chore(deps): Bump actions/setup-python from 5 to 6

Bumps [actions/setup-python](https://github.com/actions/setup-python) from 5 to 6.
- [Release notes](https://github.com/actions/setup-python/releases)
- [Commits](https://github.com/actions/setup-python/compare/v5...v6)

---
updated-dependencies:
- dependency-name: actions/setup-python
  dependency-version: '6'
  dependency-type: direct:production
  update-type: version-update:semver-major
...

Signed-off-by: dependabot[bot] <support@github.com>

.dockerignore

@@ -5,13 +5,42 @@
 !docs/
 
 # Platform - Libs
-!autogpt_platform/autogpt_libs/
+!autogpt_platform/autogpt_libs/autogpt_libs/
+!autogpt_platform/autogpt_libs/pyproject.toml
+!autogpt_platform/autogpt_libs/poetry.lock
+!autogpt_platform/autogpt_libs/README.md
 
 # Platform - Backend
-!autogpt_platform/backend/
+!autogpt_platform/backend/backend/
+!autogpt_platform/backend/test/e2e_test_data.py
+!autogpt_platform/backend/migrations/
+!autogpt_platform/backend/schema.prisma
+!autogpt_platform/backend/pyproject.toml
+!autogpt_platform/backend/poetry.lock
+!autogpt_platform/backend/README.md
+!autogpt_platform/backend/.env
+!autogpt_platform/backend/gen_prisma_types_stub.py
+
+# Platform - Market
+!autogpt_platform/market/market/
+!autogpt_platform/market/scripts.py
+!autogpt_platform/market/schema.prisma
+!autogpt_platform/market/pyproject.toml
+!autogpt_platform/market/poetry.lock
+!autogpt_platform/market/README.md
 
 # Platform - Frontend
-!autogpt_platform/frontend/
+!autogpt_platform/frontend/src/
+!autogpt_platform/frontend/public/
+!autogpt_platform/frontend/scripts/
+!autogpt_platform/frontend/package.json
+!autogpt_platform/frontend/pnpm-lock.yaml
+!autogpt_platform/frontend/tsconfig.json
+!autogpt_platform/frontend/README.md
+## config
+!autogpt_platform/frontend/*.config.*
+!autogpt_platform/frontend/.env.*
+!autogpt_platform/frontend/.env
 
 # Classic - AutoGPT
 !classic/original_autogpt/autogpt/
@@ -35,38 +64,6 @@
 # Classic - Frontend
 !classic/frontend/build/web/
 
-# Explicitly re-ignore unwanted files from whitelisted directories
-# Note: These patterns MUST come after the whitelist rules to take effect
-
-# Hidden files and directories (but keep frontend .env files needed for build)
-**/.*
-!autogpt_platform/frontend/.env
-!autogpt_platform/frontend/.env.default
-!autogpt_platform/frontend/.env.production
-
-# Python artifacts
-**/__pycache__/
-**/*.pyc
-**/*.pyo
-**/.venv/
-**/.ruff_cache/
-**/.pytest_cache/
-**/.coverage
-**/htmlcov/
-
-# Node artifacts
-**/node_modules/
-**/.next/
-**/storybook-static/
-**/playwright-report/
-**/test-results/
-
-# Build artifacts
-**/dist/
-**/build/
-!autogpt_platform/frontend/src/**/build/
-**/target/
-
-# Logs and temp files
-**/*.log
-**/*.tmp
+# Explicitly re-ignore some folders
+.*
+**/__pycache__

.github/workflows/classic-autogpt-ci.yml

@@ -72,7 +72,7 @@ jobs:
           git config --global user.email "github-bot@agpt.co"
 
       - name: Set up Python ${{ matrix.python-version }}
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
         with:
           python-version: ${{ matrix.python-version }}
 

.github/workflows/classic-autogpts-ci.yml

@@ -50,7 +50,7 @@ jobs:
           submodules: true
 
       - name: Set up Python ${{ env.min-python-version }}
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
         with:
           python-version: ${{ env.min-python-version }}
 

.github/workflows/classic-benchmark-ci.yml

@@ -48,7 +48,7 @@ jobs:
           submodules: true
 
       - name: Set up Python ${{ matrix.python-version }}
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
         with:
           python-version: ${{ matrix.python-version }}
 
@@ -120,7 +120,7 @@ jobs:
           submodules: true
 
       - name: Set up Python ${{ env.min-python-version }}
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
         with:
           python-version: ${{ env.min-python-version }}
 

.github/workflows/classic-benchmark_publish_package.yml

@@ -16,7 +16,7 @@ jobs:
         fetch-depth: 0
 
     - name: Set up Python
-      uses: actions/setup-python@v5
+      uses: actions/setup-python@v6
       with:
         python-version: 3.8
 

.github/workflows/classic-forge-ci.yml

@@ -100,7 +100,7 @@ jobs:
           fi
 
       - name: Set up Python ${{ matrix.python-version }}
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
         with:
           python-version: ${{ matrix.python-version }}
 

.github/workflows/classic-python-checks.yml

@@ -73,7 +73,7 @@ jobs:
           fetch-depth: 0
 
       - name: Set up Python ${{ env.min-python-version }}
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
         with:
           python-version: ${{ env.min-python-version }}
 
@@ -125,7 +125,7 @@ jobs:
           fetch-depth: 0
 
       - name: Set up Python ${{ env.min-python-version }}
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
         with:
           python-version: ${{ env.min-python-version }}
 

.github/workflows/claude-ci-failure-auto-fix.yml

@@ -22,7 +22,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Checkout code
-        uses: actions/checkout@v6
+        uses: actions/checkout@v4
         with:
           ref: ${{ github.event.workflow_run.head_branch }}
           fetch-depth: 0
@@ -40,48 +40,6 @@ jobs:
           git checkout -b "$BRANCH_NAME"
           echo "branch_name=$BRANCH_NAME" >> $GITHUB_OUTPUT
 
-      # Backend Python/Poetry setup (so Claude can run linting/tests)
-      - name: Set up Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: "3.11"
-
-      - name: Set up Python dependency cache
-        uses: actions/cache@v5
-        with:
-          path: ~/.cache/pypoetry
-          key: poetry-${{ runner.os }}-${{ hashFiles('autogpt_platform/backend/poetry.lock') }}
-
-      - name: Install Poetry
-        run: |
-          cd autogpt_platform/backend
-          HEAD_POETRY_VERSION=$(python3 ../../.github/workflows/scripts/get_package_version_from_lockfile.py poetry)
-          curl -sSL https://install.python-poetry.org | POETRY_VERSION=$HEAD_POETRY_VERSION python3 -
-          echo "$HOME/.local/bin" >> $GITHUB_PATH
-
-      - name: Install Python dependencies
-        working-directory: autogpt_platform/backend
-        run: poetry install
-
-      - name: Generate Prisma Client
-        working-directory: autogpt_platform/backend
-        run: poetry run prisma generate && poetry run gen-prisma-stub
-
-      # Frontend Node.js/pnpm setup (so Claude can run linting/tests)
-      - name: Enable corepack
-        run: corepack enable
-
-      - name: Set up Node.js
-        uses: actions/setup-node@v6
-        with:
-          node-version: "22"
-          cache: "pnpm"
-          cache-dependency-path: autogpt_platform/frontend/pnpm-lock.yaml
-
-      - name: Install JavaScript dependencies
-        working-directory: autogpt_platform/frontend
-        run: pnpm install --frozen-lockfile
-
       - name: Get CI failure details
         id: failure_details
         uses: actions/github-script@v8

.github/workflows/claude-dependabot.yml

@@ -30,13 +30,13 @@ jobs:
       actions: read # Required for CI access
     steps:
       - name: Checkout code
-        uses: actions/checkout@v6
+        uses: actions/checkout@v4
         with:
           fetch-depth: 1
 
       # Backend Python/Poetry setup (mirrors platform-backend-ci.yml)
       - name: Set up Python
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
         with:
           python-version: "3.11"  # Use standard version matching CI
 
@@ -77,15 +77,27 @@ jobs:
         run: poetry run prisma generate && poetry run gen-prisma-stub
 
       # Frontend Node.js/pnpm setup (mirrors platform-frontend-ci.yml)
-      - name: Enable corepack
-        run: corepack enable
-
       - name: Set up Node.js
         uses: actions/setup-node@v6
         with:
           node-version: "22"
-          cache: "pnpm"
-          cache-dependency-path: autogpt_platform/frontend/pnpm-lock.yaml
+
+      - name: Enable corepack
+        run: corepack enable
+
+      - name: Set pnpm store directory
+        run: |
+          pnpm config set store-dir ~/.pnpm-store
+          echo "PNPM_HOME=$HOME/.pnpm-store" >> $GITHUB_ENV
+
+      - name: Cache frontend dependencies
+        uses: actions/cache@v5
+        with:
+          path: ~/.pnpm-store
+          key: ${{ runner.os }}-pnpm-${{ hashFiles('autogpt_platform/frontend/pnpm-lock.yaml', 'autogpt_platform/frontend/package.json') }}
+          restore-keys: |
+            ${{ runner.os }}-pnpm-${{ hashFiles('autogpt_platform/frontend/pnpm-lock.yaml') }}
+            ${{ runner.os }}-pnpm-
 
       - name: Install JavaScript dependencies
         working-directory: autogpt_platform/frontend

.github/workflows/claude.yml

@@ -40,7 +40,7 @@ jobs:
       actions: read # Required for CI access
     steps:
       - name: Checkout code
-        uses: actions/checkout@v6
+        uses: actions/checkout@v4
         with:
           fetch-depth: 1
 
@@ -52,7 +52,7 @@ jobs:
 
       # Backend Python/Poetry setup (mirrors platform-backend-ci.yml)
       - name: Set up Python
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
         with:
           python-version: "3.11"  # Use standard version matching CI
 
@@ -93,15 +93,27 @@ jobs:
         run: poetry run prisma generate && poetry run gen-prisma-stub
 
       # Frontend Node.js/pnpm setup (mirrors platform-frontend-ci.yml)
-      - name: Enable corepack
-        run: corepack enable
-
       - name: Set up Node.js
         uses: actions/setup-node@v6
         with:
           node-version: "22"
-          cache: "pnpm"
-          cache-dependency-path: autogpt_platform/frontend/pnpm-lock.yaml
+
+      - name: Enable corepack
+        run: corepack enable
+
+      - name: Set pnpm store directory
+        run: |
+          pnpm config set store-dir ~/.pnpm-store
+          echo "PNPM_HOME=$HOME/.pnpm-store" >> $GITHUB_ENV
+
+      - name: Cache frontend dependencies
+        uses: actions/cache@v5
+        with:
+          path: ~/.pnpm-store
+          key: ${{ runner.os }}-pnpm-${{ hashFiles('autogpt_platform/frontend/pnpm-lock.yaml', 'autogpt_platform/frontend/package.json') }}
+          restore-keys: |
+            ${{ runner.os }}-pnpm-${{ hashFiles('autogpt_platform/frontend/pnpm-lock.yaml') }}
+            ${{ runner.os }}-pnpm-
 
       - name: Install JavaScript dependencies
         working-directory: autogpt_platform/frontend

.github/workflows/codeql.yml

@@ -58,11 +58,11 @@ jobs:
         # your codebase is analyzed, see https://docs.github.com/en/code-security/code-scanning/creating-an-advanced-setup-for-code-scanning/codeql-code-scanning-for-compiled-languages
     steps:
     - name: Checkout repository
-      uses: actions/checkout@v6
+      uses: actions/checkout@v4
 
     # Initializes the CodeQL tools for scanning.
     - name: Initialize CodeQL
-      uses: github/codeql-action/init@v4
+      uses: github/codeql-action/init@v3
       with:
         languages: ${{ matrix.language }}
         build-mode: ${{ matrix.build-mode }}
@@ -93,6 +93,6 @@ jobs:
         exit 1
 
     - name: Perform CodeQL Analysis
-      uses: github/codeql-action/analyze@v4
+      uses: github/codeql-action/analyze@v3
       with:
         category: "/language:${{matrix.language}}"

.github/workflows/copilot-setup-steps.yml

@@ -27,14 +27,14 @@ jobs:
     # If you do not check out your code, Copilot will do this for you.
     steps:
       - name: Checkout code
-        uses: actions/checkout@v6
+        uses: actions/checkout@v4
         with:
           fetch-depth: 0
           submodules: true
 
       # Backend Python/Poetry setup (mirrors platform-backend-ci.yml)
       - name: Set up Python
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
         with:
           python-version: "3.11"  # Use standard version matching CI
 

.github/workflows/docs-block-sync.yml

@@ -23,12 +23,12 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v6
+        uses: actions/checkout@v4
         with:
           fetch-depth: 1
 
       - name: Set up Python
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
         with:
           python-version: "3.11"
 

.github/workflows/docs-claude-review.yml

@@ -7,10 +7,6 @@ on:
       - "docs/integrations/**"
       - "autogpt_platform/backend/backend/blocks/**"
 
-concurrency:
-  group: claude-docs-review-${{ github.event.pull_request.number }}
-  cancel-in-progress: true
-
 jobs:
   claude-review:
     # Only run for PRs from members/collaborators
@@ -27,12 +23,12 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v6
+        uses: actions/checkout@v4
         with:
           fetch-depth: 0
 
       - name: Set up Python
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
         with:
           python-version: "3.11"
 
@@ -95,35 +91,5 @@ jobs:
             3. Read corresponding documentation files to verify accuracy
             4. Provide your feedback as a PR comment
 
-            ## IMPORTANT: Comment Marker
-            Start your PR comment with exactly this HTML comment marker on its own line:
-            <!-- CLAUDE_DOCS_REVIEW -->
-
-            This marker is used to identify and replace your comment on subsequent runs.
-
             Be constructive and specific. If everything looks good, say so!
             If there are issues, explain what's wrong and suggest how to fix it.
-
-      - name: Delete old Claude review comments
-        env:
-          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-        run: |
-          # Get all comment IDs with our marker, sorted by creation date (oldest first)
-          COMMENT_IDS=$(gh api \
-            repos/${{ github.repository }}/issues/${{ github.event.pull_request.number }}/comments \
-            --jq '[.[] | select(.body | contains("<!-- CLAUDE_DOCS_REVIEW -->"))] | sort_by(.created_at) | .[].id')
-
-          # Count comments
-          COMMENT_COUNT=$(echo "$COMMENT_IDS" | grep -c . || true)
-
-          if [ "$COMMENT_COUNT" -gt 1 ]; then
-            # Delete all but the last (newest) comment
-            echo "$COMMENT_IDS" | head -n -1 | while read -r COMMENT_ID; do
-              if [ -n "$COMMENT_ID" ]; then
-                echo "Deleting old review comment: $COMMENT_ID"
-                gh api -X DELETE repos/${{ github.repository }}/issues/comments/$COMMENT_ID
-              fi
-            done
-          else
-            echo "No old review comments to clean up"
-          fi

.github/workflows/docs-enhance.yml

@@ -28,12 +28,12 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v6
+        uses: actions/checkout@v4
         with:
           fetch-depth: 1
 
       - name: Set up Python
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
         with:
           python-version: "3.11"
 

.github/workflows/platform-autogpt-deploy-dev.yaml

@@ -25,12 +25,12 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v6
+        uses: actions/checkout@v4
         with:
           ref: ${{ github.event.inputs.git_ref || github.ref_name }}
 
       - name: Set up Python
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
         with:
           python-version: '3.11'
 
@@ -52,7 +52,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Trigger deploy workflow
-        uses: peter-evans/repository-dispatch@v4
+        uses: peter-evans/repository-dispatch@v3
         with:
           token: ${{ secrets.DEPLOY_TOKEN }}
           repository: Significant-Gravitas/AutoGPT_cloud_infrastructure

.github/workflows/platform-autogpt-deploy-prod.yml

@@ -17,12 +17,12 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v6
+        uses: actions/checkout@v4
         with:
           ref: ${{ github.ref_name || 'master' }}
 
       - name: Set up Python
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
         with:
           python-version: '3.11'
 
@@ -45,7 +45,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Trigger deploy workflow
-        uses: peter-evans/repository-dispatch@v4
+        uses: peter-evans/repository-dispatch@v3
         with:
           token: ${{ secrets.DEPLOY_TOKEN }}
           repository: Significant-Gravitas/AutoGPT_cloud_infrastructure

.github/workflows/platform-backend-ci.yml

@@ -41,18 +41,13 @@ jobs:
         ports:
           - 6379:6379
       rabbitmq:
-        image: rabbitmq:4.1.4
+        image: rabbitmq:3.12-management
         ports:
           - 5672:5672
+          - 15672:15672
         env:
           RABBITMQ_DEFAULT_USER: ${{ env.RABBITMQ_DEFAULT_USER }}
           RABBITMQ_DEFAULT_PASS: ${{ env.RABBITMQ_DEFAULT_PASS }}
-        options: >-
-          --health-cmd "rabbitmq-diagnostics -q ping"
-          --health-interval 30s
-          --health-timeout 10s
-          --health-retries 5
-          --health-start-period 10s
       clamav:
         image: clamav/clamav-debian:latest
         ports:
@@ -73,13 +68,13 @@ jobs:
 
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v6
+        uses: actions/checkout@v4
         with:
           fetch-depth: 0
           submodules: true
 
       - name: Set up Python ${{ matrix.python-version }}
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
         with:
           python-version: ${{ matrix.python-version }}
 

.github/workflows/platform-dev-deploy-event-dispatcher.yml

@@ -82,7 +82,7 @@ jobs:
           
       - name: Dispatch Deploy Event
         if: steps.check_status.outputs.should_deploy == 'true'
-        uses: peter-evans/repository-dispatch@v4
+        uses: peter-evans/repository-dispatch@v3
         with:
           token: ${{ secrets.DISPATCH_TOKEN }}
           repository: Significant-Gravitas/AutoGPT_cloud_infrastructure
@@ -110,7 +110,7 @@ jobs:
 
       - name: Dispatch Undeploy Event (from comment)
         if: steps.check_status.outputs.should_undeploy == 'true'
-        uses: peter-evans/repository-dispatch@v4
+        uses: peter-evans/repository-dispatch@v3
         with:
           token: ${{ secrets.DISPATCH_TOKEN }}
           repository: Significant-Gravitas/AutoGPT_cloud_infrastructure
@@ -168,7 +168,7 @@ jobs:
           github.event_name == 'pull_request' &&
           github.event.action == 'closed' &&
           steps.check_pr_close.outputs.should_undeploy == 'true'
-        uses: peter-evans/repository-dispatch@v4
+        uses: peter-evans/repository-dispatch@v3
         with:
           token: ${{ secrets.DISPATCH_TOKEN }}
           repository: Significant-Gravitas/AutoGPT_cloud_infrastructure

.github/workflows/platform-frontend-ci.yml

@@ -6,16 +6,10 @@ on:
     paths:
       - ".github/workflows/platform-frontend-ci.yml"
       - "autogpt_platform/frontend/**"
-      - "autogpt_platform/backend/Dockerfile"
-      - "autogpt_platform/docker-compose.yml"
-      - "autogpt_platform/docker-compose.platform.yml"
   pull_request:
     paths:
       - ".github/workflows/platform-frontend-ci.yml"
       - "autogpt_platform/frontend/**"
-      - "autogpt_platform/backend/Dockerfile"
-      - "autogpt_platform/docker-compose.yml"
-      - "autogpt_platform/docker-compose.platform.yml"
   merge_group:
   workflow_dispatch:
 
@@ -32,11 +26,12 @@ jobs:
   setup:
     runs-on: ubuntu-latest
     outputs:
+      cache-key: ${{ steps.cache-key.outputs.key }}
       components-changed: ${{ steps.filter.outputs.components }}
 
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v6
+        uses: actions/checkout@v4
 
       - name: Check for component changes
         uses: dorny/paths-filter@v3
@@ -46,17 +41,28 @@ jobs:
             components:
               - 'autogpt_platform/frontend/src/components/**'
 
-      - name: Enable corepack
-        run: corepack enable
-
-      - name: Set up Node
+      - name: Set up Node.js
         uses: actions/setup-node@v6
         with:
           node-version: "22.18.0"
-          cache: "pnpm"
-          cache-dependency-path: autogpt_platform/frontend/pnpm-lock.yaml
 
-      - name: Install dependencies to populate cache
+      - name: Enable corepack
+        run: corepack enable
+
+      - name: Generate cache key
+        id: cache-key
+        run: echo "key=${{ runner.os }}-pnpm-${{ hashFiles('autogpt_platform/frontend/pnpm-lock.yaml', 'autogpt_platform/frontend/package.json') }}" >> $GITHUB_OUTPUT
+
+      - name: Cache dependencies
+        uses: actions/cache@v5
+        with:
+          path: ~/.pnpm-store
+          key: ${{ steps.cache-key.outputs.key }}
+          restore-keys: |
+            ${{ runner.os }}-pnpm-${{ hashFiles('autogpt_platform/frontend/pnpm-lock.yaml') }}
+            ${{ runner.os }}-pnpm-
+
+      - name: Install dependencies
         run: pnpm install --frozen-lockfile
 
   lint:
@@ -65,17 +71,24 @@ jobs:
 
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v6
+        uses: actions/checkout@v4
+
+      - name: Set up Node.js
+        uses: actions/setup-node@v6
+        with:
+          node-version: "22.18.0"
 
       - name: Enable corepack
         run: corepack enable
 
-      - name: Set up Node
-        uses: actions/setup-node@v6
+      - name: Restore dependencies cache
+        uses: actions/cache@v5
         with:
-          node-version: "22.18.0"
-          cache: "pnpm"
-          cache-dependency-path: autogpt_platform/frontend/pnpm-lock.yaml
+          path: ~/.pnpm-store
+          key: ${{ needs.setup.outputs.cache-key }}
+          restore-keys: |
+            ${{ runner.os }}-pnpm-${{ hashFiles('autogpt_platform/frontend/pnpm-lock.yaml') }}
+            ${{ runner.os }}-pnpm-
 
       - name: Install dependencies
         run: pnpm install --frozen-lockfile
@@ -94,19 +107,26 @@ jobs:
 
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v6
+        uses: actions/checkout@v4
         with:
           fetch-depth: 0
 
