mental models
+    which are automatically consolidated from memories, directives are explicit
+    instructions that are always included in relevant prompts.
+
+    Examples:
+    - "Always respond in formal English"
+    - "Never share personal data with third parties"
+    - "Prefer conservative investment recommendations"
+    """
+
+    id: UUID = Field(description="Unique identifier")
+    bank_id: str = Field(description="Bank this directive belongs to")
+    name: str = Field(description="Human-readable name")
+    content: str = Field(description="The directive text to inject into prompts")
+    priority: int = Field(default=0, description="Higher priority directives are injected first")
+    is_active: bool = Field(default=True, description="Whether this directive is currently active")
+    tags: list[str] = Field(default_factory=list, description="Tags for filtering")
+    created_at: datetime = Field(
+        default_factory=lambda: datetime.now(timezone.utc), description="When this directive was created"
+    )
+    updated_at: datetime = Field(
+        default_factory=lambda: datetime.now(timezone.utc), description="When this directive was last updated"
+    )
+
+    class Config:
+        from_attributes = True