+      - name: Set up Node.js
+        uses: actions/setup-node@v6
+        with:
+          node-version: "22.18.0"
+
       - name: Enable corepack
         run: corepack enable
 
-      - name: Set up Node
-        uses: actions/setup-node@v6
+      - name: Restore dependencies cache
+        uses: actions/cache@v5
         with:
-          node-version: "22.18.0"
-          cache: "pnpm"
-          cache-dependency-path: autogpt_platform/frontend/pnpm-lock.yaml
+          path: ~/.pnpm-store
+          key: ${{ needs.setup.outputs.cache-key }}
+          restore-keys: |
+            ${{ runner.os }}-pnpm-${{ hashFiles('autogpt_platform/frontend/pnpm-lock.yaml') }}
+            ${{ runner.os }}-pnpm-
 
       - name: Install dependencies
         run: pnpm install --frozen-lockfile
@@ -121,20 +141,30 @@ jobs:
           exitOnceUploaded: true
 
   e2e_test:
-    name: end-to-end tests
     runs-on: big-boi
+    needs: setup
+    strategy:
+      fail-fast: false
 
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v6
+        uses: actions/checkout@v4
         with:
           submodules: recursive
 
-      - name: Set up Platform - Copy default supabase .env
+      - name: Set up Node.js
+        uses: actions/setup-node@v6
+        with:
+          node-version: "22.18.0"
+
+      - name: Enable corepack
+        run: corepack enable
+
+      - name: Copy default supabase .env
         run: |
           cp ../.env.default ../.env
 
-      - name: Set up Platform - Copy backend .env and set OpenAI API key
+      - name: Copy backend .env and set OpenAI API key
         run: |
           cp ../backend/.env.default ../backend/.env
           echo "OPENAI_INTERNAL_API_KEY=${{ secrets.OPENAI_API_KEY }}" >> ../backend/.env
@@ -142,125 +172,77 @@ jobs:
           # Used by E2E test data script to generate embeddings for approved store agents
           OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
 
-      - name: Set up Platform - Set up Docker Buildx
+      - name: Set up Docker Buildx
         uses: docker/setup-buildx-action@v3
-        with:
-          driver: docker-container
-          driver-opts: network=host
 
-      - name: Set up Platform - Expose GHA cache to docker buildx CLI
-        uses: crazy-max/ghaction-github-runtime@v3
-
-      - name: Set up Platform - Build Docker images (with cache)
-        working-directory: autogpt_platform
-        run: |
-          pip install pyyaml
-
-          # Resolve extends and generate a flat compose file that bake can understand
-          docker compose -f docker-compose.yml config > docker-compose.resolved.yml
-
-          # Add cache configuration to the resolved compose file
-          python ../.github/workflows/scripts/docker-ci-fix-compose-build-cache.py \
-            --source docker-compose.resolved.yml \
-            --cache-from "type=gha" \
-            --cache-to "type=gha,mode=max" \
-            --backend-hash "${{ hashFiles('autogpt_platform/backend/Dockerfile', 'autogpt_platform/backend/poetry.lock', 'autogpt_platform/backend/backend') }}" \
-            --frontend-hash "${{ hashFiles('autogpt_platform/frontend/Dockerfile', 'autogpt_platform/frontend/pnpm-lock.yaml', 'autogpt_platform/frontend/src') }}" \
-            --git-ref "${{ github.ref }}"
-
-          # Build with bake using the resolved compose file (now includes cache config)
-          docker buildx bake --allow=fs.read=.. -f docker-compose.resolved.yml --load
-        env:
-          NEXT_PUBLIC_PW_TEST: true
-
-      - name: Set up tests - Cache E2E test data
-        id: e2e-data-cache
+      - name: Cache Docker layers
         uses: actions/cache@v5
         with:
-          path: /tmp/e2e_test_data.sql
-          key: e2e-test-data-${{ hashFiles('autogpt_platform/backend/test/e2e_test_data.py', 'autogpt_platform/backend/migrations/**', '.github/workflows/platform-frontend-ci.yml') }}
+          path: /tmp/.buildx-cache
+          key: ${{ runner.os }}-buildx-frontend-test-${{ hashFiles('autogpt_platform/docker-compose.yml', 'autogpt_platform/backend/Dockerfile', 'autogpt_platform/backend/pyproject.toml', 'autogpt_platform/backend/poetry.lock') }}
+          restore-keys: |
+            ${{ runner.os }}-buildx-frontend-test-
 
-      - name: Set up Platform - Start Supabase DB + Auth
+      - name: Run docker compose
         run: |
-          docker compose -f ../docker-compose.resolved.yml up -d db auth --no-build
-          echo "Waiting for database to be ready..."
-          timeout 60 sh -c 'until docker compose -f ../docker-compose.resolved.yml exec -T db pg_isready -U postgres 2>/dev/null; do sleep 2; done'
-          echo "Waiting for auth service to be ready..."
-          timeout 60 sh -c 'until docker compose -f ../docker-compose.resolved.yml exec -T db psql -U postgres -d postgres -c "SELECT 1 FROM auth.users LIMIT 1" 2>/dev/null; do sleep 2; done' || echo "Auth schema check timeout, continuing..."
-
-      - name: Set up Platform - Run migrations
-        run: |
-          echo "Running migrations..."
-          docker compose -f ../docker-compose.resolved.yml run --rm migrate
-          echo "✅ Migrations completed"
+          NEXT_PUBLIC_PW_TEST=true docker compose -f ../docker-compose.yml up -d
         env:
-          NEXT_PUBLIC_PW_TEST: true
+          DOCKER_BUILDKIT: 1
+          BUILDX_CACHE_FROM: type=local,src=/tmp/.buildx-cache
+          BUILDX_CACHE_TO: type=local,dest=/tmp/.buildx-cache-new,mode=max
 
-      - name: Set up tests - Load cached E2E test data
-        if: steps.e2e-data-cache.outputs.cache-hit == 'true'
+      - name: Move cache
         run: |
-          echo "✅ Found cached E2E test data, restoring..."
-          {
-            echo "SET session_replication_role = 'replica';"
-            cat /tmp/e2e_test_data.sql
-            echo "SET session_replication_role = 'origin';"
-          } | docker compose -f ../docker-compose.resolved.yml exec -T db psql -U postgres -d postgres -b
-          # Refresh materialized views after restore
-          docker compose -f ../docker-compose.resolved.yml exec -T db \
-            psql -U postgres -d postgres -b -c "SET search_path TO platform; SELECT refresh_store_materialized_views();" || true
+          rm -rf /tmp/.buildx-cache
+          if [ -d "/tmp/.buildx-cache-new" ]; then
+            mv /tmp/.buildx-cache-new /tmp/.buildx-cache
+          fi
 
-          echo "✅ E2E test data restored from cache"
-
-      - name: Set up Platform - Start (all other services)
+      - name: Wait for services to be ready
         run: |
-          docker compose -f ../docker-compose.resolved.yml up -d --no-build
           echo "Waiting for rest_server to be ready..."
           timeout 60 sh -c 'until curl -f http://localhost:8006/health 2>/dev/null; do sleep 2; done' || echo "Rest server health check timeout, continuing..."
-        env:
-          NEXT_PUBLIC_PW_TEST: true
+          echo "Waiting for database to be ready..."
+          timeout 60 sh -c 'until docker compose -f ../docker-compose.yml exec -T db pg_isready -U postgres 2>/dev/null; do sleep 2; done' || echo "Database ready check timeout, continuing..."
 
-      - name: Set up tests - Create E2E test data
-        if: steps.e2e-data-cache.outputs.cache-hit != 'true'
+      - name: Create E2E test data
         run: |
           echo "Creating E2E test data..."
-          docker cp ../backend/test/e2e_test_data.py $(docker compose -f ../docker-compose.resolved.yml ps -q rest_server):/tmp/e2e_test_data.py
-          docker compose -f ../docker-compose.resolved.yml exec -T rest_server sh -c "cd /app/autogpt_platform && python /tmp/e2e_test_data.py" || {
-            echo "❌ E2E test data creation failed!"
-            docker compose -f ../docker-compose.resolved.yml logs --tail=50 rest_server
-            exit 1
-          }
+          # First try to run the script from inside the container
+          if docker compose -f ../docker-compose.yml exec -T rest_server test -f /app/autogpt_platform/backend/test/e2e_test_data.py; then
+            echo "✅ Found e2e_test_data.py in container, running it..."
+            docker compose -f ../docker-compose.yml exec -T rest_server sh -c "cd /app/autogpt_platform && python backend/test/e2e_test_data.py" || {
+              echo "❌ E2E test data creation failed!"
+              docker compose -f ../docker-compose.yml logs --tail=50 rest_server
+              exit 1
+            }
+          else
+            echo "⚠️ e2e_test_data.py not found in container, copying and running..."
+            # Copy the script into the container and run it
+            docker cp ../backend/test/e2e_test_data.py $(docker compose -f ../docker-compose.yml ps -q rest_server):/tmp/e2e_test_data.py || {
+              echo "❌ Failed to copy script to container"
+              exit 1
+            }
+            docker compose -f ../docker-compose.yml exec -T rest_server sh -c "cd /app/autogpt_platform && python /tmp/e2e_test_data.py" || {
+              echo "❌ E2E test data creation failed!"
+              docker compose -f ../docker-compose.yml logs --tail=50 rest_server
+              exit 1
+            }
+          fi
 
-          # Dump auth.users + platform schema for cache (two separate dumps)
-          echo "Dumping database for cache..."
-          {
-            docker compose -f ../docker-compose.resolved.yml exec -T db \
-              pg_dump -U postgres --data-only --column-inserts \
-              --table='auth.users' postgres
-            docker compose -f ../docker-compose.resolved.yml exec -T db \
-              pg_dump -U postgres --data-only --column-inserts \
-              --schema=platform \
-              --exclude-table='platform._prisma_migrations' \
-              --exclude-table='platform.apscheduler_jobs' \
-              --exclude-table='platform.apscheduler_jobs_batched_notifications' \
-              postgres
-          } > /tmp/e2e_test_data.sql
-
-          echo "✅ Database dump created for caching ($(wc -l < /tmp/e2e_test_data.sql) lines)"
-
-      - name: Set up tests - Enable corepack
-        run: corepack enable
-
-      - name: Set up tests - Set up Node
-        uses: actions/setup-node@v6
+      - name: Restore dependencies cache
+        uses: actions/cache@v5
         with:
-          node-version: "22.18.0"
-          cache: "pnpm"
-          cache-dependency-path: autogpt_platform/frontend/pnpm-lock.yaml
+          path: ~/.pnpm-store
+          key: ${{ needs.setup.outputs.cache-key }}
+          restore-keys: |
+            ${{ runner.os }}-pnpm-${{ hashFiles('autogpt_platform/frontend/pnpm-lock.yaml') }}
+            ${{ runner.os }}-pnpm-
 
-      - name: Set up tests - Install dependencies
+      - name: Install dependencies
         run: pnpm install --frozen-lockfile
 
-      - name: Set up tests - Install browser 'chromium'
+      - name: Install Browser 'chromium'
         run: pnpm playwright install --with-deps chromium
 
       - name: Run Playwright tests
@@ -287,7 +269,7 @@ jobs:
 
       - name: Print Final Docker Compose logs
         if: always()
-        run: docker compose -f ../docker-compose.resolved.yml logs
+        run: docker compose -f ../docker-compose.yml logs
 
   integration_test:
     runs-on: ubuntu-latest
@@ -295,19 +277,26 @@ jobs:
 
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v6
+        uses: actions/checkout@v4
         with:
           submodules: recursive
 
+      - name: Set up Node.js
+        uses: actions/setup-node@v6
+        with:
+          node-version: "22.18.0"
+
       - name: Enable corepack
         run: corepack enable
 
-      - name: Set up Node
-        uses: actions/setup-node@v6
+      - name: Restore dependencies cache
+        uses: actions/cache@v5
         with:
-          node-version: "22.18.0"
-          cache: "pnpm"
-          cache-dependency-path: autogpt_platform/frontend/pnpm-lock.yaml
+          path: ~/.pnpm-store
+          key: ${{ needs.setup.outputs.cache-key }}
+          restore-keys: |
+            ${{ runner.os }}-pnpm-${{ hashFiles('autogpt_platform/frontend/pnpm-lock.yaml') }}
+            ${{ runner.os }}-pnpm-
 
       - name: Install dependencies
         run: pnpm install --frozen-lockfile

.github/workflows/platform-fullstack-ci.yml

@@ -29,7 +29,7 @@ jobs:
 
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v6
+        uses: actions/checkout@v4
 
       - name: Set up Node.js
         uses: actions/setup-node@v6
@@ -63,7 +63,7 @@ jobs:
 
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v6
+        uses: actions/checkout@v4
         with:
           submodules: recursive
 

.github/workflows/pr-overlap-check.yml

@@ -1,39 +0,0 @@
-name: PR Overlap Detection
-
-on:
-  pull_request:
-    types: [opened, synchronize, reopened]
-    branches:
-      - dev
-      - master
-
-permissions:
-  contents: read
-  pull-requests: write
-
-jobs:
-  check-overlaps:
-    runs-on: ubuntu-latest
-    steps:
-      - name: Checkout repository
-        uses: actions/checkout@v4
-        with:
-          fetch-depth: 0  # Need full history for merge testing
-
-      - name: Set up Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: '3.11'
-
-      - name: Configure git
-        run: |
-          git config user.email "github-actions[bot]@users.noreply.github.com"
-          git config user.name "github-actions[bot]"
-
-      - name: Run overlap detection
-        env:
-          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-        # Always succeed - this check informs contributors, it shouldn't block merging
-        continue-on-error: true
-        run: |
-          python .github/scripts/detect_overlaps.py ${{ github.event.pull_request.number }}

.github/workflows/repo-workflow-checker.yml

@@ -11,11 +11,11 @@ jobs:
     steps:
       # - name: Wait some time for all actions to start
       #   run: sleep 30
-      - uses: actions/checkout@v6
+      - uses: actions/checkout@v4
         # with:
           # fetch-depth: 0
       - name: Set up Python
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
         with:
           python-version: "3.10"
       - name: Install dependencies

.github/workflows/scripts/docker-ci-fix-compose-build-cache.py

@@ -1,195 +0,0 @@
-#!/usr/bin/env python3
-"""
-Add cache configuration to a resolved docker-compose file for all services
-that have a build key, and ensure image names match what docker compose expects.
-"""
-
-import argparse
-
-import yaml
-
-
-DEFAULT_BRANCH = "dev"
-CACHE_BUILDS_FOR_COMPONENTS = ["backend", "frontend"]
-
-
-def main():
-    parser = argparse.ArgumentParser(
-        description="Add cache config to a resolved compose file"
-    )
-    parser.add_argument(
-        "--source",
-        required=True,
-        help="Source compose file to read (should be output of `docker compose config`)",
-    )
-    parser.add_argument(
-        "--cache-from",
-        default="type=gha",
-        help="Cache source configuration",
-    )
-    parser.add_argument(
-        "--cache-to",
-        default="type=gha,mode=max",
-        help="Cache destination configuration",
-    )
-    for component in CACHE_BUILDS_FOR_COMPONENTS:
-        parser.add_argument(
-            f"--{component}-hash",
-            default="",
-            help=f"Hash for {component} cache scope (e.g., from hashFiles())",
-        )
-    parser.add_argument(
-        "--git-ref",
-        default="",
-        help="Git ref for branch-based cache scope (e.g., refs/heads/master)",
-    )
-    args = parser.parse_args()
-
-    # Normalize git ref to a safe scope name (e.g., refs/heads/master -> master)
-    git_ref_scope = ""
-    if args.git_ref:
-        git_ref_scope = args.git_ref.replace("refs/heads/", "").replace("/", "-")
-
-    with open(args.source, "r") as f:
-        compose = yaml.safe_load(f)
-
-    # Get project name from compose file or default
-    project_name = compose.get("name", "autogpt_platform")
-
-    def get_image_name(dockerfile: str, target: str) -> str:
-        """Generate image name based on Dockerfile folder and build target."""
-        dockerfile_parts = dockerfile.replace("\\", "/").split("/")
-        if len(dockerfile_parts) >= 2:
-            folder_name = dockerfile_parts[-2]  # e.g., "backend" or "frontend"
-        else:
-            folder_name = "app"
-        return f"{project_name}-{folder_name}:{target}"
-
-    def get_build_key(dockerfile: str, target: str) -> str:
-        """Generate a unique key for a Dockerfile+target combination."""
-        return f"{dockerfile}:{target}"
-
-    def get_component(dockerfile: str) -> str | None:
-        """Get component name (frontend/backend) from dockerfile path."""
-        for component in CACHE_BUILDS_FOR_COMPONENTS:
-            if component in dockerfile:
-                return component
-        return None
-
-    # First pass: collect all services with build configs and identify duplicates
-    # Track which (dockerfile, target) combinations we've seen
-    build_key_to_first_service: dict[str, str] = {}
-    services_to_build: list[str] = []
-    services_to_dedupe: list[str] = []
-
-    for service_name, service_config in compose.get("services", {}).items():
-        if "build" not in service_config:
-            continue
-
-        build_config = service_config["build"]
-        dockerfile = build_config.get("dockerfile", "Dockerfile")
-        target = build_config.get("target", "default")
-        build_key = get_build_key(dockerfile, target)
-
-        if build_key not in build_key_to_first_service:
-            # First service with this build config - it will do the actual build
-            build_key_to_first_service[build_key] = service_name
-            services_to_build.append(service_name)
-        else:
-            # Duplicate - will just use the image from the first service
-            services_to_dedupe.append(service_name)
-
-    # Second pass: configure builds and deduplicate
-    modified_services = []
-    for service_name, service_config in compose.get("services", {}).items():
-        if "build" not in service_config:
-            continue
-
-        build_config = service_config["build"]
-        dockerfile = build_config.get("dockerfile", "Dockerfile")
-        target = build_config.get("target", "latest")
-        image_name = get_image_name(dockerfile, target)
-
-        # Set image name for all services (needed for both builders and deduped)
-        service_config["image"] = image_name
-
-        if service_name in services_to_dedupe:
-            # Remove build config - this service will use the pre-built image
-            del service_config["build"]
-            continue
-
-        # This service will do the actual build - add cache config
-        cache_from_list = []
-        cache_to_list = []
-
-        component = get_component(dockerfile)
-        if not component:
-            # Skip services that don't clearly match frontend/backend
-            continue
-
-        # Get the hash for this component
-        component_hash = getattr(args, f"{component}_hash")
-
-        # Scope format: platform-{component}-{target}-{hash|ref}
-        # Example: platform-backend-server-abc123
-
-        if "type=gha" in args.cache_from:
-            # 1. Primary: exact hash match (most specific)
-            if component_hash:
-                hash_scope = f"platform-{component}-{target}-{component_hash}"
-                cache_from_list.append(f"{args.cache_from},scope={hash_scope}")
-
-            # 2. Fallback: branch-based cache
-            if git_ref_scope:
-                ref_scope = f"platform-{component}-{target}-{git_ref_scope}"
-                cache_from_list.append(f"{args.cache_from},scope={ref_scope}")
-
-            # 3. Fallback: dev branch cache (for PRs/feature branches)
-            if git_ref_scope and git_ref_scope != DEFAULT_BRANCH:
-                master_scope = f"platform-{component}-{target}-{DEFAULT_BRANCH}"
-                cache_from_list.append(f"{args.cache_from},scope={master_scope}")
-
-        if "type=gha" in args.cache_to:
-            # Write to both hash-based and branch-based scopes
-            if component_hash:
-                hash_scope = f"platform-{component}-{target}-{component_hash}"
-                cache_to_list.append(f"{args.cache_to},scope={hash_scope}")
-
-            if git_ref_scope:
-                ref_scope = f"platform-{component}-{target}-{git_ref_scope}"
-                cache_to_list.append(f"{args.cache_to},scope={ref_scope}")
-
-        # Ensure we have at least one cache source/target
-        if not cache_from_list:
-            cache_from_list.append(args.cache_from)
-        if not cache_to_list:
-            cache_to_list.append(args.cache_to)
-
-        build_config["cache_from"] = cache_from_list
-        build_config["cache_to"] = cache_to_list
-        modified_services.append(service_name)
-
-    # Write back to the same file
-    with open(args.source, "w") as f:
-        yaml.dump(compose, f, default_flow_style=False, sort_keys=False)
-
-    print(f"Added cache config to {len(modified_services)} services in {args.source}:")
-    for svc in modified_services:
-        svc_config = compose["services"][svc]
-        build_cfg = svc_config.get("build", {})
-        cache_from_list = build_cfg.get("cache_from", ["none"])
-        cache_to_list = build_cfg.get("cache_to", ["none"])
-        print(f"  - {svc}")
-        print(f"      image: {svc_config.get('image', 'N/A')}")
-        print(f"      cache_from: {cache_from_list}")
-        print(f"      cache_to: {cache_to_list}")
-    if services_to_dedupe:
-        print(
-            f"Deduplicated {len(services_to_dedupe)} services (will use pre-built images):"
-        )
-        for svc in services_to_dedupe:
-            print(f"  - {svc} -> {compose['services'][svc].get('image', 'N/A')}")
-
-
-if __name__ == "__main__":
-    main()

autogpt_platform/CLAUDE.md

@@ -45,11 +45,6 @@ AutoGPT Platform is a monorepo containing:
 - Backend/Frontend services use YAML anchors for consistent configuration
 - Supabase services (`db/docker/docker-compose.yml`) follow the same pattern
 
-### Branching Strategy
-
-- **`dev`** is the main development branch. All PRs should target `dev`.
-- **`master`** is the production branch. Only used for production releases.
-
 ### Creating Pull Requests
 
 - Create the PR against the `dev` branch of the repository.

autogpt_platform/autogpt_libs/poetry.lock

@@ -448,61 +448,61 @@ toml = ["tomli ; python_full_version <= \"3.11.0a6\""]
 
 [[package]]
 name = "cryptography"
-version = "46.0.5"
+version = "46.0.4"
 description = "cryptography is a package which provides cryptographic recipes and primitives to Python developers."
 optional = false
 python-versions = "!=3.9.0,!=3.9.1,>=3.8"
 groups = ["main"]
 files = [
-    {file = "cryptography-46.0.5-cp311-abi3-macosx_10_9_universal2.whl", hash = "sha256:351695ada9ea9618b3500b490ad54c739860883df6c1f555e088eaf25b1bbaad"},
-    {file = "cryptography-46.0.5-cp311-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:c18ff11e86df2e28854939acde2d003f7984f721eba450b56a200ad90eeb0e6b"},
-    {file = "cryptography-46.0.5-cp311-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:4d7e3d356b8cd4ea5aff04f129d5f66ebdc7b6f8eae802b93739ed520c47c79b"},
-    {file = "cryptography-46.0.5-cp311-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:50bfb6925eff619c9c023b967d5b77a54e04256c4281b0e21336a130cd7fc263"},
-    {file = "cryptography-46.0.5-cp311-abi3-manylinux_2_28_ppc64le.whl", hash = "sha256:803812e111e75d1aa73690d2facc295eaefd4439be1023fefc4995eaea2af90d"},
-    {file = "cryptography-46.0.5-cp311-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:3ee190460e2fbe447175cda91b88b84ae8322a104fc27766ad09428754a618ed"},
-    {file = "cryptography-46.0.5-cp311-abi3-manylinux_2_31_armv7l.whl", hash = "sha256:f145bba11b878005c496e93e257c1e88f154d278d2638e6450d17e0f31e558d2"},
-    {file = "cryptography-46.0.5-cp311-abi3-manylinux_2_34_aarch64.whl", hash = "sha256:e9251e3be159d1020c4030bd2e5f84d6a43fe54b6c19c12f51cde9542a2817b2"},
-    {file = "cryptography-46.0.5-cp311-abi3-manylinux_2_34_ppc64le.whl", hash = "sha256:47fb8a66058b80e509c47118ef8a75d14c455e81ac369050f20ba0d23e77fee0"},
-    {file = "cryptography-46.0.5-cp311-abi3-manylinux_2_34_x86_64.whl", hash = "sha256:4c3341037c136030cb46e4b1e17b7418ea4cbd9dd207e4a6f3b2b24e0d4ac731"},
-    {file = "cryptography-46.0.5-cp311-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:890bcb4abd5a2d3f852196437129eb3667d62630333aacc13dfd470fad3aaa82"},
-    {file = "cryptography-46.0.5-cp311-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:80a8d7bfdf38f87ca30a5391c0c9ce4ed2926918e017c29ddf643d0ed2778ea1"},
-    {file = "cryptography-46.0.5-cp311-abi3-win32.whl", hash = "sha256:60ee7e19e95104d4c03871d7d7dfb3d22ef8a9b9c6778c94e1c8fcc8365afd48"},
-    {file = "cryptography-46.0.5-cp311-abi3-win_amd64.whl", hash = "sha256:38946c54b16c885c72c4f59846be9743d699eee2b69b6988e0a00a01f46a61a4"},
-    {file = "cryptography-46.0.5-cp314-cp314t-macosx_10_9_universal2.whl", hash = "sha256:94a76daa32eb78d61339aff7952ea819b1734b46f73646a07decb40e5b3448e2"},
-    {file = "cryptography-46.0.5-cp314-cp314t-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:5be7bf2fb40769e05739dd0046e7b26f9d4670badc7b032d6ce4db64dddc0678"},
-    {file = "cryptography-46.0.5-cp314-cp314t-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:fe346b143ff9685e40192a4960938545c699054ba11d4f9029f94751e3f71d87"},
-    {file = "cryptography-46.0.5-cp314-cp314t-manylinux_2_28_aarch64.whl", hash = "sha256:c69fd885df7d089548a42d5ec05be26050ebcd2283d89b3d30676eb32ff87dee"},
-    {file = "cryptography-46.0.5-cp314-cp314t-manylinux_2_28_ppc64le.whl", hash = "sha256:8293f3dea7fc929ef7240796ba231413afa7b68ce38fd21da2995549f5961981"},
-    {file = "cryptography-46.0.5-cp314-cp314t-manylinux_2_28_x86_64.whl", hash = "sha256:1abfdb89b41c3be0365328a410baa9df3ff8a9110fb75e7b52e66803ddabc9a9"},
-    {file = "cryptography-46.0.5-cp314-cp314t-manylinux_2_31_armv7l.whl", hash = "sha256:d66e421495fdb797610a08f43b05269e0a5ea7f5e652a89bfd5a7d3c1dee3648"},
-    {file = "cryptography-46.0.5-cp314-cp314t-manylinux_2_34_aarch64.whl", hash = "sha256:4e817a8920bfbcff8940ecfd60f23d01836408242b30f1a708d93198393a80b4"},
-    {file = "cryptography-46.0.5-cp314-cp314t-manylinux_2_34_ppc64le.whl", hash = "sha256:68f68d13f2e1cb95163fa3b4db4bf9a159a418f5f6e7242564fc75fcae667fd0"},
-    {file = "cryptography-46.0.5-cp314-cp314t-manylinux_2_34_x86_64.whl", hash = "sha256:a3d1fae9863299076f05cb8a778c467578262fae09f9dc0ee9b12eb4268ce663"},
-    {file = "cryptography-46.0.5-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:c4143987a42a2397f2fc3b4d7e3a7d313fbe684f67ff443999e803dd75a76826"},
-    {file = "cryptography-46.0.5-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:7d731d4b107030987fd61a7f8ab512b25b53cef8f233a97379ede116f30eb67d"},
-    {file = "cryptography-46.0.5-cp314-cp314t-win32.whl", hash = "sha256:c3bcce8521d785d510b2aad26ae2c966092b7daa8f45dd8f44734a104dc0bc1a"},
-    {file = "cryptography-46.0.5-cp314-cp314t-win_amd64.whl", hash = "sha256:4d8ae8659ab18c65ced284993c2265910f6c9e650189d4e3f68445ef82a810e4"},
-    {file = "cryptography-46.0.5-cp38-abi3-macosx_10_9_universal2.whl", hash = "sha256:4108d4c09fbbf2789d0c926eb4152ae1760d5a2d97612b92d508d96c861e4d31"},
-    {file = "cryptography-46.0.5-cp38-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:7d1f30a86d2757199cb2d56e48cce14deddf1f9c95f1ef1b64ee91ea43fe2e18"},
-    {file = "cryptography-46.0.5-cp38-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:039917b0dc418bb9f6edce8a906572d69e74bd330b0b3fea4f79dab7f8ddd235"},
-    {file = "cryptography-46.0.5-cp38-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:ba2a27ff02f48193fc4daeadf8ad2590516fa3d0adeeb34336b96f7fa64c1e3a"},
-    {file = "cryptography-46.0.5-cp38-abi3-manylinux_2_28_ppc64le.whl", hash = "sha256:61aa400dce22cb001a98014f647dc21cda08f7915ceb95df0c9eaf84b4b6af76"},
-    {file = "cryptography-46.0.5-cp38-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:3ce58ba46e1bc2aac4f7d9290223cead56743fa6ab94a5d53292ffaac6a91614"},
-    {file = "cryptography-46.0.5-cp38-abi3-manylinux_2_31_armv7l.whl", hash = "sha256:420d0e909050490d04359e7fdb5ed7e667ca5c3c402b809ae2563d7e66a92229"},
-    {file = "cryptography-46.0.5-cp38-abi3-manylinux_2_34_aarch64.whl", hash = "sha256:582f5fcd2afa31622f317f80426a027f30dc792e9c80ffee87b993200ea115f1"},
-    {file = "cryptography-46.0.5-cp38-abi3-manylinux_2_34_ppc64le.whl", hash = "sha256:bfd56bb4b37ed4f330b82402f6f435845a5f5648edf1ad497da51a8452d5d62d"},
-    {file = "cryptography-46.0.5-cp38-abi3-manylinux_2_34_x86_64.whl", hash = "sha256:a3d507bb6a513ca96ba84443226af944b0f7f47dcc9a399d110cd6146481d24c"},
-    {file = "cryptography-46.0.5-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:9f16fbdf4da055efb21c22d81b89f155f02ba420558db21288b3d0035bafd5f4"},
-    {file = "cryptography-46.0.5-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:ced80795227d70549a411a4ab66e8ce307899fad2220ce5ab2f296e687eacde9"},
-    {file = "cryptography-46.0.5-cp38-abi3-win32.whl", hash = "sha256:02f547fce831f5096c9a567fd41bc12ca8f11df260959ecc7c3202555cc47a72"},
-    {file = "cryptography-46.0.5-cp38-abi3-win_amd64.whl", hash = "sha256:556e106ee01aa13484ce9b0239bca667be5004efb0aabbed28d353df86445595"},
-    {file = "cryptography-46.0.5-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:3b4995dc971c9fb83c25aa44cf45f02ba86f71ee600d81091c2f0cbae116b06c"},
-    {file = "cryptography-46.0.5-pp311-pypy311_pp73-manylinux_2_28_aarch64.whl", hash = "sha256:bc84e875994c3b445871ea7181d424588171efec3e185dced958dad9e001950a"},
-    {file = "cryptography-46.0.5-pp311-pypy311_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:2ae6971afd6246710480e3f15824ed3029a60fc16991db250034efd0b9fb4356"},
-    {file = "cryptography-46.0.5-pp311-pypy311_pp73-manylinux_2_34_aarch64.whl", hash = "sha256:d861ee9e76ace6cf36a6a89b959ec08e7bc2493ee39d07ffe5acb23ef46d27da"},
-    {file = "cryptography-46.0.5-pp311-pypy311_pp73-manylinux_2_34_x86_64.whl", hash = "sha256:2b7a67c9cd56372f3249b39699f2ad479f6991e62ea15800973b956f4b73e257"},
-    {file = "cryptography-46.0.5-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:8456928655f856c6e1533ff59d5be76578a7157224dbd9ce6872f25055ab9ab7"},
-    {file = "cryptography-46.0.5.tar.gz", hash = "sha256:abace499247268e3757271b2f1e244b36b06f8515cf27c4d49468fc9eb16e93d"},
+    {file = "cryptography-46.0.4-cp311-abi3-macosx_10_9_universal2.whl", hash = "sha256:281526e865ed4166009e235afadf3a4c4cba6056f99336a99efba65336fd5485"},
+    {file = "cryptography-46.0.4-cp311-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:5f14fba5bf6f4390d7ff8f086c566454bff0411f6d8aa7af79c88b6f9267aecc"},
+    {file = "cryptography-46.0.4-cp311-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:47bcd19517e6389132f76e2d5303ded6cf3f78903da2158a671be8de024f4cd0"},
+    {file = "cryptography-46.0.4-cp311-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:01df4f50f314fbe7009f54046e908d1754f19d0c6d3070df1e6268c5a4af09fa"},
+    {file = "cryptography-46.0.4-cp311-abi3-manylinux_2_28_ppc64le.whl", hash = "sha256:5aa3e463596b0087b3da0dbe2b2487e9fc261d25da85754e30e3b40637d61f81"},
+    {file = "cryptography-46.0.4-cp311-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:0a9ad24359fee86f131836a9ac3bffc9329e956624a2d379b613f8f8abaf5255"},
+    {file = "cryptography-46.0.4-cp311-abi3-manylinux_2_31_armv7l.whl", hash = "sha256:dc1272e25ef673efe72f2096e92ae39dea1a1a450dd44918b15351f72c5a168e"},
+    {file = "cryptography-46.0.4-cp311-abi3-manylinux_2_34_aarch64.whl", hash = "sha256:de0f5f4ec8711ebc555f54735d4c673fc34b65c44283895f1a08c2b49d2fd99c"},
+    {file = "cryptography-46.0.4-cp311-abi3-manylinux_2_34_ppc64le.whl", hash = "sha256:eeeb2e33d8dbcccc34d64651f00a98cb41b2dc69cef866771a5717e6734dfa32"},
+    {file = "cryptography-46.0.4-cp311-abi3-manylinux_2_34_x86_64.whl", hash = "sha256:3d425eacbc9aceafd2cb429e42f4e5d5633c6f873f5e567077043ef1b9bbf616"},
+    {file = "cryptography-46.0.4-cp311-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:91627ebf691d1ea3976a031b61fb7bac1ccd745afa03602275dda443e11c8de0"},
+    {file = "cryptography-46.0.4-cp311-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:2d08bc22efd73e8854b0b7caff402d735b354862f1145d7be3b9c0f740fef6a0"},
+    {file = "cryptography-46.0.4-cp311-abi3-win32.whl", hash = "sha256:82a62483daf20b8134f6e92898da70d04d0ef9a75829d732ea1018678185f4f5"},
+    {file = "cryptography-46.0.4-cp311-abi3-win_amd64.whl", hash = "sha256:6225d3ebe26a55dbc8ead5ad1265c0403552a63336499564675b29eb3184c09b"},
+    {file = "cryptography-46.0.4-cp314-cp314t-macosx_10_9_universal2.whl", hash = "sha256:485e2b65d25ec0d901bca7bcae0f53b00133bf3173916d8e421f6fddde103908"},
+    {file = "cryptography-46.0.4-cp314-cp314t-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:078e5f06bd2fa5aea5a324f2a09f914b1484f1d0c2a4d6a8a28c74e72f65f2da"},
+    {file = "cryptography-46.0.4-cp314-cp314t-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:dce1e4f068f03008da7fa51cc7abc6ddc5e5de3e3d1550334eaf8393982a5829"},
+    {file = "cryptography-46.0.4-cp314-cp314t-manylinux_2_28_aarch64.whl", hash = "sha256:2067461c80271f422ee7bdbe79b9b4be54a5162e90345f86a23445a0cf3fd8a2"},
+    {file = "cryptography-46.0.4-cp314-cp314t-manylinux_2_28_ppc64le.whl", hash = "sha256:c92010b58a51196a5f41c3795190203ac52edfd5dc3ff99149b4659eba9d2085"},
+    {file = "cryptography-46.0.4-cp314-cp314t-manylinux_2_28_x86_64.whl", hash = "sha256:829c2b12bbc5428ab02d6b7f7e9bbfd53e33efd6672d21341f2177470171ad8b"},
+    {file = "cryptography-46.0.4-cp314-cp314t-manylinux_2_31_armv7l.whl", hash = "sha256:62217ba44bf81b30abaeda1488686a04a702a261e26f87db51ff61d9d3510abd"},
+    {file = "cryptography-46.0.4-cp314-cp314t-manylinux_2_34_aarch64.whl", hash = "sha256:9c2da296c8d3415b93e6053f5a728649a87a48ce084a9aaf51d6e46c87c7f2d2"},
+    {file = "cryptography-46.0.4-cp314-cp314t-manylinux_2_34_ppc64le.whl", hash = "sha256:9b34d8ba84454641a6bf4d6762d15847ecbd85c1316c0a7984e6e4e9f748ec2e"},
+    {file = "cryptography-46.0.4-cp314-cp314t-manylinux_2_34_x86_64.whl", hash = "sha256:df4a817fa7138dd0c96c8c8c20f04b8aaa1fac3bbf610913dcad8ea82e1bfd3f"},
+    {file = "cryptography-46.0.4-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:b1de0ebf7587f28f9190b9cb526e901bf448c9e6a99655d2b07fff60e8212a82"},
+    {file = "cryptography-46.0.4-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:9b4d17bc7bd7cdd98e3af40b441feaea4c68225e2eb2341026c84511ad246c0c"},
+    {file = "cryptography-46.0.4-cp314-cp314t-win32.whl", hash = "sha256:c411f16275b0dea722d76544a61d6421e2cc829ad76eec79280dbdc9ddf50061"},
+    {file = "cryptography-46.0.4-cp314-cp314t-win_amd64.whl", hash = "sha256:728fedc529efc1439eb6107b677f7f7558adab4553ef8669f0d02d42d7b959a7"},
+    {file = "cryptography-46.0.4-cp38-abi3-macosx_10_9_universal2.whl", hash = "sha256:a9556ba711f7c23f77b151d5798f3ac44a13455cc68db7697a1096e6d0563cab"},
+    {file = "cryptography-46.0.4-cp38-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:8bf75b0259e87fa70bddc0b8b4078b76e7fd512fd9afae6c1193bcf440a4dbef"},
+    {file = "cryptography-46.0.4-cp38-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:3c268a3490df22270955966ba236d6bc4a8f9b6e4ffddb78aac535f1a5ea471d"},
+    {file = "cryptography-46.0.4-cp38-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:812815182f6a0c1d49a37893a303b44eaac827d7f0d582cecfc81b6427f22973"},
+    {file = "cryptography-46.0.4-cp38-abi3-manylinux_2_28_ppc64le.whl", hash = "sha256:a90e43e3ef65e6dcf969dfe3bb40cbf5aef0d523dff95bfa24256be172a845f4"},
+    {file = "cryptography-46.0.4-cp38-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:a05177ff6296644ef2876fce50518dffb5bcdf903c85250974fc8bc85d54c0af"},
+    {file = "cryptography-46.0.4-cp38-abi3-manylinux_2_31_armv7l.whl", hash = "sha256:daa392191f626d50f1b136c9b4cf08af69ca8279d110ea24f5c2700054d2e263"},
+    {file = "cryptography-46.0.4-cp38-abi3-manylinux_2_34_aarch64.whl", hash = "sha256:e07ea39c5b048e085f15923511d8121e4a9dc45cee4e3b970ca4f0d338f23095"},
+    {file = "cryptography-46.0.4-cp38-abi3-manylinux_2_34_ppc64le.whl", hash = "sha256:d5a45ddc256f492ce42a4e35879c5e5528c09cd9ad12420828c972951d8e016b"},
+    {file = "cryptography-46.0.4-cp38-abi3-manylinux_2_34_x86_64.whl", hash = "sha256:6bb5157bf6a350e5b28aee23beb2d84ae6f5be390b2f8ee7ea179cda077e1019"},
+    {file = "cryptography-46.0.4-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:dd5aba870a2c40f87a3af043e0dee7d9eb02d4aff88a797b48f2b43eff8c3ab4"},
+    {file = "cryptography-46.0.4-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:93d8291da8d71024379ab2cb0b5c57915300155ad42e07f76bea6ad838d7e59b"},
+    {file = "cryptography-46.0.4-cp38-abi3-win32.whl", hash = "sha256:0563655cb3c6d05fb2afe693340bc050c30f9f34e15763361cf08e94749401fc"},
+    {file = "cryptography-46.0.4-cp38-abi3-win_amd64.whl", hash = "sha256:fa0900b9ef9c49728887d1576fd8d9e7e3ea872fa9b25ef9b64888adc434e976"},
+    {file = "cryptography-46.0.4-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:766330cce7416c92b5e90c3bb71b1b79521760cdcfc3a6a1a182d4c9fab23d2b"},
+    {file = "cryptography-46.0.4-pp311-pypy311_pp73-manylinux_2_28_aarch64.whl", hash = "sha256:c236a44acfb610e70f6b3e1c3ca20ff24459659231ef2f8c48e879e2d32b73da"},
+    {file = "cryptography-46.0.4-pp311-pypy311_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:8a15fb869670efa8f83cbffbc8753c1abf236883225aed74cd179b720ac9ec80"},
+    {file = "cryptography-46.0.4-pp311-pypy311_pp73-manylinux_2_34_aarch64.whl", hash = "sha256:fdc3daab53b212472f1524d070735b2f0c214239df131903bae1d598016fa822"},
+    {file = "cryptography-46.0.4-pp311-pypy311_pp73-manylinux_2_34_x86_64.whl", hash = "sha256:44cc0675b27cadb71bdbb96099cca1fa051cd11d2ade09e5cd3a2edb929ed947"},
+    {file = "cryptography-46.0.4-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:be8c01a7d5a55f9a47d1888162b76c8f49d62b234d88f0ff91a9fbebe32ffbc3"},
+    {file = "cryptography-46.0.4.tar.gz", hash = "sha256:bfd019f60f8abc2ed1b9be4ddc21cfef059c841d86d710bb69909a688cbb8f59"},
 ]
 
 [package.dependencies]
@@ -516,7 +516,7 @@ nox = ["nox[uv] (>=2024.4.15)"]
 pep8test = ["check-sdist", "click (>=8.0.1)", "mypy (>=1.14)", "ruff (>=0.11.11)"]
 sdist = ["build (>=1.0.0)"]
 ssh = ["bcrypt (>=3.1.5)"]
-test = ["certifi (>=2024)", "cryptography-vectors (==46.0.5)", "pretend (>=0.7)", "pytest (>=7.4.0)", "pytest-benchmark (>=4.0)", "pytest-cov (>=2.10.1)", "pytest-xdist (>=3.5.0)"]
+test = ["certifi (>=2024)", "cryptography-vectors (==46.0.4)", "pretend (>=0.7)", "pytest (>=7.4.0)", "pytest-benchmark (>=4.0)", "pytest-cov (>=2.10.1)", "pytest-xdist (>=3.5.0)"]
 test-randomorder = ["pytest-randomly"]
 
 [[package]]
@@ -570,25 +570,24 @@ tests = ["coverage", "coveralls", "dill", "mock", "nose"]
 
 [[package]]
 name = "fastapi"
-version = "0.128.7"
+version = "0.128.0"
 description = "FastAPI framework, high performance, easy to learn, fast to code, ready for production"
 optional = false
 python-versions = ">=3.9"
 groups = ["main"]
 files = [
-    {file = "fastapi-0.128.7-py3-none-any.whl", hash = "sha256:6bd9bd31cb7047465f2d3fa3ba3f33b0870b17d4eaf7cdb36d1576ab060ad662"},
-    {file = "fastapi-0.128.7.tar.gz", hash = "sha256:783c273416995486c155ad2c0e2b45905dedfaf20b9ef8d9f6a9124670639a24"},
+    {file = "fastapi-0.128.0-py3-none-any.whl", hash = "sha256:aebd93f9716ee3b4f4fcfe13ffb7cf308d99c9f3ab5622d8877441072561582d"},
+    {file = "fastapi-0.128.0.tar.gz", hash = "sha256:1cc179e1cef10a6be60ffe429f79b829dce99d8de32d7acb7e6c8dfdf7f2645a"},
 ]
 
 [package.dependencies]
 annotated-doc = ">=0.0.2"
 pydantic = ">=2.7.0"
-starlette = ">=0.40.0,<1.0.0"
+starlette = ">=0.40.0,<0.51.0"
 typing-extensions = ">=4.8.0"
-typing-inspection = ">=0.4.2"
 
 [package.extras]
-all = ["email-validator (>=2.0.0)", "fastapi-cli[standard] (>=0.0.8)", "httpx (>=0.23.0,<1.0.0)", "itsdangerous (>=1.1.0)", "jinja2 (>=3.1.5)", "orjson (>=3.9.3)", "pydantic-extra-types (>=2.0.0)", "pydantic-settings (>=2.0.0)", "python-multipart (>=0.0.18)", "pyyaml (>=5.3.1)", "ujson (>=5.8.0)", "uvicorn[standard] (>=0.12.0)"]
+all = ["email-validator (>=2.0.0)", "fastapi-cli[standard] (>=0.0.8)", "httpx (>=0.23.0,<1.0.0)", "itsdangerous (>=1.1.0)", "jinja2 (>=3.1.5)", "orjson (>=3.2.1)", "pydantic-extra-types (>=2.0.0)", "pydantic-settings (>=2.0.0)", "python-multipart (>=0.0.18)", "pyyaml (>=5.3.1)", "ujson (>=4.0.1,!=4.0.2,!=4.1.0,!=4.2.0,!=4.3.0,!=5.0.0,!=5.1.0)", "uvicorn[standard] (>=0.12.0)"]
 standard = ["email-validator (>=2.0.0)", "fastapi-cli[standard] (>=0.0.8)", "httpx (>=0.23.0,<1.0.0)", "jinja2 (>=3.1.5)", "pydantic-extra-types (>=2.0.0)", "pydantic-settings (>=2.0.0)", "python-multipart (>=0.0.18)", "uvicorn[standard] (>=0.12.0)"]
 standard-no-fastapi-cloud-cli = ["email-validator (>=2.0.0)", "fastapi-cli[standard-no-fastapi-cloud-cli] (>=0.0.8)", "httpx (>=0.23.0,<1.0.0)", "jinja2 (>=3.1.5)", "pydantic-extra-types (>=2.0.0)", "pydantic-settings (>=2.0.0)", "python-multipart (>=0.0.18)", "uvicorn[standard] (>=0.12.0)"]
 
@@ -1063,14 +1062,14 @@ urllib3 = ">=1.26.0,<3"
 
 [[package]]
 name = "launchdarkly-server-sdk"
-version = "9.15.0"
+version = "9.14.1"
 description = "LaunchDarkly SDK for Python"
 optional = false
-python-versions = ">=3.10"
+python-versions = ">=3.9"
 groups = ["main"]
 files = [
-    {file = "launchdarkly_server_sdk-9.15.0-py3-none-any.whl", hash = "sha256:c267e29bfa3fb5e2a06a208448ada6ed5557a2924979b8d79c970b45d227c668"},
-    {file = "launchdarkly_server_sdk-9.15.0.tar.gz", hash = "sha256:f31441b74bc1a69c381db57c33116509e407a2612628ad6dff0a7dbb39d5020b"},
+    {file = "launchdarkly_server_sdk-9.14.1-py3-none-any.whl", hash = "sha256:a9e2bd9ecdef845cd631ae0d4334a1115e5b44257c42eb2349492be4bac7815c"},
+    {file = "launchdarkly_server_sdk-9.14.1.tar.gz", hash = "sha256:1df44baf0a0efa74d8c1dad7a00592b98bce7d19edded7f770da8dbc49922213"},
 ]
 
 [package.dependencies]
@@ -1479,14 +1478,14 @@ testing = ["coverage", "pytest", "pytest-benchmark"]
 
 [[package]]
 name = "postgrest"
-version = "2.28.0"
+version = "2.27.2"
 description = "PostgREST client for Python. This library provides an ORM interface to PostgREST."
 optional = false
 python-versions = ">=3.9"
 groups = ["main"]
 files = [
-    {file = "postgrest-2.28.0-py3-none-any.whl", hash = "sha256:7bca2f24dd1a1bf8a3d586c7482aba6cd41662da6733045fad585b63b7f7df75"},
-    {file = "postgrest-2.28.0.tar.gz", hash = "sha256:c36b38646d25ea4255321d3d924ce70f8d20ec7799cb42c1221d6a818d4f6515"},
+    {file = "postgrest-2.27.2-py3-none-any.whl", hash = "sha256:1666fef3de05ca097a314433dd5ae2f2d71c613cb7b233d0f468c4ffe37277da"},
+    {file = "postgrest-2.27.2.tar.gz", hash = "sha256:55407d530b5af3d64e883a71fec1f345d369958f723ce4a8ab0b7d169e313242"},
 ]
 
 [package.dependencies]
@@ -2249,14 +2248,14 @@ cli = ["click (>=5.0)"]
 
 [[package]]
 name = "realtime"
-version = "2.28.0"
+version = "2.27.2"
 description = ""
 optional = false
 python-versions = ">=3.9"
 groups = ["main"]
 files = [
-    {file = "realtime-2.28.0-py3-none-any.whl", hash = "sha256:db1bd59bab9b1fcc9f9d3b1a073bed35bf4994d720e6751f10031a58d57a3836"},
-    {file = "realtime-2.28.0.tar.gz", hash = "sha256:d18cedcebd6a8f22fcd509bc767f639761eb218b7b2b6f14fc4205b6259b50fc"},
+    {file = "realtime-2.27.2-py3-none-any.whl", hash = "sha256:34a9cbb26a274e707e8fc9e3ee0a66de944beac0fe604dc336d1e985db2c830f"},
+    {file = "realtime-2.27.2.tar.gz", hash = "sha256:b960a90294d2cea1b3f1275ecb89204304728e08fff1c393cc1b3150739556b3"},
 ]
 
 [package.dependencies]
@@ -2437,14 +2436,14 @@ full = ["httpx (>=0.27.0,<0.29.0)", "itsdangerous", "jinja2", "python-multipart
 
 [[package]]
 name = "storage3"
-version = "2.28.0"
+version = "2.27.2"
 description = "Supabase Storage client for Python."
 optional = false
 python-versions = ">=3.9"
 groups = ["main"]
 files = [
-    {file = "storage3-2.28.0-py3-none-any.whl", hash = "sha256:ecb50efd2ac71dabbdf97e99ad346eafa630c4c627a8e5a138ceb5fbbadae716"},
-    {file = "storage3-2.28.0.tar.gz", hash = "sha256:bc1d008aff67de7a0f2bd867baee7aadbcdb6f78f5a310b4f7a38e8c13c19865"},
+    {file = "storage3-2.27.2-py3-none-any.whl", hash = "sha256:e6f16e7a260729e7b1f46e9bf61746805a02e30f5e419ee1291007c432e3ec63"},
+    {file = "storage3-2.27.2.tar.gz", hash = "sha256:cb4807b7f86b4bb1272ac6fdd2f3cfd8ba577297046fa5f88557425200275af5"},
 ]
 
 [package.dependencies]
@@ -2488,35 +2487,35 @@ python-dateutil = ">=2.6.0"
 
 [[package]]
 name = "supabase"
-version = "2.28.0"
+version = "2.27.2"
 description = "Supabase client for Python."
 optional = false
 python-versions = ">=3.9"
 groups = ["main"]
 files = [
-    {file = "supabase-2.28.0-py3-none-any.whl", hash = "sha256:42776971c7d0ccca16034df1ab96a31c50228eb1eb19da4249ad2f756fc20272"},
-    {file = "supabase-2.28.0.tar.gz", hash = "sha256:aea299aaab2a2eed3c57e0be7fc035c6807214194cce795a3575add20268ece1"},
+    {file = "supabase-2.27.2-py3-none-any.whl", hash = "sha256:d4dce00b3a418ee578017ec577c0e5be47a9a636355009c76f20ed2faa15bc54"},
+    {file = "supabase-2.27.2.tar.gz", hash = "sha256:2aed40e4f3454438822442a1e94a47be6694c2c70392e7ae99b51a226d4293f7"},
 ]
 
 [package.dependencies]
 httpx = ">=0.26,<0.29"
-postgrest = "2.28.0"
-realtime = "2.28.0"
-storage3 = "2.28.0"
-supabase-auth = "2.28.0"
-supabase-functions = "2.28.0"
+postgrest = "2.27.2"
+realtime = "2.27.2"
+storage3 = "2.27.2"
+supabase-auth = "2.27.2"
+supabase-functions = "2.27.2"
 yarl = ">=1.22.0"
 
 [[package]]
 name = "supabase-auth"
-version = "2.28.0"
+version = "2.27.2"
 description = "Python Client Library for Supabase Auth"
 optional = false
 python-versions = ">=3.9"
 groups = ["main"]
 files = [
-    {file = "supabase_auth-2.28.0-py3-none-any.whl", hash = "sha256:2ac85026cc285054c7fa6d41924f3a333e9ec298c013e5b5e1754039ba7caec9"},
-    {file = "supabase_auth-2.28.0.tar.gz", hash = "sha256:2bb8f18ff39934e44b28f10918db965659f3735cd6fbfcc022fe0b82dbf8233e"},
+    {file = "supabase_auth-2.27.2-py3-none-any.whl", hash = "sha256:78ec25b11314d0a9527a7205f3b1c72560dccdc11b38392f80297ef98664ee91"},
+    {file = "supabase_auth-2.27.2.tar.gz", hash = "sha256:0f5bcc79b3677cb42e9d321f3c559070cfa40d6a29a67672cc8382fb7dc2fe97"},
 ]
 
 [package.dependencies]
@@ -2526,14 +2525,14 @@ pyjwt = {version = ">=2.10.1", extras = ["crypto"]}
 
 [[package]]
 name = "supabase-functions"
-version = "2.28.0"
+version = "2.27.2"
 description = "Library for Supabase Functions"
 optional = false
 python-versions = ">=3.9"
 groups = ["main"]
 files = [
-    {file = "supabase_functions-2.28.0-py3-none-any.whl", hash = "sha256:30bf2d586f8df285faf0621bb5d5bb3ec3157234fc820553ca156f009475e4ae"},
-    {file = "supabase_functions-2.28.0.tar.gz", hash = "sha256:db3dddfc37aca5858819eb461130968473bd8c75bd284581013958526dac718b"},
+    {file = "supabase_functions-2.27.2-py3-none-any.whl", hash = "sha256:db480efc669d0bca07605b9b6f167312af43121adcc842a111f79bea416ef754"},
+    {file = "supabase_functions-2.27.2.tar.gz", hash = "sha256:d0c8266207a94371cb3fd35ad3c7f025b78a97cf026861e04ccd35ac1775f80b"},
 ]
 
 [package.dependencies]
@@ -2912,4 +2911,4 @@ type = ["pytest-mypy"]
 [metadata]
 lock-version = "2.1"
 python-versions = ">=3.10,<4.0"
-content-hash = "9619cae908ad38fa2c48016a58bcf4241f6f5793aa0e6cc140276e91c433cbbb"
+content-hash = "40eae94995dc0a388fa832ed4af9b6137f28d5b5ced3aaea70d5f91d4d9a179d"

autogpt_platform/autogpt_libs/pyproject.toml

@@ -11,14 +11,14 @@ python = ">=3.10,<4.0"
 colorama = "^0.4.6"
 cryptography = "^46.0"
 expiringdict = "^1.2.2"
-fastapi = "^0.128.7"
+fastapi = "^0.128.0"
 google-cloud-logging = "^3.13.0"
-launchdarkly-server-sdk = "^9.15.0"
+launchdarkly-server-sdk = "^9.14.1"
 pydantic = "^2.12.5"
 pydantic-settings = "^2.12.0"
 pyjwt = { version = "^2.11.0", extras = ["crypto"] }
 redis = "^6.2.0"
-supabase = "^2.28.0"
+supabase = "^2.27.2"
 uvicorn = "^0.40.0"
 
 [tool.poetry.group.dev.dependencies]

autogpt_platform/backend/.env.default

@@ -104,12 +104,6 @@ TWITTER_CLIENT_SECRET=
 # Make a new workspace for your OAuth APP -- trust me
 # https://linear.app/settings/api/applications/new
 # Callback URL: http://localhost:3000/auth/integrations/oauth_callback
-LINEAR_API_KEY=
-# Linear project and team IDs for the feature request tracker.
-# Find these in your Linear workspace URL: linear.app/<workspace>/project/<project-id>
-# and in team settings. Used by the chat copilot to file and search feature requests.
-LINEAR_FEATURE_REQUEST_PROJECT_ID=
-LINEAR_FEATURE_REQUEST_TEAM_ID=
 LINEAR_CLIENT_ID=
 LINEAR_CLIENT_SECRET=
 

autogpt_platform/backend/Dockerfile

@@ -1,5 +1,3 @@
-# ============================ DEPENDENCY BUILDER ============================ #
-
 FROM debian:13-slim AS builder
 
 # Set environment variables
@@ -53,62 +51,27 @@ COPY autogpt_platform/backend/backend/data/partial_types.py ./backend/data/parti
 COPY autogpt_platform/backend/gen_prisma_types_stub.py ./
 RUN poetry run prisma generate && poetry run gen-prisma-stub
 
-# =============================== DB MIGRATOR =============================== #
-
-# Lightweight migrate stage - only needs Prisma CLI, not full Python environment
-FROM debian:13-slim AS migrate
-
-WORKDIR /app/autogpt_platform/backend
-
-ENV DEBIAN_FRONTEND=noninteractive
-
-# Install only what's needed for prisma migrate: Node.js and minimal Python for prisma-python
-RUN apt-get update && apt-get install -y --no-install-recommends \
-    python3.13 \
-    python3-pip \
-    ca-certificates \
-    && rm -rf /var/lib/apt/lists/*
-
-# Copy Node.js from builder (needed for Prisma CLI)
-COPY --from=builder /usr/bin/node /usr/bin/node
-COPY --from=builder /usr/lib/node_modules /usr/lib/node_modules
-COPY --from=builder /usr/bin/npm /usr/bin/npm
-
-# Copy Prisma binaries
-COPY --from=builder /root/.cache/prisma-python/binaries /root/.cache/prisma-python/binaries
-
-# Install prisma-client-py directly (much smaller than copying full venv)
-RUN pip3 install prisma>=0.15.0 --break-system-packages
-
-COPY autogpt_platform/backend/schema.prisma ./
-COPY autogpt_platform/backend/backend/data/partial_types.py ./backend/data/partial_types.py
-COPY autogpt_platform/backend/gen_prisma_types_stub.py ./
-COPY autogpt_platform/backend/migrations ./migrations
-
-# ============================== BACKEND SERVER ============================== #
-
-FROM debian:13-slim AS server
+FROM debian:13-slim AS server_dependencies
 
 WORKDIR /app
 
-ENV DEBIAN_FRONTEND=noninteractive
+ENV POETRY_HOME=/opt/poetry \
+    POETRY_NO_INTERACTION=1 \
+    POETRY_VIRTUALENVS_CREATE=true \
+    POETRY_VIRTUALENVS_IN_PROJECT=true \
+    DEBIAN_FRONTEND=noninteractive
+ENV PATH=/opt/poetry/bin:$PATH
 
-# Install Python, FFmpeg, ImageMagick, and CLI tools for agent use.
-# bubblewrap provides OS-level sandbox (whitelist-only FS + no network)
-# for the bash_exec MCP tool.
-# Using --no-install-recommends saves ~650MB by skipping unnecessary deps like llvm, mesa, etc.
-RUN apt-get update && apt-get install -y --no-install-recommends \
+# Install Python, FFmpeg, and ImageMagick (required for video processing blocks)
+RUN apt-get update && apt-get install -y \
     python3.13 \
     python3-pip \
     ffmpeg \
     imagemagick \
-    jq \
-    ripgrep \
-    tree \
-    bubblewrap \
     && rm -rf /var/lib/apt/lists/*
 
-# Copy poetry (build-time only, for `poetry install --only-root` to create entry points)
+# Copy only necessary files from builder
+COPY --from=builder /app /app
 COPY --from=builder /usr/local/lib/python3* /usr/local/lib/python3*
 COPY --from=builder /usr/local/bin/poetry /usr/local/bin/poetry
 # Copy Node.js installation for Prisma
@@ -118,25 +81,30 @@ COPY --from=builder /usr/bin/npm /usr/bin/npm
 COPY --from=builder /usr/bin/npx /usr/bin/npx
 COPY --from=builder /root/.cache/prisma-python/binaries /root/.cache/prisma-python/binaries
 
-WORKDIR /app/autogpt_platform/backend
-
-# Copy only the .venv from builder (not the entire /app directory)
-# The .venv includes the generated Prisma client
-COPY --from=builder /app/autogpt_platform/backend/.venv ./.venv
 ENV PATH="/app/autogpt_platform/backend/.venv/bin:$PATH"
 
-# Copy dependency files + autogpt_libs (path dependency)
-COPY autogpt_platform/autogpt_libs /app/autogpt_platform/autogpt_libs
-COPY autogpt_platform/backend/poetry.lock autogpt_platform/backend/pyproject.toml ./
+RUN mkdir -p /app/autogpt_platform/autogpt_libs
+RUN mkdir -p /app/autogpt_platform/backend
 
-# Copy backend code + docs (for Copilot docs search)
-COPY autogpt_platform/backend ./
+COPY autogpt_platform/autogpt_libs /app/autogpt_platform/autogpt_libs
+
+COPY autogpt_platform/backend/poetry.lock autogpt_platform/backend/pyproject.toml /app/autogpt_platform/backend/
+
+WORKDIR /app/autogpt_platform/backend
+
+FROM server_dependencies AS migrate
+
+# Migration stage only needs schema and migrations - much lighter than full backend
+COPY autogpt_platform/backend/schema.prisma /app/autogpt_platform/backend/
+COPY autogpt_platform/backend/backend/data/partial_types.py /app/autogpt_platform/backend/backend/data/partial_types.py
+COPY autogpt_platform/backend/migrations /app/autogpt_platform/backend/migrations
+
+FROM server_dependencies AS server
+
+COPY autogpt_platform/backend /app/autogpt_platform/backend
 COPY docs /app/docs
-# Install the project package to create entry point scripts in .venv/bin/
-# (e.g., rest, executor, ws, db, scheduler, notification - see [tool.poetry.scripts])
-RUN POETRY_VIRTUALENVS_CREATE=true POETRY_VIRTUALENVS_IN_PROJECT=true \
-    poetry install --no-ansi --only-root
+RUN poetry install --no-ansi --only-root
 
 ENV PORT=8000
 
-CMD ["rest"]
+CMD ["poetry", "run", "rest"]

autogpt_platform/backend/backend/api/conftest.py

@@ -1,9 +1,4 @@
-"""Common test fixtures for server tests.
-
-Note: Common fixtures like test_user_id, admin_user_id, target_user_id,
-setup_test_user, and setup_admin_user are defined in the parent conftest.py
-(backend/conftest.py) and are available here automatically.
-"""
+"""Common test fixtures for server tests."""
 
 import pytest
 from pytest_snapshot.plugin import Snapshot
@@ -16,6 +11,54 @@ def configured_snapshot(snapshot: Snapshot) -> Snapshot:
     return snapshot
 
 
+@pytest.fixture
+def test_user_id() -> str:
+    """Test user ID fixture."""
+    return "3e53486c-cf57-477e-ba2a-cb02dc828e1a"
+
+
+@pytest.fixture
+def admin_user_id() -> str:
+    """Admin user ID fixture."""
+    return "4e53486c-cf57-477e-ba2a-cb02dc828e1b"
+
+
+@pytest.fixture
+def target_user_id() -> str:
+    """Target user ID fixture."""
+    return "5e53486c-cf57-477e-ba2a-cb02dc828e1c"
+
+
+@pytest.fixture
+async def setup_test_user(test_user_id):
+    """Create test user in database before tests."""
+    from backend.data.user import get_or_create_user
+
+    # Create the test user in the database using JWT token format
+    user_data = {
+        "sub": test_user_id,
+        "email": "test@example.com",
+        "user_metadata": {"name": "Test User"},
+    }
+    await get_or_create_user(user_data)
+    return test_user_id
+
+
+@pytest.fixture
+async def setup_admin_user(admin_user_id):
+    """Create admin user in database before tests."""
+    from backend.data.user import get_or_create_user
+
+    # Create the admin user in the database using JWT token format
+    user_data = {
+        "sub": admin_user_id,
+        "email": "test-admin@example.com",
+        "user_metadata": {"name": "Test Admin"},
+    }
+    await get_or_create_user(user_data)
+    return admin_user_id
+
+
 @pytest.fixture
 def mock_jwt_user(test_user_id):
     """Provide mock JWT payload for regular user testing."""

autogpt_platform/backend/backend/api/external/v1/routes.py

@@ -10,7 +10,7 @@ from typing_extensions import TypedDict
 
 import backend.api.features.store.cache as store_cache
 import backend.api.features.store.model as store_model
-import backend.blocks
+import backend.data.block
 from backend.api.external.middleware import require_permission
 from backend.data import execution as execution_db
 from backend.data import graph as graph_db
@@ -67,7 +67,7 @@ async def get_user_info(
     dependencies=[Security(require_permission(APIKeyPermission.READ_BLOCK))],
 )
 async def get_graph_blocks() -> Sequence[dict[Any, Any]]:
-    blocks = [block() for block in backend.blocks.get_blocks().values()]
+    blocks = [block() for block in backend.data.block.get_blocks().values()]
     return [b.to_dict() for b in blocks if not b.disabled]
 
 
@@ -83,7 +83,7 @@ async def execute_graph_block(
         require_permission(APIKeyPermission.EXECUTE_BLOCK)
     ),
 ) -> CompletedBlockOutput:
-    obj = backend.blocks.get_block(block_id)
+    obj = backend.data.block.get_block(block_id)
     if not obj:
         raise HTTPException(status_code=404, detail=f"Block #{block_id} not found.")
     if obj.disabled:

autogpt_platform/backend/backend/api/external/v1/tools.py

@@ -15,9 +15,9 @@ from prisma.enums import APIKeyPermission
 from pydantic import BaseModel, Field
 
 from backend.api.external.middleware import require_permission
-from backend.copilot.model import ChatSession
-from backend.copilot.tools import find_agent_tool, run_agent_tool
-from backend.copilot.tools.models import ToolResponseBase
+from backend.api.features.chat.model import ChatSession
+from backend.api.features.chat.tools import find_agent_tool, run_agent_tool
+from backend.api.features.chat.tools.models import ToolResponseBase
 from backend.data.auth.base import APIAuthorizationInfo
 
 logger = logging.getLogger(__name__)

autogpt_platform/backend/backend/api/features/builder/db.py

@@ -10,15 +10,10 @@ import backend.api.features.library.db as library_db
 import backend.api.features.library.model as library_model
 import backend.api.features.store.db as store_db
 import backend.api.features.store.model as store_model
+import backend.data.block
 from backend.blocks import load_all_blocks
-from backend.blocks._base import (
-    AnyBlockSchema,
-    BlockCategory,
-    BlockInfo,
-    BlockSchema,
-    BlockType,
-)
 from backend.blocks.llm import LlmModel
+from backend.data.block import AnyBlockSchema, BlockCategory, BlockInfo, BlockSchema
 from backend.data.db import query_raw_with_schema
 from backend.integrations.providers import ProviderName
 from backend.util.cache import cached
@@ -27,7 +22,7 @@ from backend.util.models import Pagination
 from .model import (
     BlockCategoryResponse,
     BlockResponse,
-    BlockTypeFilter,
+    BlockType,
     CountResponse,
     FilterType,
     Provider,
@@ -93,7 +88,7 @@ def get_block_categories(category_blocks: int = 3) -> list[BlockCategoryResponse
 def get_blocks(
     *,
     category: str | None = None,
-    type: BlockTypeFilter | None = None,
+    type: BlockType | None = None,
     provider: ProviderName | None = None,
     page: int = 1,
     page_size: int = 50,
@@ -674,9 +669,9 @@ async def get_suggested_blocks(count: int = 5) -> list[BlockInfo]:
     for block_type in load_all_blocks().values():
         block: AnyBlockSchema = block_type()
         if block.disabled or block.block_type in (
-            BlockType.INPUT,
-            BlockType.OUTPUT,
-            BlockType.AGENT,
+            backend.data.block.BlockType.INPUT,
+            backend.data.block.BlockType.OUTPUT,
+            backend.data.block.BlockType.AGENT,
         ):
             continue
         # Find the execution count for this block

autogpt_platform/backend/backend/api/features/builder/model.py

@@ -4,7 +4,7 @@ from pydantic import BaseModel
 
 import backend.api.features.library.model as library_model
 import backend.api.features.store.model as store_model
-from backend.blocks._base import BlockInfo
+from backend.data.block import BlockInfo
 from backend.integrations.providers import ProviderName
 from backend.util.models import Pagination
 
@@ -15,7 +15,7 @@ FilterType = Literal[
     "my_agents",
 ]
 
-BlockTypeFilter = Literal["all", "input", "action", "output"]
+BlockType = Literal["all", "input", "action", "output"]
 
 
 class SearchEntry(BaseModel):

autogpt_platform/backend/backend/api/features/builder/routes.py

@@ -88,7 +88,7 @@ async def get_block_categories(
 )
 async def get_blocks(
     category: Annotated[str | None, fastapi.Query()] = None,
-    type: Annotated[builder_model.BlockTypeFilter | None, fastapi.Query()] = None,
+    type: Annotated[builder_model.BlockType | None, fastapi.Query()] = None,
     provider: Annotated[ProviderName | None, fastapi.Query()] = None,
     page: Annotated[int, fastapi.Query()] = 1,
     page_size: Annotated[int, fastapi.Query()] = 50,

autogpt_platform/backend/backend/api/features/chat/completion_consumer.py

@@ -0,0 +1,368 @@
+"""Redis Streams consumer for operation completion messages.
+
+This module provides a consumer (ChatCompletionConsumer) that listens for
+completion notifications (OperationCompleteMessage) from external services
+(like Agent Generator) and triggers the appropriate stream registry and
+chat service updates via process_operation_success/process_operation_failure.
+
+Why Redis Streams instead of RabbitMQ?
+--------------------------------------
+While the project typically uses RabbitMQ for async task queues (e.g., execution
+queue), Redis Streams was chosen for chat completion notifications because:
+
+1. **Unified Infrastructure**: The SSE reconnection feature already uses Redis
+   Streams (via stream_registry) for message persistence and replay. Using Redis
+   Streams for completion notifications keeps all chat streaming infrastructure
+   in one system, simplifying operations and reducing cross-system coordination.
+
+2. **Message Replay**: Redis Streams support XREAD with arbitrary message IDs,
+   allowing consumers to replay missed messages after reconnection. This aligns
+   with the SSE reconnection pattern where clients can resume from last_message_id.
+
+3. **Consumer Groups with XAUTOCLAIM**: Redis consumer groups provide automatic
+   load balancing across pods with explicit message claiming (XAUTOCLAIM) for
+   recovering from dead consumers - ideal for the completion callback pattern.
+
+4. **Lower Latency**: For real-time SSE updates, Redis (already in-memory for
+   stream_registry) provides lower latency than an additional RabbitMQ hop.
+
+5. **Atomicity with Task State**: Completion processing often needs to update
+   task metadata stored in Redis. Keeping both in Redis enables simpler
+   transactional semantics without distributed coordination.
+
+The consumer uses Redis Streams with consumer groups for reliable message
+processing across multiple platform pods, with XAUTOCLAIM for reclaiming
+stale pending messages from dead consumers.
+"""
+
+import asyncio
+import logging
+import os
+import uuid
+from typing import Any
+
+import orjson
+from prisma import Prisma
+from pydantic import BaseModel
+from redis.exceptions import ResponseError
+
+from backend.data.redis_client import get_redis_async
+
+from . import stream_registry
+from .completion_handler import process_operation_failure, process_operation_success
+from .config import ChatConfig
+
+logger = logging.getLogger(__name__)
+config = ChatConfig()
+
+
+class OperationCompleteMessage(BaseModel):
+    """Message format for operation completion notifications."""
+
+    operation_id: str
+    task_id: str
+    success: bool
+    result: dict | str | None = None
+    error: str | None = None
+
+
+class ChatCompletionConsumer:
+    """Consumer for chat operation completion messages from Redis Streams.
+
+    This consumer initializes its own Prisma client in start() to ensure
+    database operations work correctly within this async context.
+
+    Uses Redis consumer groups to allow multiple platform pods to consume
+    messages reliably with automatic redelivery on failure.
+    """
+
+    def __init__(self):
+        self._consumer_task: asyncio.Task | None = None
+        self._running = False
+        self._prisma: Prisma | None = None
+        self._consumer_name = f"consumer-{uuid.uuid4().hex[:8]}"
+
+    async def start(self) -> None:
+        """Start the completion consumer."""
+        if self._running:
+            logger.warning("Completion consumer already running")
+            return
+
+        # Create consumer group if it doesn't exist
+        try:
+            redis = await get_redis_async()
+            await redis.xgroup_create(
+                config.stream_completion_name,
+                config.stream_consumer_group,
+                id="0",
+                mkstream=True,
+            )
+            logger.info(
+                f"Created consumer group '{config.stream_consumer_group}' "
+                f"on stream '{config.stream_completion_name}'"
+            )
+        except ResponseError as e:
+            if "BUSYGROUP" in str(e):
+                logger.debug(
+                    f"Consumer group '{config.stream_consumer_group}' already exists"
+                )
+            else:
+                raise
+
+        self._running = True
+        self._consumer_task = asyncio.create_task(self._consume_messages())
+        logger.info(
+            f"Chat completion consumer started (consumer: {self._consumer_name})"
+        )
+
+    async def _ensure_prisma(self) -> Prisma:
+        """Lazily initialize Prisma client on first use."""
+        if self._prisma is None:
+            database_url = os.getenv("DATABASE_URL", "postgresql://localhost:5432")
+            self._prisma = Prisma(datasource={"url": database_url})
+            await self._prisma.connect()
+            logger.info("[COMPLETION] Consumer Prisma client connected (lazy init)")
+        return self._prisma
+
+    async def stop(self) -> None:
+        """Stop the completion consumer."""
+        self._running = False
+
+        if self._consumer_task:
+            self._consumer_task.cancel()
+            try:
+                await self._consumer_task
+            except asyncio.CancelledError:
+                pass
+            self._consumer_task = None
+
+        if self._prisma:
+            await self._prisma.disconnect()
+            self._prisma = None
+            logger.info("[COMPLETION] Consumer Prisma client disconnected")
+
+        logger.info("Chat completion consumer stopped")
+
+    async def _consume_messages(self) -> None:
+        """Main message consumption loop with retry logic."""
+        max_retries = 10
+        retry_delay = 5  # seconds
+        retry_count = 0
+        block_timeout = 5000  # milliseconds
+
+        while self._running and retry_count < max_retries:
+            try:
+                redis = await get_redis_async()
+
+                # Reset retry count on successful connection
+                retry_count = 0
+
+                while self._running:
+                    # First, claim any stale pending messages from dead consumers
+                    # Redis does NOT auto-redeliver pending messages; we must explicitly
+                    # claim them using XAUTOCLAIM
+                    try:
+                        claimed_result = await redis.xautoclaim(
+                            name=config.stream_completion_name,
+                            groupname=config.stream_consumer_group,
+                            consumername=self._consumer_name,
+                            min_idle_time=config.stream_claim_min_idle_ms,
+                            start_id="0-0",
+                            count=10,
+                        )
+                        # xautoclaim returns: (next_start_id, [(id, data), ...], [deleted_ids])
+                        if claimed_result and len(claimed_result) >= 2:
+                            claimed_entries = claimed_result[1]
+                            if claimed_entries:
+                                logger.info(
+                                    f"Claimed {len(claimed_entries)} stale pending messages"
+                                )
+                                for entry_id, data in claimed_entries:
+                                    if not self._running:
+                                        return
+                                    await self._process_entry(redis, entry_id, data)
+                    except Exception as e:
+                        logger.warning(f"XAUTOCLAIM failed (non-fatal): {e}")
+
+                    # Read new messages from the stream
+                    messages = await redis.xreadgroup(
+                        groupname=config.stream_consumer_group,
+                        consumername=self._consumer_name,
+                        streams={config.stream_completion_name: ">"},
+                        block=block_timeout,
+                        count=10,
+                    )
+
+                    if not messages:
+                        continue
+
+                    for stream_name, entries in messages:
+                        for entry_id, data in entries:
+                            if not self._running:
+                                return
+                            await self._process_entry(redis, entry_id, data)
+
+            except asyncio.CancelledError:
+                logger.info("Consumer cancelled")
+                return
+            except Exception as e:
+                retry_count += 1
+                logger.error(
+                    f"Consumer error (retry {retry_count}/{max_retries}): {e}",
+                    exc_info=True,
+                )
+                if self._running and retry_count < max_retries:
+                    await asyncio.sleep(retry_delay)
+                else:
+                    logger.error("Max retries reached, stopping consumer")
+                    return
+
+    async def _process_entry(
+        self, redis: Any, entry_id: str, data: dict[str, Any]
+    ) -> None:
+        """Process a single stream entry and acknowledge it on success.
+
+        Args:
+            redis: Redis client connection
+            entry_id: The stream entry ID
+            data: The entry data dict
+        """
+        try:
+            # Handle the message
+            message_data = data.get("data")
+            if message_data:
+                await self._handle_message(
+                    message_data.encode()
+                    if isinstance(message_data, str)
+                    else message_data
+                )
+
+            # Acknowledge the message after successful processing
+            await redis.xack(
+                config.stream_completion_name,
+                config.stream_consumer_group,
+                entry_id,
+            )
+        except Exception as e:
+            logger.error(
+                f"Error processing completion message {entry_id}: {e}",
+                exc_info=True,
+            )
+            # Message remains in pending state and will be claimed by
+            # XAUTOCLAIM after min_idle_time expires
+
+    async def _handle_message(self, body: bytes) -> None:
+        """Handle a completion message using our own Prisma client."""
+        try:
+            data = orjson.loads(body)
+            message = OperationCompleteMessage(**data)
+        except Exception as e:
+            logger.error(f"Failed to parse completion message: {e}")
+            return
+
+        logger.info(
+            f"[COMPLETION] Received completion for operation {message.operation_id} "
+            f"(task_id={message.task_id}, success={message.success})"
+        )
+
+        # Find task in registry
+        task = await stream_registry.find_task_by_operation_id(message.operation_id)
+        if task is None:
+            task = await stream_registry.get_task(message.task_id)
+
+        if task is None:
+            logger.warning(
+                f"[COMPLETION] Task not found for operation {message.operation_id} "
+                f"(task_id={message.task_id})"
+            )
+            return
+
+        logger.info(
+            f"[COMPLETION] Found task: task_id={task.task_id}, "
+            f"session_id={task.session_id}, tool_call_id={task.tool_call_id}"
+        )
+
+        # Guard against empty task fields
+        if not task.task_id or not task.session_id or not task.tool_call_id:
+            logger.error(
+                f"[COMPLETION] Task has empty critical fields! "
+                f"task_id={task.task_id!r}, session_id={task.session_id!r}, "
+                f"tool_call_id={task.tool_call_id!r}"
+            )
+            return
+
+        if message.success:
+            await self._handle_success(task, message)
+        else:
+            await self._handle_failure(task, message)
+
+    async def _handle_success(
+        self,
+        task: stream_registry.ActiveTask,
+        message: OperationCompleteMessage,
+    ) -> None:
+        """Handle successful operation completion."""
+        prisma = await self._ensure_prisma()
+        await process_operation_success(task, message.result, prisma)
+
+    async def _handle_failure(
+        self,
+        task: stream_registry.ActiveTask,
+        message: OperationCompleteMessage,
+    ) -> None:
+        """Handle failed operation completion."""
+        prisma = await self._ensure_prisma()
+        await process_operation_failure(task, message.error, prisma)
+
+
+# Module-level consumer instance
+_consumer: ChatCompletionConsumer | None = None
+
+
+async def start_completion_consumer() -> None:
+    """Start the global completion consumer."""
+    global _consumer
+    if _consumer is None:
+        _consumer = ChatCompletionConsumer()
+    await _consumer.start()
+
+
+async def stop_completion_consumer() -> None:
+    """Stop the global completion consumer."""
+    global _consumer
+    if _consumer:
+        await _consumer.stop()
+        _consumer = None
+
+
+async def publish_operation_complete(
+    operation_id: str,
+    task_id: str,
+    success: bool,
+    result: dict | str | None = None,
+    error: str | None = None,
+) -> None:
+    """Publish an operation completion message to Redis Streams.
+
+    Args:
+        operation_id: The operation ID that completed.
+        task_id: The task ID associated with the operation.
+        success: Whether the operation succeeded.
+        result: The result data (for success).
+        error: The error message (for failure).
+    """
+    message = OperationCompleteMessage(
+        operation_id=operation_id,
+        task_id=task_id,
+        success=success,
+        result=result,
+        error=error,
+    )
+
+    redis = await get_redis_async()
+    await redis.xadd(
+        config.stream_completion_name,
+        {"data": message.model_dump_json()},
+        maxlen=config.stream_max_length,
+    )
+    logger.info(f"Published completion for operation {operation_id}")

autogpt_platform/backend/backend/api/features/chat/completion_handler.py

@@ -0,0 +1,344 @@
+"""Shared completion handling for operation success and failure.
+
+This module provides common logic for handling operation completion from both:
+- The Redis Streams consumer (completion_consumer.py)
+- The HTTP webhook endpoint (routes.py)
+"""
+
+import logging
+from typing import Any
+
+import orjson
+from prisma import Prisma
+
+from . import service as chat_service
+from . import stream_registry
+from .response_model import StreamError, StreamToolOutputAvailable
+from .tools.models import ErrorResponse
+
+logger = logging.getLogger(__name__)
+
+# Tools that produce agent_json that needs to be saved to library
+AGENT_GENERATION_TOOLS = {"create_agent", "edit_agent"}
+
+# Keys that should be stripped from agent_json when returning in error responses
+SENSITIVE_KEYS = frozenset(
+    {
+        "api_key",
+        "apikey",
+        "api_secret",
+        "password",
+        "secret",
+        "credentials",
+        "credential",
+        "token",
+        "access_token",
+        "refresh_token",
+        "private_key",
+        "privatekey",
+        "auth",
+        "authorization",
+    }
+)
+
+
+def _sanitize_agent_json(obj: Any) -> Any:
+    """Recursively sanitize agent_json by removing sensitive keys.
+
+    Args:
+        obj: The object to sanitize (dict, list, or primitive)
+
+    Returns:
+        Sanitized copy with sensitive keys removed/redacted
+    """
+    if isinstance(obj, dict):
+        return {
+            k: "[REDACTED]" if k.lower() in SENSITIVE_KEYS else _sanitize_agent_json(v)
+            for k, v in obj.items()
+        }
+    elif isinstance(obj, list):
+        return [_sanitize_agent_json(item) for item in obj]
+    else:
+        return obj
+
+
+class ToolMessageUpdateError(Exception):
+    """Raised when updating a tool message in the database fails."""
+
+    pass
+
+
+async def _update_tool_message(
+    session_id: str,
+    tool_call_id: str,
+    content: str,
+    prisma_client: Prisma | None,
+) -> None:
+    """Update tool message in database.
+
+    Args:
+        session_id: The session ID
+        tool_call_id: The tool call ID to update
+        content: The new content for the message
+        prisma_client: Optional Prisma client. If None, uses chat_service.
+
+    Raises:
+        ToolMessageUpdateError: If the database update fails. The caller should
+            handle this to avoid marking the task as completed with inconsistent state.
+    """
+    try:
+        if prisma_client:
+            # Use provided Prisma client (for consumer with its own connection)
+            updated_count = await prisma_client.chatmessage.update_many(
+                where={
+                    "sessionId": session_id,
+                    "toolCallId": tool_call_id,
+                },
+                data={"content": content},
+            )
+            # Check if any rows were updated - 0 means message not found
+            if updated_count == 0:
+                raise ToolMessageUpdateError(
+                    f"No message found with tool_call_id={tool_call_id} in session {session_id}"
+                )
+        else:
+            # Use service function (for webhook endpoint)
+            await chat_service._update_pending_operation(
+                session_id=session_id,
+                tool_call_id=tool_call_id,
+                result=content,
+            )
+    except ToolMessageUpdateError:
+        raise
+    except Exception as e:
+        logger.error(f"[COMPLETION] Failed to update tool message: {e}", exc_info=True)
+        raise ToolMessageUpdateError(
+            f"Failed to update tool message for tool_call_id={tool_call_id}: {e}"
+        ) from e
+
+
+def serialize_result(result: dict | list | str | int | float | bool | None) -> str:
+    """Serialize result to JSON string with sensible defaults.
+
+    Args:
+        result: The result to serialize. Can be a dict, list, string,
+            number, boolean, or None.
+
+    Returns:
+        JSON string representation of the result. Returns '{"status": "completed"}'
+        only when result is explicitly None.
+    """
+    if isinstance(result, str):
+        return result
+    if result is None:
+        return '{"status": "completed"}'
+    return orjson.dumps(result).decode("utf-8")
+
+
+async def _save_agent_from_result(
+    result: dict[str, Any],
+    user_id: str | None,
+    tool_name: str,
+) -> dict[str, Any]:
+    """Save agent to library if result contains agent_json.
+
+    Args:
+        result: The result dict that may contain agent_json
+        user_id: The user ID to save the agent for
+        tool_name: The tool name (create_agent or edit_agent)
+
+    Returns:
+        Updated result dict with saved agent details, or original result if no agent_json
+    """
+    if not user_id:
+        logger.warning("[COMPLETION] Cannot save agent: no user_id in task")
+        return result
+
+    agent_json = result.get("agent_json")
+    if not agent_json:
+        logger.warning(
+            f"[COMPLETION] {tool_name} completed but no agent_json in result"
+        )
+        return result
+
+    try:
+        from .tools.agent_generator import save_agent_to_library
+
+        is_update = tool_name == "edit_agent"
+        created_graph, library_agent = await save_agent_to_library(
+            agent_json, user_id, is_update=is_update
+        )
+
+        logger.info(
+            f"[COMPLETION] Saved agent '{created_graph.name}' to library "
+            f"(graph_id={created_graph.id}, library_agent_id={library_agent.id})"
+        )
+
+        # Return a response similar to AgentSavedResponse
+        return {
+            "type": "agent_saved",
+            "message": f"Agent '{created_graph.name}' has been saved to your library!",
+            "agent_id": created_graph.id,
+            "agent_name": created_graph.name,
+            "library_agent_id": library_agent.id,
+            "library_agent_link": f"/library/agents/{library_agent.id}",
+            "agent_page_link": f"/build?flowID={created_graph.id}",
+        }
+    except Exception as e:
+        logger.error(
+            f"[COMPLETION] Failed to save agent to library: {e}",
+            exc_info=True,
+        )
+        # Return error but don't fail the whole operation
+        # Sanitize agent_json to remove sensitive keys before returning
+        return {
+            "type": "error",
+            "message": f"Agent was generated but failed to save: {str(e)}",
+            "error": str(e),
+            "agent_json": _sanitize_agent_json(agent_json),
+        }
+
+
+async def process_operation_success(
+    task: stream_registry.ActiveTask,
+    result: dict | str | None,
+    prisma_client: Prisma | None = None,
+) -> None:
+    """Handle successful operation completion.
+
+    Publishes the result to the stream registry, updates the database,
+    generates LLM continuation, and marks the task as completed.
+
+    Args:
+        task: The active task that completed
+        result: The result data from the operation
+        prisma_client: Optional Prisma client for database operations.
+            If None, uses chat_service._update_pending_operation instead.
+
+    Raises:
+        ToolMessageUpdateError: If the database update fails. The task will be
+            marked as failed instead of completed to avoid inconsistent state.
+    """
+    # For agent generation tools, save the agent to library
+    if task.tool_name in AGENT_GENERATION_TOOLS and isinstance(result, dict):
+        result = await _save_agent_from_result(result, task.user_id, task.tool_name)
+
+    # Serialize result for output (only substitute default when result is exactly None)
+    result_output = result if result is not None else {"status": "completed"}
+    output_str = (
+        result_output
+        if isinstance(result_output, str)
+        else orjson.dumps(result_output).decode("utf-8")
+    )
+
+    # Publish result to stream registry
+    await stream_registry.publish_chunk(
+        task.task_id,
+        StreamToolOutputAvailable(
+            toolCallId=task.tool_call_id,
+            toolName=task.tool_name,
+            output=output_str,
+            success=True,
+        ),
+    )
+
+    # Update pending operation in database
+    # If this fails, we must not continue to mark the task as completed
+    result_str = serialize_result(result)
+    try:
+        await _update_tool_message(
+            session_id=task.session_id,
+            tool_call_id=task.tool_call_id,
+            content=result_str,
+            prisma_client=prisma_client,
+        )
+    except ToolMessageUpdateError:
+        # DB update failed - mark task as failed to avoid inconsistent state
+        logger.error(
+            f"[COMPLETION] DB update failed for task {task.task_id}, "
+            "marking as failed instead of completed"
+        )
+        await stream_registry.publish_chunk(
+            task.task_id,
+            StreamError(errorText="Failed to save operation result to database"),
+        )
+        await stream_registry.mark_task_completed(task.task_id, status="failed")
+        raise
+
+    # Generate LLM continuation with streaming
+    try:
+        await chat_service._generate_llm_continuation_with_streaming(
+            session_id=task.session_id,
+            user_id=task.user_id,
+            task_id=task.task_id,
+        )
+    except Exception as e:
+        logger.error(
+            f"[COMPLETION] Failed to generate LLM continuation: {e}",
+            exc_info=True,
+        )
+
+    # Mark task as completed and release Redis lock
+    await stream_registry.mark_task_completed(task.task_id, status="completed")
+    try:
+        await chat_service._mark_operation_completed(task.tool_call_id)
+    except Exception as e:
+        logger.error(f"[COMPLETION] Failed to mark operation completed: {e}")
+
+    logger.info(
+        f"[COMPLETION] Successfully processed completion for task {task.task_id}"
+    )
+
+
+async def process_operation_failure(
+    task: stream_registry.ActiveTask,
+    error: str | None,
+    prisma_client: Prisma | None = None,
+) -> None:
+    """Handle failed operation completion.
+
+    Publishes the error to the stream registry, updates the database with
+    the error response, and marks the task as failed.
+
+    Args:
+        task: The active task that failed
+        error: The error message from the operation
+        prisma_client: Optional Prisma client for database operations.
+            If None, uses chat_service._update_pending_operation instead.
+    """
+    error_msg = error or "Operation failed"
+
+    # Publish error to stream registry
+    await stream_registry.publish_chunk(
+        task.task_id,
+        StreamError(errorText=error_msg),
+    )
+
+    # Update pending operation with error
+    # If this fails, we still continue to mark the task as failed
+    error_response = ErrorResponse(
+        message=error_msg,
+        error=error,
+    )
+    try:
+        await _update_tool_message(
+            session_id=task.session_id,
+            tool_call_id=task.tool_call_id,
+            content=error_response.model_dump_json(),
+            prisma_client=prisma_client,
+        )
+    except ToolMessageUpdateError:
+        # DB update failed - log but continue with cleanup
+        logger.error(
+            f"[COMPLETION] DB update failed while processing failure for task {task.task_id}, "
+            "continuing with cleanup"
+        )
+
+    # Mark task as failed and release Redis lock
+    await stream_registry.mark_task_completed(task.task_id, status="failed")
+    try:
+        await chat_service._mark_operation_completed(task.tool_call_id)
+    except Exception as e:
+        logger.error(f"[COMPLETION] Failed to mark operation completed: {e}")
+
+    logger.info(f"[COMPLETION] Processed failure for task {task.task_id}: {error_msg}")

autogpt_platform/backend/backend/api/features/chat/config.py

@@ -27,38 +27,63 @@ class ChatConfig(BaseSettings):
     session_ttl: int = Field(default=43200, description="Session TTL in seconds")
 
     # Streaming Configuration
-    max_retries: int = Field(
-        default=3,
-        description="Max retries for fallback path (SDK handles retries internally)",
+    max_context_messages: int = Field(
+        default=50, ge=1, le=200, description="Maximum context messages"
     )
+
+    stream_timeout: int = Field(default=300, description="Stream timeout in seconds")
+    max_retries: int = Field(default=3, description="Maximum number of retries")
     max_agent_runs: int = Field(default=30, description="Maximum number of agent runs")
     max_agent_schedules: int = Field(
         default=30, description="Maximum number of agent schedules"
     )
 
+    # Long-running operation configuration
+    long_running_operation_ttl: int = Field(
+        default=600,
+        description="TTL in seconds for long-running operation tracking in Redis (safety net if pod dies)",
+    )
+
     # Stream registry configuration for SSE reconnection
     stream_ttl: int = Field(
         default=3600,
         description="TTL in seconds for stream data in Redis (1 hour)",
     )
-    stream_lock_ttl: int = Field(
-        default=120,
-        description="TTL in seconds for stream lock (2 minutes). Short timeout allows "
-        "reconnection after refresh/crash without long waits.",
-    )
     stream_max_length: int = Field(
         default=10000,
         description="Maximum number of messages to store per stream",
     )
 
-    # Redis key prefixes for stream registry
-    session_meta_prefix: str = Field(
-        default="chat:task:meta:",
-        description="Prefix for session metadata hash keys",
+    # Redis Streams configuration for completion consumer
+    stream_completion_name: str = Field(
+        default="chat:completions",
+        description="Redis Stream name for operation completions",
     )
-    turn_stream_prefix: str = Field(
+    stream_consumer_group: str = Field(
+        default="chat_consumers",
+        description="Consumer group name for completion stream",
+    )
+    stream_claim_min_idle_ms: int = Field(
+        default=60000,
+        description="Minimum idle time in milliseconds before claiming pending messages from dead consumers",
+    )
+
+    # Redis key prefixes for stream registry
+    task_meta_prefix: str = Field(
+        default="chat:task:meta:",
+        description="Prefix for task metadata hash keys",
+    )
+    task_stream_prefix: str = Field(
         default="chat:stream:",
-        description="Prefix for turn message stream keys",
+        description="Prefix for task message stream keys",
+    )
+    task_op_prefix: str = Field(
+        default="chat:task:op:",
+        description="Prefix for operation ID to task ID mapping keys",
+    )
+    internal_api_key: str | None = Field(
+        default=None,
+        description="API key for internal webhook callbacks (env: CHAT_INTERNAL_API_KEY)",
     )
 
     # Langfuse Prompt Management Configuration
@@ -68,31 +93,6 @@ class ChatConfig(BaseSettings):
         description="Name of the prompt in Langfuse to fetch",
     )
 
-    # Claude Agent SDK Configuration
-    use_claude_agent_sdk: bool = Field(
-        default=True,
-        description="Use Claude Agent SDK for chat completions",
-    )
-    claude_agent_model: str | None = Field(
-        default=None,
-        description="Model for the Claude Agent SDK path. If None, derives from "
-        "the `model` field by stripping the OpenRouter provider prefix.",
-    )
-    claude_agent_max_buffer_size: int = Field(
-        default=10 * 1024 * 1024,  # 10MB (default SDK is 1MB)
-        description="Max buffer size in bytes for Claude Agent SDK JSON message parsing. "
-        "Increase if tool outputs exceed the limit.",
-    )
-    claude_agent_max_subtasks: int = Field(
-        default=10,
-        description="Max number of sub-agent Tasks the SDK can spawn per session.",
-    )
-    claude_agent_use_resume: bool = Field(
-        default=True,
-        description="Use --resume for multi-turn conversations instead of "
-        "history compression. Falls back to compression when unavailable.",
-    )
-
     # Extended thinking configuration for Claude models
     thinking_enabled: bool = Field(
         default=True,
@@ -130,16 +130,13 @@ class ChatConfig(BaseSettings):
                 v = "https://openrouter.ai/api/v1"
         return v
 
-    @field_validator("use_claude_agent_sdk", mode="before")
+    @field_validator("internal_api_key", mode="before")
     @classmethod
-    def get_use_claude_agent_sdk(cls, v):
-        """Get use_claude_agent_sdk from environment if not provided."""
-        # Check environment variable - default to True if not set
-        env_val = os.getenv("CHAT_USE_CLAUDE_AGENT_SDK", "").lower()
-        if env_val:
-            return env_val in ("true", "1", "yes", "on")
-        # Default to True (SDK enabled by default)
-        return True if v is None else v
+    def get_internal_api_key(cls, v):
+        """Get internal API key from environment if not provided."""
+        if v is None:
+            v = os.getenv("CHAT_INTERNAL_API_KEY")
+        return v
 
     # Prompt paths for different contexts
     PROMPT_PATHS: dict[str, str] = {

autogpt_platform/backend/backend/api/features/chat/db.py

@@ -3,9 +3,8 @@
 import asyncio
 import logging
 from datetime import UTC, datetime
-from typing import Any
+from typing import Any, cast
 
-from prisma.errors import UniqueViolationError
 from prisma.models import ChatMessage as PrismaChatMessage
 from prisma.models import ChatSession as PrismaChatSession
 from prisma.types import (
@@ -15,27 +14,29 @@ from prisma.types import (
     ChatSessionWhereInput,
 )
 
-from backend.data import db
+from backend.data.db import transaction
 from backend.util.json import SafeJson
 
-from .model import ChatMessage, ChatSession, ChatSessionInfo
-
 logger = logging.getLogger(__name__)
 
 
-async def get_chat_session(session_id: str) -> ChatSession | None:
+async def get_chat_session(session_id: str) -> PrismaChatSession | None:
     """Get a chat session by ID from the database."""
     session = await PrismaChatSession.prisma().find_unique(
         where={"id": session_id},
-        include={"Messages": {"order_by": {"sequence": "asc"}}},
+        include={"Messages": True},
     )
-    return ChatSession.from_db(session) if session else None
+    if session and session.Messages:
+        # Sort messages by sequence in Python - Prisma Python client doesn't support
+        # order_by in include clauses (unlike Prisma JS), so we sort after fetching
+        session.Messages.sort(key=lambda m: m.sequence)
+    return session
 
 
 async def create_chat_session(
     session_id: str,
     user_id: str,
-) -> ChatSessionInfo:
+) -> PrismaChatSession:
     """Create a new chat session in the database."""
     data = ChatSessionCreateInput(
         id=session_id,
@@ -44,8 +45,7 @@ async def create_chat_session(
         successfulAgentRuns=SafeJson({}),
         successfulAgentSchedules=SafeJson({}),
     )
-    prisma_session = await PrismaChatSession.prisma().create(data=data)
-    return ChatSessionInfo.from_db(prisma_session)
+    return await PrismaChatSession.prisma().create(data=data)
 
 
 async def update_chat_session(
@@ -56,7 +56,7 @@ async def update_chat_session(
     total_prompt_tokens: int | None = None,
     total_completion_tokens: int | None = None,
     title: str | None = None,
-) -> ChatSession | None:
+) -> PrismaChatSession | None:
     """Update a chat session's metadata."""
     data: ChatSessionUpdateInput = {"updatedAt": datetime.now(UTC)}
 
@@ -76,9 +76,12 @@ async def update_chat_session(
     session = await PrismaChatSession.prisma().update(
         where={"id": session_id},
         data=data,
-        include={"Messages": {"order_by": {"sequence": "asc"}}},
+        include={"Messages": True},
     )
-    return ChatSession.from_db(session) if session else None
+    if session and session.Messages:
+        # Sort in Python - Prisma Python doesn't support order_by in include clauses
+        session.Messages.sort(key=lambda m: m.sequence)
+    return session
 
 
 async def add_chat_message(
@@ -91,11 +94,12 @@ async def add_chat_message(
     refusal: str | None = None,
     tool_calls: list[dict[str, Any]] | None = None,
     function_call: dict[str, Any] | None = None,
-) -> ChatMessage:
+) -> PrismaChatMessage:
     """Add a message to a chat session."""
-    # Build ChatMessageCreateInput with only non-None values
-    # (Prisma TypedDict rejects optional fields set to None)
-    data: ChatMessageCreateInput = {
+    # Build input dict dynamically rather than using ChatMessageCreateInput directly
+    # because Prisma's TypedDict validation rejects optional fields set to None.
+    # We only include fields that have values, then cast at the end.
+    data: dict[str, Any] = {
         "Session": {"connect": {"id": session_id}},
         "role": role,
         "sequence": sequence,
@@ -123,117 +127,81 @@ async def add_chat_message(
             where={"id": session_id},
             data={"updatedAt": datetime.now(UTC)},
         ),
-        PrismaChatMessage.prisma().create(data=data),
+        PrismaChatMessage.prisma().create(data=cast(ChatMessageCreateInput, data)),
     )
-    return ChatMessage.from_db(message)
+    return message
 
 
 async def add_chat_messages_batch(
     session_id: str,
     messages: list[dict[str, Any]],
     start_sequence: int,
-) -> int:
+) -> list[PrismaChatMessage]:
     """Add multiple messages to a chat session in a batch.
 
-    Uses collision detection with retry: tries to create messages starting
-    at start_sequence. If a unique constraint violation occurs (e.g., the
-    streaming loop and long-running callback race), queries the latest
-    sequence and retries with the correct offset. This avoids unnecessary
-    upserts and DB queries in the common case (no collision).
-
-    Returns:
-        Next sequence number for the next message to be inserted. This equals
-        start_sequence + len(messages) and allows callers to update their
-        counters even when collision detection adjusts start_sequence.
+    Uses a transaction for atomicity - if any message creation fails,
+    the entire batch is rolled back.
     """
     if not messages:
-        # No messages to add - return current count
-        return start_sequence
+        return []
 
-    max_retries = 5
-    for attempt in range(max_retries):
-        try:
-            # Single timestamp for all messages and session update
-            now = datetime.now(UTC)
+    created_messages = []
 
-            async with db.transaction() as tx:
-                # Build all message data
-                messages_data = []
-                for i, msg in enumerate(messages):
-                    # Build ChatMessageCreateInput with only non-None values
-                    # (Prisma TypedDict rejects optional fields set to None)
-                    # Note: create_many doesn't support nested creates, use sessionId directly
-                    data: ChatMessageCreateInput = {
-                        "sessionId": session_id,
-                        "role": msg["role"],
-                        "sequence": start_sequence + i,
-                        "createdAt": now,
-                    }
+    async with transaction() as tx:
+        for i, msg in enumerate(messages):
+            # Build input dict dynamically rather than using ChatMessageCreateInput
+            # directly because Prisma's TypedDict validation rejects optional fields
+            # set to None. We only include fields that have values, then cast.
+            data: dict[str, Any] = {
+                "Session": {"connect": {"id": session_id}},
+                "role": msg["role"],
+                "sequence": start_sequence + i,
+            }
 
-                    # Add optional string fields
-                    if msg.get("content") is not None:
-                        data["content"] = msg["content"]
-                    if msg.get("name") is not None:
-                        data["name"] = msg["name"]
-                    if msg.get("tool_call_id") is not None:
-                        data["toolCallId"] = msg["tool_call_id"]
-                    if msg.get("refusal") is not None:
-                        data["refusal"] = msg["refusal"]
+            # Add optional string fields
+            if msg.get("content") is not None:
+                data["content"] = msg["content"]
+            if msg.get("name") is not None:
+                data["name"] = msg["name"]
+            if msg.get("tool_call_id") is not None:
+                data["toolCallId"] = msg["tool_call_id"]
+            if msg.get("refusal") is not None:
+                data["refusal"] = msg["refusal"]
 
-                    # Add optional JSON fields only when they have values
-                    if msg.get("tool_calls") is not None:
-                        data["toolCalls"] = SafeJson(msg["tool_calls"])
-                    if msg.get("function_call") is not None:
-                        data["functionCall"] = SafeJson(msg["function_call"])
+            # Add optional JSON fields only when they have values
+            if msg.get("tool_calls") is not None:
+                data["toolCalls"] = SafeJson(msg["tool_calls"])
+            if msg.get("function_call") is not None:
+                data["functionCall"] = SafeJson(msg["function_call"])
 
-                    messages_data.append(data)
+            created = await PrismaChatMessage.prisma(tx).create(
+                data=cast(ChatMessageCreateInput, data)
+            )
+            created_messages.append(created)
 
-                # Run create_many and session update in parallel within transaction
-                # Both use the same timestamp for consistency
-                await asyncio.gather(
-                    PrismaChatMessage.prisma(tx).create_many(data=messages_data),
-                    PrismaChatSession.prisma(tx).update(
-                        where={"id": session_id},
-                        data={"updatedAt": now},
-                    ),
-                )
+        # Update session's updatedAt timestamp within the same transaction.
+        # Note: Token usage (total_prompt_tokens, total_completion_tokens) is updated
+        # separately via update_chat_session() after streaming completes.
+        await PrismaChatSession.prisma(tx).update(
+            where={"id": session_id},
+            data={"updatedAt": datetime.now(UTC)},
+        )
 
-            # Return next sequence number for counter sync
-            return start_sequence + len(messages)
-
-        except UniqueViolationError:
-            if attempt < max_retries - 1:
-                # Collision detected - query MAX(sequence)+1 and retry with correct offset
-                logger.info(
-                    f"Collision detected for session {session_id} at sequence "
-                    f"{start_sequence}, querying DB for latest sequence"
-                )
-                start_sequence = await get_next_sequence(session_id)
-                logger.info(
-                    f"Retrying batch insert with start_sequence={start_sequence}"
-                )
-                continue
-            else:
-                # Max retries exceeded - propagate error
-                raise
-
-    # Should never reach here due to raise in exception handler
-    raise RuntimeError(f"Failed to insert messages after {max_retries} attempts")
+    return created_messages
 
 
 async def get_user_chat_sessions(
     user_id: str,
     limit: int = 50,
     offset: int = 0,
-) -> list[ChatSessionInfo]:
+) -> list[PrismaChatSession]:
     """Get chat sessions for a user, ordered by most recent."""
-    prisma_sessions = await PrismaChatSession.prisma().find_many(
+    return await PrismaChatSession.prisma().find_many(
         where={"userId": user_id},
         order={"updatedAt": "desc"},
         take=limit,
         skip=offset,
     )
-    return [ChatSessionInfo.from_db(s) for s in prisma_sessions]
 
 
 async def get_user_session_count(user_id: str) -> int:
@@ -272,20 +240,10 @@ async def delete_chat_session(session_id: str, user_id: str | None = None) -> bo
         return False
 
 
-async def get_next_sequence(session_id: str) -> int:
-    """Get the next sequence number for a new message in this session.
-
-    Uses MAX(sequence) + 1 for robustness. Returns 0 if no messages exist.
-    More robust than COUNT(*) because it's immune to deleted messages.
-
-    Optimized to select only the sequence column using raw SQL.
-    The unique index on (sessionId, sequence) makes this query fast.
-    """
-    results = await db.query_raw_with_schema(
-        'SELECT "sequence" FROM {schema_prefix}"ChatMessage" WHERE "sessionId" = $1 ORDER BY "sequence" DESC LIMIT 1',
-        session_id,
-    )
-    return 0 if not results else results[0]["sequence"] + 1
+async def get_chat_session_message_count(session_id: str) -> int:
+    """Get the number of messages in a chat session."""
+    count = await PrismaChatMessage.prisma().count(where={"sessionId": session_id})
+    return count
 
 
 async def update_tool_message_content(

autogpt_platform/backend/backend/api/features/chat/model.py

@@ -2,7 +2,7 @@ import asyncio
 import logging
 import uuid
 from datetime import UTC, datetime
-from typing import Any, Self, cast
+from typing import Any
 from weakref import WeakValueDictionary
 
 from openai.types.chat import (
@@ -23,17 +23,26 @@ from prisma.models import ChatMessage as PrismaChatMessage
 from prisma.models import ChatSession as PrismaChatSession
 from pydantic import BaseModel
 
-from backend.data.db_accessors import chat_db
 from backend.data.redis_client import get_redis_async
 from backend.util import json
 from backend.util.exceptions import DatabaseError, RedisError
 
+from . import db as chat_db
 from .config import ChatConfig
 
 logger = logging.getLogger(__name__)
 config = ChatConfig()
 
 
+def _parse_json_field(value: str | dict | list | None, default: Any = None) -> Any:
+    """Parse a JSON field that may be stored as string or already parsed."""
+    if value is None:
+        return default
+    if isinstance(value, str):
+        return json.loads(value)
+    return value
+
+
 # Redis cache key prefix for chat sessions
 CHAT_SESSION_CACHE_PREFIX = "chat:session:"
 
@@ -43,7 +52,28 @@ def _get_session_cache_key(session_id: str) -> str:
     return f"{CHAT_SESSION_CACHE_PREFIX}{session_id}"
 
 
-# ===================== Chat data models ===================== #
+# Session-level locks to prevent race conditions during concurrent upserts.
+# Uses WeakValueDictionary to automatically garbage collect locks when no longer referenced,
+# preventing unbounded memory growth while maintaining lock semantics for active sessions.
+# Invalidation: Locks are auto-removed by GC when no coroutine holds a reference (after
+# async with lock: completes). Explicit cleanup also occurs in delete_chat_session().
+_session_locks: WeakValueDictionary[str, asyncio.Lock] = WeakValueDictionary()
+_session_locks_mutex = asyncio.Lock()
+
+
+async def _get_session_lock(session_id: str) -> asyncio.Lock:
+    """Get or create a lock for a specific session to prevent concurrent upserts.
+
+    Uses WeakValueDictionary for automatic cleanup: locks are garbage collected
+    when no coroutine holds a reference to them, preventing memory leaks from
+    unbounded growth of session locks.
+    """
+    async with _session_locks_mutex:
+        lock = _session_locks.get(session_id)
+        if lock is None:
+            lock = asyncio.Lock()
+            _session_locks[session_id] = lock
+        return lock
 
 
 class ChatMessage(BaseModel):
@@ -55,19 +85,6 @@ class ChatMessage(BaseModel):
     tool_calls: list[dict] | None = None
     function_call: dict | None = None
 
-    @staticmethod
-    def from_db(prisma_message: PrismaChatMessage) -> "ChatMessage":
-        """Convert a Prisma ChatMessage to a Pydantic ChatMessage."""
-        return ChatMessage(
-            role=prisma_message.role,
-            content=prisma_message.content,
-            name=prisma_message.name,
-            tool_call_id=prisma_message.toolCallId,
-            refusal=prisma_message.refusal,
-            tool_calls=_parse_json_field(prisma_message.toolCalls),
-            function_call=_parse_json_field(prisma_message.functionCall),
-        )
-
 
 class Usage(BaseModel):
     prompt_tokens: int
@@ -75,10 +92,11 @@ class Usage(BaseModel):
     total_tokens: int
 
 
-class ChatSessionInfo(BaseModel):
+class ChatSession(BaseModel):
     session_id: str
     user_id: str
     title: str | None = None
+    messages: list[ChatMessage]
     usage: list[Usage]
     credentials: dict[str, dict] = {}  # Map of provider -> credential metadata
     started_at: datetime
@@ -86,9 +104,40 @@ class ChatSessionInfo(BaseModel):
     successful_agent_runs: dict[str, int] = {}
     successful_agent_schedules: dict[str, int] = {}
 
-    @classmethod
-    def from_db(cls, prisma_session: PrismaChatSession) -> Self:
-        """Convert Prisma ChatSession to Pydantic ChatSession."""
+    @staticmethod
+    def new(user_id: str) -> "ChatSession":
+        return ChatSession(
+            session_id=str(uuid.uuid4()),
+            user_id=user_id,
+            title=None,
+            messages=[],
+            usage=[],
+            credentials={},
+            started_at=datetime.now(UTC),
+            updated_at=datetime.now(UTC),
+        )
+
+    @staticmethod
+    def from_db(
+        prisma_session: PrismaChatSession,
+        prisma_messages: list[PrismaChatMessage] | None = None,
+    ) -> "ChatSession":
+        """Convert Prisma models to Pydantic ChatSession."""
+        messages = []
+        if prisma_messages:
+            for msg in prisma_messages:
+                messages.append(
+                    ChatMessage(
+                        role=msg.role,
+                        content=msg.content,
+                        name=msg.name,
+                        tool_call_id=msg.toolCallId,
+                        refusal=msg.refusal,
+                        tool_calls=_parse_json_field(msg.toolCalls),
+                        function_call=_parse_json_field(msg.functionCall),
+                    )
+                )
+
         # Parse JSON fields from Prisma
         credentials = _parse_json_field(prisma_session.credentials, default={})
         successful_agent_runs = _parse_json_field(
@@ -110,10 +159,11 @@ class ChatSessionInfo(BaseModel):
                 )
             )
 
-        return cls(
+        return ChatSession(
             session_id=prisma_session.id,
             user_id=prisma_session.userId,
             title=prisma_session.title,
+            messages=messages,
             usage=usage,
             credentials=credentials,
             started_at=prisma_session.createdAt,
@@ -122,56 +172,6 @@ class ChatSessionInfo(BaseModel):
             successful_agent_schedules=successful_agent_schedules,
         )
 
-
-class ChatSession(ChatSessionInfo):
-    messages: list[ChatMessage]
-
-    @classmethod
-    def new(cls, user_id: str) -> Self:
-        return cls(
-            session_id=str(uuid.uuid4()),
-            user_id=user_id,
-            title=None,
-            messages=[],
-            usage=[],
-            credentials={},
-            started_at=datetime.now(UTC),
-            updated_at=datetime.now(UTC),
-        )
-
-    @classmethod
-    def from_db(cls, prisma_session: PrismaChatSession) -> Self:
-        """Convert Prisma ChatSession to Pydantic ChatSession."""
-        if prisma_session.Messages is None:
-            raise ValueError(
-                f"Prisma session {prisma_session.id} is missing Messages relation"
-            )
-
-        return cls(
-            **ChatSessionInfo.from_db(prisma_session).model_dump(),
-            messages=[ChatMessage.from_db(m) for m in prisma_session.Messages],
-        )
-
-    def add_tool_call_to_current_turn(self, tool_call: dict) -> None:
-        """Attach a tool_call to the current turn's assistant message.
-
-        Searches backwards for the most recent assistant message (stopping at
-        any user message boundary). If found, appends the tool_call to it.
-        Otherwise creates a new assistant message with the tool_call.
-        """
-        for msg in reversed(self.messages):
-            if msg.role == "user":
-                break
-            if msg.role == "assistant":
-                if not msg.tool_calls:
-                    msg.tool_calls = []
-                msg.tool_calls.append(tool_call)
-                return
-
-        self.messages.append(
-            ChatMessage(role="assistant", content="", tool_calls=[tool_call])
-        )
-
     def to_openai_messages(self) -> list[ChatCompletionMessageParam]:
         messages = []
         for message in self.messages:
@@ -258,72 +258,43 @@ class ChatSession(ChatSessionInfo):
                         name=message.name or "",
                     )
                 )
-        return self._merge_consecutive_assistant_messages(messages)
-
-    @staticmethod
-    def _merge_consecutive_assistant_messages(
-        messages: list[ChatCompletionMessageParam],
-    ) -> list[ChatCompletionMessageParam]:
-        """Merge consecutive assistant messages into single messages.
-
-        Long-running tool flows can create split assistant messages: one with
-        text content and another with tool_calls. Anthropic's API requires
-        tool_result blocks to reference a tool_use in the immediately preceding
-        assistant message, so these splits cause 400 errors via OpenRouter.
-        """
-        if len(messages) < 2:
-            return messages
-
-        result: list[ChatCompletionMessageParam] = [messages[0]]
-        for msg in messages[1:]:
-            prev = result[-1]
-            if prev.get("role") != "assistant" or msg.get("role") != "assistant":
-                result.append(msg)
-                continue
-
-            prev = cast(ChatCompletionAssistantMessageParam, prev)
-            curr = cast(ChatCompletionAssistantMessageParam, msg)
-
-            curr_content = curr.get("content") or ""
-            if curr_content:
-                prev_content = prev.get("content") or ""
-                prev["content"] = (
-                    f"{prev_content}\n{curr_content}" if prev_content else curr_content
-                )
-
-            curr_tool_calls = curr.get("tool_calls")
-            if curr_tool_calls:
-                prev_tool_calls = prev.get("tool_calls")
-                prev["tool_calls"] = (
-                    list(prev_tool_calls) + list(curr_tool_calls)
-                    if prev_tool_calls
-                    else list(curr_tool_calls)
-                )
-        return result
+        return messages
 
 
-def _parse_json_field(value: str | dict | list | None, default: Any = None) -> Any:
-    """Parse a JSON field that may be stored as string or already parsed."""
-    if value is None:
-        return default
-    if isinstance(value, str):
-        return json.loads(value)
-    return value
+async def _get_session_from_cache(session_id: str) -> ChatSession | None:
+    """Get a chat session from Redis cache."""
+    redis_key = _get_session_cache_key(session_id)
+    async_redis = await get_redis_async()
+    raw_session: bytes | None = await async_redis.get(redis_key)
+
+    if raw_session is None:
+        return None
+
+    try:
+        session = ChatSession.model_validate_json(raw_session)
+        logger.info(
+            f"Loading session {session_id} from cache: "
+            f"message_count={len(session.messages)}, "
+            f"roles={[m.role for m in session.messages]}"
+        )
+        return session
+    except Exception as e:
+        logger.error(f"Failed to deserialize session {session_id}: {e}", exc_info=True)
+        raise RedisError(f"Corrupted session data for {session_id}") from e
 
 
-# ================ Chat cache + DB operations ================ #
-
-# NOTE: Database calls are automatically routed through DatabaseManager if Prisma is not
-#       connected directly.
-
-
-async def cache_chat_session(session: ChatSession) -> None:
-    """Cache a chat session in Redis (without persisting to the database)."""
+async def _cache_session(session: ChatSession) -> None:
+    """Cache a chat session in Redis."""
     redis_key = _get_session_cache_key(session.session_id)
     async_redis = await get_redis_async()
     await async_redis.setex(redis_key, config.session_ttl, session.model_dump_json())
 
 
+async def cache_chat_session(session: ChatSession) -> None:
+    """Cache a chat session without persisting to the database."""
+    await _cache_session(session)
+
+
 async def invalidate_session_cache(session_id: str) -> None:
     """Invalidate a chat session from Redis cache.
 
@@ -339,6 +310,80 @@ async def invalidate_session_cache(session_id: str) -> None:
         logger.warning(f"Failed to invalidate session cache for {session_id}: {e}")
 
 
+async def _get_session_from_db(session_id: str) -> ChatSession | None:
+    """Get a chat session from the database."""
+    prisma_session = await chat_db.get_chat_session(session_id)
+    if not prisma_session:
+        return None
+
+    messages = prisma_session.Messages
+    logger.info(
+        f"Loading session {session_id} from DB: "
+        f"has_messages={messages is not None}, "
+        f"message_count={len(messages) if messages else 0}, "
+        f"roles={[m.role for m in messages] if messages else []}"
+    )
+
+    return ChatSession.from_db(prisma_session, messages)
+
+
+async def _save_session_to_db(
+    session: ChatSession, existing_message_count: int
+) -> None:
+    """Save or update a chat session in the database."""
+    # Check if session exists in DB
+    existing = await chat_db.get_chat_session(session.session_id)
+
+    if not existing:
+        # Create new session
+        await chat_db.create_chat_session(
+            session_id=session.session_id,
+            user_id=session.user_id,
+        )
+        existing_message_count = 0
+
+    # Calculate total tokens from usage
+    total_prompt = sum(u.prompt_tokens for u in session.usage)
+    total_completion = sum(u.completion_tokens for u in session.usage)
+
+    # Update session metadata
+    await chat_db.update_chat_session(
+        session_id=session.session_id,
+        credentials=session.credentials,
+        successful_agent_runs=session.successful_agent_runs,
+        successful_agent_schedules=session.successful_agent_schedules,
+        total_prompt_tokens=total_prompt,
+        total_completion_tokens=total_completion,
+    )
+
+    # Add new messages (only those after existing count)
+    new_messages = session.messages[existing_message_count:]
+    if new_messages:
+        messages_data = []
+        for msg in new_messages:
+            messages_data.append(
+                {
+                    "role": msg.role,
+                    "content": msg.content,
+                    "name": msg.name,
+                    "tool_call_id": msg.tool_call_id,
+                    "refusal": msg.refusal,
+                    "tool_calls": msg.tool_calls,
+                    "function_call": msg.function_call,
+                }
+            )
+        logger.info(
+            f"Saving {len(new_messages)} new messages to DB for session {session.session_id}: "
+            f"roles={[m['role'] for m in messages_data]}, "
+            f"start_sequence={existing_message_count}"
+        )
+        await chat_db.add_chat_messages_batch(
+            session_id=session.session_id,
+            messages=messages_data,
+            start_sequence=existing_message_count,
+        )
+
+
 async def get_chat_session(
     session_id: str,
     user_id: str | None = None,
@@ -370,7 +415,7 @@ async def get_chat_session(
         logger.warning(f"Unexpected cache error for session {session_id}: {e}")
 
     # Fall back to database
-    logger.debug(f"Session {session_id} not in cache, checking database")
+    logger.info(f"Session {session_id} not in cache, checking database")
     session = await _get_session_from_db(session_id)
 
     if session is None:
@@ -386,7 +431,7 @@ async def get_chat_session(
 
     # Cache the session from DB
     try:
-        await cache_chat_session(session)
+        await _cache_session(session)
         logger.info(f"Cached session {session_id} from database")
     except Exception as e:
         logger.warning(f"Failed to cache session {session_id}: {e}")
@@ -394,44 +439,6 @@ async def get_chat_session(
     return session
 
 
-async def _get_session_from_cache(session_id: str) -> ChatSession | None:
-    """Get a chat session from Redis cache."""
-    redis_key = _get_session_cache_key(session_id)
-    async_redis = await get_redis_async()
-    raw_session: bytes | None = await async_redis.get(redis_key)
-
-    if raw_session is None:
-        return None
-
-    try:
-        session = ChatSession.model_validate_json(raw_session)
-        logger.info(
-            f"Loading session {session_id} from cache: "
-            f"message_count={len(session.messages)}, "
-            f"roles={[m.role for m in session.messages]}"
-        )
-        return session
-    except Exception as e:
-        logger.error(f"Failed to deserialize session {session_id}: {e}", exc_info=True)
-        raise RedisError(f"Corrupted session data for {session_id}") from e
-
-
-async def _get_session_from_db(session_id: str) -> ChatSession | None:
-    """Get a chat session from the database."""
-    session = await chat_db().get_chat_session(session_id)
-    if not session:
-        return None
-
-    logger.info(
-        f"Loaded session {session_id} from DB: "
-        f"has_messages={bool(session.messages)}, "
-        f"message_count={len(session.messages)}, "
-        f"roles={[m.role for m in session.messages]}"
-    )
-
-    return session
-
-
 async def upsert_chat_session(
     session: ChatSession,
 ) -> ChatSession:
@@ -451,18 +458,16 @@ async def upsert_chat_session(
     lock = await _get_session_lock(session.session_id)
 
     async with lock:
-        # Always query DB for existing message count to ensure consistency
-        existing_message_count = await chat_db().get_next_sequence(session.session_id)
+        # Get existing message count from DB for incremental saves
+        existing_message_count = await chat_db.get_chat_session_message_count(
+            session.session_id
+        )
 
         db_error: Exception | None = None
 
         # Save to database (primary storage)
         try:
-            await _save_session_to_db(
-                session,
-                existing_message_count,
-                skip_existence_check=existing_message_count > 0,
-            )
+            await _save_session_to_db(session, existing_message_count)
         except Exception as e:
             logger.error(
                 f"Failed to save session {session.session_id} to database: {e}"
@@ -471,7 +476,7 @@ async def upsert_chat_session(
 
         # Save to cache (best-effort, even if DB failed)
         try:
-            await cache_chat_session(session)
+            await _cache_session(session)
         except Exception as e:
             # If DB succeeded but cache failed, raise cache error
             if db_error is None:
@@ -492,107 +497,6 @@ async def upsert_chat_session(
         return session
 
 
-async def _save_session_to_db(
-    session: ChatSession,
-    existing_message_count: int,
-    *,
-    skip_existence_check: bool = False,
-) -> None:
-    """Save or update a chat session in the database.
-
-    Args:
-        skip_existence_check: When True, skip the ``get_chat_session`` query
-            and assume the session row already exists.  Saves one DB round trip
-            for incremental saves during streaming.
-    """
-    db = chat_db()
-
-    if not skip_existence_check:
-        # Check if session exists in DB
-        existing = await db.get_chat_session(session.session_id)
-
-        if not existing:
-            # Create new session
-            await db.create_chat_session(
-                session_id=session.session_id,
-                user_id=session.user_id,
-            )
-            existing_message_count = 0
-
-    # Calculate total tokens from usage
-    total_prompt = sum(u.prompt_tokens for u in session.usage)
-    total_completion = sum(u.completion_tokens for u in session.usage)
-
-    # Update session metadata
-    await db.update_chat_session(
-        session_id=session.session_id,
-        credentials=session.credentials,
-        successful_agent_runs=session.successful_agent_runs,
-        successful_agent_schedules=session.successful_agent_schedules,
-        total_prompt_tokens=total_prompt,
-        total_completion_tokens=total_completion,
-    )
-
-    # Add new messages (only those after existing count)
-    new_messages = session.messages[existing_message_count:]
-    if new_messages:
-        messages_data = []
-        for msg in new_messages:
-            messages_data.append(
-                {
-                    "role": msg.role,
-                    "content": msg.content,
-                    "name": msg.name,
-                    "tool_call_id": msg.tool_call_id,
-                    "refusal": msg.refusal,
-                    "tool_calls": msg.tool_calls,
-                    "function_call": msg.function_call,
-                }
-            )
-        logger.info(
-            f"Saving {len(new_messages)} new messages to DB for session {session.session_id}: "
-            f"roles={[m['role'] for m in messages_data]}, "
-            f"start_sequence={existing_message_count}"
-        )
-        await db.add_chat_messages_batch(
-            session_id=session.session_id,
-            messages=messages_data,
-            start_sequence=existing_message_count,
-        )
-
-
-async def append_and_save_message(session_id: str, message: ChatMessage) -> ChatSession:
-    """Atomically append a message to a session and persist it.
-
-    Acquires the session lock, re-fetches the latest session state,
-    appends the message, and saves — preventing message loss when
-    concurrent requests modify the same session.
-    """
-    lock = await _get_session_lock(session_id)
-
-    async with lock:
-        session = await get_chat_session(session_id)
-        if session is None:
-            raise ValueError(f"Session {session_id} not found")
-
-        session.messages.append(message)
-        existing_message_count = await chat_db().get_next_sequence(session_id)
-
-        try:
-            await _save_session_to_db(session, existing_message_count)
-        except Exception as e:
-            raise DatabaseError(
-                f"Failed to persist message to session {session_id}"
-            ) from e
-
-        try:
-            await cache_chat_session(session)
-        except Exception as e:
-            logger.warning(f"Cache write failed for session {session_id}: {e}")
-
-        return session
-
-
 async def create_chat_session(user_id: str) -> ChatSession:
     """Create a new chat session and persist it.
 
@@ -605,7 +509,7 @@ async def create_chat_session(user_id: str) -> ChatSession:
 
     # Create in database first - fail fast if this fails
     try:
-        await chat_db().create_chat_session(
+        await chat_db.create_chat_session(
             session_id=session.session_id,
             user_id=user_id,
         )
@@ -617,7 +521,7 @@ async def create_chat_session(user_id: str) -> ChatSession:
 
     # Cache the session (best-effort optimization, DB is source of truth)
     try:
-        await cache_chat_session(session)
+        await _cache_session(session)
     except Exception as e:
         logger.warning(f"Failed to cache new session {session.session_id}: {e}")
 
@@ -628,16 +532,20 @@ async def get_user_sessions(
     user_id: str,
     limit: int = 50,
     offset: int = 0,
-) -> tuple[list[ChatSessionInfo], int]:
+) -> tuple[list[ChatSession], int]:
     """Get chat sessions for a user from the database with total count.
 
     Returns:
         A tuple of (sessions, total_count) where total_count is the overall
         number of sessions for the user (not just the current page).
     """
-    db = chat_db()
-    sessions = await db.get_user_chat_sessions(user_id, limit, offset)
-    total_count = await db.get_user_session_count(user_id)
+    prisma_sessions = await chat_db.get_user_chat_sessions(user_id, limit, offset)
+    total_count = await chat_db.get_user_session_count(user_id)
+
+    sessions = []
+    for prisma_session in prisma_sessions:
+        # Convert without messages for listing (lighter weight)
+        sessions.append(ChatSession.from_db(prisma_session, None))
 
     return sessions, total_count
 
@@ -655,7 +563,7 @@ async def delete_chat_session(session_id: str, user_id: str | None = None) -> bo
     """
     # Delete from database first (with optional user_id validation)
     # This confirms ownership before invalidating cache
-    deleted = await chat_db().delete_chat_session(session_id, user_id)
+    deleted = await chat_db.delete_chat_session(session_id, user_id)
 
     if not deleted:
         return False
@@ -690,52 +598,20 @@ async def update_session_title(session_id: str, title: str) -> bool:
         True if updated successfully, False otherwise.
     """
     try:
-        result = await chat_db().update_chat_session(session_id=session_id, title=title)
+        result = await chat_db.update_chat_session(session_id=session_id, title=title)
         if result is None:
             logger.warning(f"Session {session_id} not found for title update")
             return False
 
-        # Update title in cache if it exists (instead of invalidating).
-        # This prevents race conditions where cache invalidation causes
-        # the frontend to see stale DB data while streaming is still in progress.
+        # Invalidate cache so next fetch gets updated title
         try:
-            cached = await _get_session_from_cache(session_id)
-            if cached:
-                cached.title = title
-                await cache_chat_session(cached)
+            redis_key = _get_session_cache_key(session_id)
+            async_redis = await get_redis_async()
+            await async_redis.delete(redis_key)
         except Exception as e:
-            # Not critical - title will be correct on next full cache refresh
-            logger.warning(
-                f"Failed to update title in cache for session {session_id}: {e}"
-            )
+            logger.warning(f"Failed to invalidate cache for session {session_id}: {e}")
 
         return True
     except Exception as e:
         logger.error(f"Failed to update title for session {session_id}: {e}")
         return False
-
-
-# ==================== Chat session locks ==================== #
-
-_session_locks: WeakValueDictionary[str, asyncio.Lock] = WeakValueDictionary()
-_session_locks_mutex = asyncio.Lock()
-
-
-async def _get_session_lock(session_id: str) -> asyncio.Lock:
-    """Get or create a lock for a specific session to prevent concurrent upserts.
-
-    This was originally added to solve the specific problem of race conditions between
-    the session title thread and the conversation thread, which always occurs on the
-    same instance as we prevent rapid request sends on the frontend.
-
-    Uses WeakValueDictionary for automatic cleanup: locks are garbage collected
-    when no coroutine holds a reference to them, preventing memory leaks from
-    unbounded growth of session locks. Explicit cleanup also occurs
-    in `delete_chat_session()`.
-    """
-    async with _session_locks_mutex:
-        lock = _session_locks.get(session_id)
-        if lock is None:
-            lock = asyncio.Lock()
-            _session_locks[session_id] = lock
-        return lock

autogpt_platform/backend/backend/api/features/chat/model_test.py

@@ -0,0 +1,119 @@
+import pytest
+
+from .model import (
+    ChatMessage,
+    ChatSession,
+    Usage,
+    get_chat_session,
+    upsert_chat_session,
+)
+
+messages = [
+    ChatMessage(content="Hello, how are you?", role="user"),
+    ChatMessage(
+        content="I'm fine, thank you!",
+        role="assistant",
+        tool_calls=[
+            {
+                "id": "t123",
+                "type": "function",
+                "function": {
+                    "name": "get_weather",
+                    "arguments": '{"city": "New York"}',
+                },
+            }
+        ],
+    ),
+    ChatMessage(
+        content="I'm using the tool to get the weather",
+        role="tool",
+        tool_call_id="t123",
+    ),
+]
+
+
+@pytest.mark.asyncio(loop_scope="session")
+async def test_chatsession_serialization_deserialization():
+    s = ChatSession.new(user_id="abc123")
+    s.messages = messages
+    s.usage = [Usage(prompt_tokens=100, completion_tokens=200, total_tokens=300)]
+    serialized = s.model_dump_json()
+    s2 = ChatSession.model_validate_json(serialized)
+    assert s2.model_dump() == s.model_dump()
+
+
+@pytest.mark.asyncio(loop_scope="session")
+async def test_chatsession_redis_storage(setup_test_user, test_user_id):
+
+    s = ChatSession.new(user_id=test_user_id)
+    s.messages = messages
+
+    s = await upsert_chat_session(s)
+
+    s2 = await get_chat_session(
+        session_id=s.session_id,
+        user_id=s.user_id,
+    )
+
+    assert s2 == s
+
+
+@pytest.mark.asyncio(loop_scope="session")
+async def test_chatsession_redis_storage_user_id_mismatch(
+    setup_test_user, test_user_id
+):
+
+    s = ChatSession.new(user_id=test_user_id)
+    s.messages = messages
+    s = await upsert_chat_session(s)
+
+    s2 = await get_chat_session(s.session_id, "different_user_id")
+
+    assert s2 is None
+
+
+@pytest.mark.asyncio(loop_scope="session")
+async def test_chatsession_db_storage(setup_test_user, test_user_id):
+    """Test that messages are correctly saved to and loaded from DB (not cache)."""
+    from backend.data.redis_client import get_redis_async
+
+    # Create session with messages including assistant message
+    s = ChatSession.new(user_id=test_user_id)
+    s.messages = messages  # Contains user, assistant, and tool messages
+    assert s.session_id is not None, "Session id is not set"
+    # Upsert to save to both cache and DB
+    s = await upsert_chat_session(s)
+
+    # Clear the Redis cache to force DB load
+    redis_key = f"chat:session:{s.session_id}"
+    async_redis = await get_redis_async()
+    await async_redis.delete(redis_key)
+
+    # Load from DB (cache was cleared)
+    s2 = await get_chat_session(
+        session_id=s.session_id,
+        user_id=s.user_id,
+    )
+
+    assert s2 is not None, "Session not found after loading from DB"
+    assert len(s2.messages) == len(
+        s.messages
+    ), f"Message count mismatch: expected {len(s.messages)}, got {len(s2.messages)}"
+
+    # Verify all roles are present
+    roles = [m.role for m in s2.messages]
+    assert "user" in roles, f"User message missing. Roles found: {roles}"
+    assert "assistant" in roles, f"Assistant message missing. Roles found: {roles}"
+    assert "tool" in roles, f"Tool message missing. Roles found: {roles}"
+
+    # Verify message content
+    for orig, loaded in zip(s.messages, s2.messages):
+        assert orig.role == loaded.role, f"Role mismatch: {orig.role} != {loaded.role}"
+        assert (
+            orig.content == loaded.content
+        ), f"Content mismatch for {orig.role}: {orig.content} != {loaded.content}"
+        if orig.tool_calls:
+            assert (
+                loaded.tool_calls is not None
+            ), f"Tool calls missing for {orig.role} message"
+            assert len(orig.tool_calls) == len(loaded.tool_calls)

autogpt_platform/backend/backend/api/features/chat/response_model.py

@@ -5,17 +5,11 @@ This module implements the AI SDK UI Stream Protocol (v1) for streaming chat res
 See: https://ai-sdk.dev/docs/ai-sdk-ui/stream-protocol
 """
 
-import json
-import logging
 from enum import Enum
 from typing import Any
 
 from pydantic import BaseModel, Field
 
-from backend.util.json import dumps as json_dumps
-
-logger = logging.getLogger(__name__)
-
 
 class ResponseType(str, Enum):
     """Types of streaming responses following AI SDK protocol."""
@@ -51,8 +45,7 @@ class StreamBaseResponse(BaseModel):
 
     def to_sse(self) -> str:
         """Convert to SSE format."""
-        json_str = self.model_dump_json(exclude_none=True)
-        return f"data: {json_str}\n\n"
+        return f"data: {self.model_dump_json()}\n\n"
 
 
 # ========== Message Lifecycle ==========
@@ -63,13 +56,15 @@ class StreamStart(StreamBaseResponse):
 
     type: ResponseType = ResponseType.START
     messageId: str = Field(..., description="Unique message ID")
-    sessionId: str | None = Field(
+    taskId: str | None = Field(
         default=None,
-        description="Session ID for SSE reconnection.",
+        description="Task ID for SSE reconnection. Clients can reconnect using GET /tasks/{taskId}/stream",
     )
 
     def to_sse(self) -> str:
-        """Convert to SSE format, excluding non-protocol fields like sessionId."""
+        """Convert to SSE format, excluding non-protocol fields like taskId."""
+        import json
+
         data: dict[str, Any] = {
             "type": self.type.value,
             "messageId": self.messageId,
@@ -166,6 +161,8 @@ class StreamToolOutputAvailable(StreamBaseResponse):
 
     def to_sse(self) -> str:
         """Convert to SSE format, excluding non-spec fields."""
+        import json
+
         data = {
             "type": self.type.value,
             "toolCallId": self.toolCallId,
@@ -196,18 +193,6 @@ class StreamError(StreamBaseResponse):
         default=None, description="Additional error details"
     )
 
-    def to_sse(self) -> str:
-        """Convert to SSE format, only emitting fields required by AI SDK protocol.
-
-        The AI SDK uses z.strictObject({type, errorText}) which rejects
-        any extra fields like `code` or `details`.
-        """
-        data = {
-            "type": self.type.value,
-            "errorText": self.errorText,
-        }
-        return f"data: {json_dumps(data)}\n\n"
-
 
 class StreamHeartbeat(StreamBaseResponse):
     """Heartbeat to keep SSE connection alive during long-running operations.

autogpt_platform/backend/backend/api/features/chat/routes.py

@@ -1,37 +1,29 @@
 """Chat API routes for chat session management and streaming via SSE."""
 
-import asyncio
 import logging
+import uuid as uuid_module
 from collections.abc import AsyncGenerator
 from typing import Annotated
-from uuid import uuid4
 
 from autogpt_libs import auth
-from fastapi import APIRouter, Depends, HTTPException, Query, Response, Security
+from fastapi import APIRouter, Depends, Header, HTTPException, Query, Response, Security
 from fastapi.responses import StreamingResponse
 from pydantic import BaseModel
 
-from backend.copilot import service as chat_service
-from backend.copilot import stream_registry
-from backend.copilot.config import ChatConfig
-from backend.copilot.executor.utils import enqueue_cancel_task, enqueue_copilot_task
-from backend.copilot.model import (
-    ChatMessage,
-    ChatSession,
-    append_and_save_message,
-    create_chat_session,
-    delete_chat_session,
-    get_chat_session,
-    get_user_sessions,
-)
-from backend.copilot.response_model import StreamError, StreamFinish, StreamHeartbeat
-from backend.copilot.tools.models import (
+from backend.util.exceptions import NotFoundError
+
+from . import service as chat_service
+from . import stream_registry
+from .completion_handler import process_operation_failure, process_operation_success
+from .config import ChatConfig
+from .model import ChatSession, create_chat_session, get_chat_session, get_user_sessions
+from .response_model import StreamFinish, StreamHeartbeat
+from .tools.models import (
     AgentDetailsResponse,
     AgentOutputResponse,
     AgentPreviewResponse,
     AgentSavedResponse,
     AgentsFoundResponse,
-    BlockDetailsResponse,
     BlockListResponse,
     BlockOutputResponse,
     ClarificationNeededResponse,
@@ -42,12 +34,12 @@ from backend.copilot.tools.models import (
     InputValidationErrorResponse,
     NeedLoginResponse,
     NoResultsResponse,
+    OperationInProgressResponse,
+    OperationPendingResponse,
+    OperationStartedResponse,
     SetupRequirementsResponse,
-    SuggestedGoalResponse,
     UnderstandingUpdatedResponse,
 )
-from backend.copilot.tracking import track_user_message
-from backend.util.exceptions import NotFoundError
 
 config = ChatConfig()
 
@@ -92,8 +84,10 @@ class CreateSessionResponse(BaseModel):
 class ActiveStreamInfo(BaseModel):
     """Information about an active stream for reconnection."""
 
-    turn_id: str
+    task_id: str
     last_message_id: str  # Redis Stream message ID for resumption
+    operation_id: str  # Operation ID for completion tracking
+    tool_name: str  # Name of the tool being executed
 
 
 class SessionDetailResponse(BaseModel):
@@ -123,11 +117,12 @@ class ListSessionsResponse(BaseModel):
     total: int
 
 
-class CancelTaskResponse(BaseModel):
-    """Response model for the cancel task endpoint."""
+class OperationCompleteRequest(BaseModel):
+    """Request model for external completion webhook."""
 
-    cancelled: bool
-    reason: str | None = None
+    success: bool
+    result: dict | str | None = None
+    error: str | None = None
 
 
 # ========== Routes ==========
@@ -204,43 +199,6 @@ async def create_session(
     )
 
 
-@router.delete(
-    "/sessions/{session_id}",
-    dependencies=[Security(auth.requires_user)],
-    status_code=204,
-    responses={404: {"description": "Session not found or access denied"}},
-)
-async def delete_session(
-    session_id: str,
-    user_id: Annotated[str, Security(auth.get_user_id)],
-) -> Response:
-    """
-    Delete a chat session.
-
-    Permanently removes a chat session and all its messages.
-    Only the owner can delete their sessions.
-
-    Args:
-        session_id: The session ID to delete.
-        user_id: The authenticated user's ID.
-
-    Returns:
-        204 No Content on success.
-
-    Raises:
-        HTTPException: 404 if session not found or not owned by user.
-    """
-    deleted = await delete_chat_session(session_id, user_id)
-
-    if not deleted:
-        raise HTTPException(
-            status_code=404,
-            detail=f"Session {session_id} not found or access denied",
-        )
-
-    return Response(status_code=204)
-
-
 @router.get(
     "/sessions/{session_id}",
 )
@@ -252,7 +210,7 @@ async def get_session(
     Retrieve the details of a specific chat session.
 
     Looks up a chat session by ID for the given user (if authenticated) and returns all session data including messages.
-    If there's an active stream for this session, returns active_stream info for reconnection.
+    If there's an active stream for this session, returns the task_id for reconnection.
 
     Args:
         session_id: The unique identifier for the desired chat session.
@@ -270,21 +228,24 @@ async def get_session(
 
     # Check if there's an active stream for this session
     active_stream_info = None
-    active_session, last_message_id = await stream_registry.get_active_session(
+    active_task, last_message_id = await stream_registry.get_active_task_for_session(
         session_id, user_id
     )
-    logger.info(
-        f"[GET_SESSION] session={session_id}, active_session={active_session is not None}, "
-        f"msg_count={len(messages)}, last_role={messages[-1].get('role') if messages else 'none'}"
-    )
-    if active_session:
-        # Keep the assistant message (including tool_calls) so the frontend can
-        # render the correct tool UI (e.g. CreateAgent with mini game).
-        # convertChatSessionToUiMessages handles isComplete=false by setting
-        # tool parts without output to state "input-available".
+    if active_task:
+        # Filter out the in-progress assistant message from the session response.
+        # The client will receive the complete assistant response through the SSE
+        # stream replay instead, preventing duplicate content.
+        if messages and messages[-1].get("role") == "assistant":
+            messages = messages[:-1]
+
+        # Use "0-0" as last_message_id to replay the stream from the beginning.
+        # Since we filtered out the cached assistant message, the client needs
+        # the full stream to reconstruct the response.
         active_stream_info = ActiveStreamInfo(
-            turn_id=active_session.turn_id,
-            last_message_id=last_message_id,
+            task_id=active_task.task_id,
+            last_message_id="0-0",
+            operation_id=active_task.operation_id,
+            tool_name=active_task.tool_name,
         )
 
     return SessionDetailResponse(
@@ -297,50 +258,6 @@ async def get_session(
     )
 
 
-@router.post(
-    "/sessions/{session_id}/cancel",
-    status_code=200,
-)
-async def cancel_session_task(
-    session_id: str,
-    user_id: Annotated[str | None, Depends(auth.get_user_id)],
-) -> CancelTaskResponse:
-    """Cancel the active streaming task for a session.
-
-    Publishes a cancel event to the executor via RabbitMQ FANOUT, then
-    polls Redis until the task status flips from ``running`` or a timeout
-    (5 s) is reached.  Returns only after the cancellation is confirmed.
-    """
-    await _validate_and_get_session(session_id, user_id)
-
-    active_session, _ = await stream_registry.get_active_session(session_id, user_id)
-    if not active_session:
-        return CancelTaskResponse(cancelled=False, reason="no_active_task")
-
-    await enqueue_cancel_task(session_id)
-    logger.info(f"[CANCEL] Published cancel for session ...{session_id[-8:]}")
-
-    # Poll until the executor confirms the task is no longer running.
-    poll_interval = 0.5
-    max_wait = 5.0
-    waited = 0.0
-    while waited < max_wait:
-        await asyncio.sleep(poll_interval)
-        waited += poll_interval
-        task = await stream_registry.get_session(session_id)
-        if task is None or task.status != "running":
-            logger.info(
-                f"[CANCEL] Session ...{session_id[-8:]} confirmed stopped "
-                f"(status={task.status if task else 'gone'}) after {waited:.1f}s"
-            )
-            return CancelTaskResponse(cancelled=True)
-
-    logger.warning(
-        f"[CANCEL] Session ...{session_id[-8:]} not confirmed after {max_wait}s"
-    )
-    return CancelTaskResponse(cancelled=True, reason="cancel_published_not_confirmed")
-
-
 @router.post(
     "/sessions/{session_id}/stream",
 )
@@ -358,15 +275,16 @@ async def stream_chat_post(
       - Tool execution results
 
     The AI generation runs in a background task that continues even if the client disconnects.
-    All chunks are written to a per-turn Redis stream for reconnection support. If the client
-    disconnects, they can reconnect using GET /sessions/{session_id}/stream to resume.
+    All chunks are written to Redis for reconnection support. If the client disconnects,
+    they can reconnect using GET /tasks/{task_id}/stream to resume from where they left off.
 
     Args:
         session_id: The chat session identifier to associate with the streamed messages.
         request: Request body containing message, is_user_message, and optional context.
         user_id: Optional authenticated user ID.
     Returns:
-        StreamingResponse: SSE-formatted response chunks.
+        StreamingResponse: SSE-formatted response chunks. First chunk is a "start" event
+        containing the task_id for reconnection.
 
     """
     import asyncio
@@ -382,9 +300,10 @@ async def stream_chat_post(
         f"user={user_id}, message_len={len(request.message)}",
         extra={"json_fields": log_meta},
     )
-    await _validate_and_get_session(session_id, user_id)
+
+    session = await _validate_and_get_session(session_id, user_id)
     logger.info(
-        f"[TIMING] session validated in {(time.perf_counter() - stream_start_time) * 1000:.1f}ms",
+        f"[TIMING] session validated in {(time.perf_counter() - stream_start_time)*1000:.1f}ms",
         extra={
             "json_fields": {
                 **log_meta,
@@ -393,39 +312,22 @@ async def stream_chat_post(
         },
     )
 
-    # Atomically append user message to session BEFORE creating task to avoid
-    # race condition where GET_SESSION sees task as "running" but message isn't
-    # saved yet.  append_and_save_message re-fetches inside a lock to prevent
-    # message loss from concurrent requests.
-    if request.message:
-        message = ChatMessage(
-            role="user" if request.is_user_message else "assistant",
-            content=request.message,
-        )
-        if request.is_user_message:
-            track_user_message(
-                user_id=user_id,
-                session_id=session_id,
-                message_length=len(request.message),
-            )
-        logger.info(f"[STREAM] Saving user message to session {session_id}")
-        await append_and_save_message(session_id, message)
-        logger.info(f"[STREAM] User message saved for session {session_id}")
-
     # Create a task in the stream registry for reconnection support
-    turn_id = str(uuid4())
-    log_meta["turn_id"] = turn_id
+    task_id = str(uuid_module.uuid4())
+    operation_id = str(uuid_module.uuid4())
+    log_meta["task_id"] = task_id
 
     task_create_start = time.perf_counter()
-    await stream_registry.create_session_task(
+    await stream_registry.create_task(
+        task_id=task_id,
         session_id=session_id,
         user_id=user_id,
-        tool_call_id="chat_stream",
+        tool_call_id="chat_stream",  # Not a tool call, but needed for the model
         tool_name="chat",
-        turn_id=turn_id,
+        operation_id=operation_id,
     )
     logger.info(
-        f"[TIMING] create_session_task completed in {(time.perf_counter() - task_create_start) * 1000:.1f}ms",
+        f"[TIMING] create_task completed in {(time.perf_counter() - task_create_start)*1000:.1f}ms",
         extra={
             "json_fields": {
                 **log_meta,
@@ -434,21 +336,82 @@ async def stream_chat_post(
         },
     )
 
-    # Per-turn stream is always fresh (unique turn_id), subscribe from beginning
-    subscribe_from_id = "0-0"
+    # Background task that runs the AI generation independently of SSE connection
+    async def run_ai_generation():
+        import time as time_module
 
-    await enqueue_copilot_task(
-        session_id=session_id,
-        user_id=user_id,
-        message=request.message,
-        turn_id=turn_id,
-        is_user_message=request.is_user_message,
-        context=request.context,
-    )
+        gen_start_time = time_module.perf_counter()
+        logger.info(
+            f"[TIMING] run_ai_generation STARTED, task={task_id}, session={session_id}, user={user_id}",
+            extra={"json_fields": log_meta},
+        )
+        first_chunk_time, ttfc = None, None
+        chunk_count = 0
+        try:
+            async for chunk in chat_service.stream_chat_completion(
+                session_id,
+                request.message,
+                is_user_message=request.is_user_message,
+                user_id=user_id,
+                session=session,  # Pass pre-fetched session to avoid double-fetch
+                context=request.context,
+                _task_id=task_id,  # Pass task_id so service emits start with taskId for reconnection
+            ):
+                chunk_count += 1
+                if first_chunk_time is None:
+                    first_chunk_time = time_module.perf_counter()
+                    ttfc = first_chunk_time - gen_start_time
+                    logger.info(
+                        f"[TIMING] FIRST AI CHUNK at {ttfc:.2f}s, type={type(chunk).__name__}",
+                        extra={
+                            "json_fields": {
+                                **log_meta,
+                                "chunk_type": type(chunk).__name__,
+                                "time_to_first_chunk_ms": ttfc * 1000,
+                            }
+                        },
+                    )
+                # Write to Redis (subscribers will receive via XREAD)
+                await stream_registry.publish_chunk(task_id, chunk)
 
+            gen_end_time = time_module.perf_counter()
+            total_time = (gen_end_time - gen_start_time) * 1000
+            logger.info(
+                f"[TIMING] run_ai_generation FINISHED in {total_time/1000:.1f}s; "
+                f"task={task_id}, session={session_id}, "
+                f"ttfc={ttfc or -1:.2f}s, n_chunks={chunk_count}",
+                extra={
+                    "json_fields": {
+                        **log_meta,
+                        "total_time_ms": total_time,
+                        "time_to_first_chunk_ms": (
+                            ttfc * 1000 if ttfc is not None else None
+                        ),
+                        "n_chunks": chunk_count,
+                    }
+                },
+            )
+            await stream_registry.mark_task_completed(task_id, "completed")
+        except Exception as e:
+            elapsed = time_module.perf_counter() - gen_start_time
+            logger.error(
+                f"[TIMING] run_ai_generation ERROR after {elapsed:.2f}s: {e}",
+                extra={
+                    "json_fields": {
+                        **log_meta,
+                        "elapsed_ms": elapsed * 1000,
+                        "error": str(e),
+                    }
+                },
+            )
+            await stream_registry.mark_task_completed(task_id, "failed")
+
+    # Start the AI generation in a background task
+    bg_task = asyncio.create_task(run_ai_generation())
+    await stream_registry.set_task_asyncio_task(task_id, bg_task)
     setup_time = (time.perf_counter() - stream_start_time) * 1000
     logger.info(
-        f"[TIMING] Task enqueued to RabbitMQ, setup={setup_time:.1f}ms",
+        f"[TIMING] Background task started, setup={setup_time:.1f}ms",
         extra={"json_fields": {**log_meta, "setup_time_ms": setup_time}},
     )
 
@@ -458,7 +421,7 @@ async def stream_chat_post(
 
         event_gen_start = time_module.perf_counter()
         logger.info(
-            f"[TIMING] event_generator STARTED, turn={turn_id}, session={session_id}, "
+            f"[TIMING] event_generator STARTED, task={task_id}, session={session_id}, "
             f"user={user_id}",
             extra={"json_fields": log_meta},
         )
@@ -466,12 +429,11 @@ async def stream_chat_post(
         first_chunk_yielded = False
         chunks_yielded = 0
         try:
-            # Subscribe from the position we captured before enqueuing
-            # This avoids replaying old messages while catching all new ones
-            subscriber_queue = await stream_registry.subscribe_to_session(
-                session_id=session_id,
+            # Subscribe to the task stream (this replays existing messages + live updates)
+            subscriber_queue = await stream_registry.subscribe_to_task(
+                task_id=task_id,
                 user_id=user_id,
-                last_message_id=subscribe_from_id,
+                last_message_id="0-0",  # Get all messages from the beginning
             )
 
             if subscriber_queue is None:
@@ -544,29 +506,23 @@ async def stream_chat_post(
                     "json_fields": {**log_meta, "elapsed_ms": elapsed, "error": str(e)}
                 },
             )
-            # Surface error to frontend so it doesn't appear stuck
-            yield StreamError(
-                errorText="An error occurred. Please try again.",
-                code="stream_error",
-            ).to_sse()
-            yield StreamFinish().to_sse()
         finally:
-            # Unsubscribe when client disconnects or stream ends
+            # Unsubscribe when client disconnects or stream ends to prevent resource leak
             if subscriber_queue is not None:
                 try:
-                    await stream_registry.unsubscribe_from_session(
-                        session_id, subscriber_queue
+                    await stream_registry.unsubscribe_from_task(
+                        task_id, subscriber_queue
                     )
                 except Exception as unsub_err:
                     logger.error(
-                        f"Error unsubscribing from session {session_id}: {unsub_err}",
+                        f"Error unsubscribing from task {task_id}: {unsub_err}",
                         exc_info=True,
                     )
             # AI SDK protocol termination - always yield even if unsubscribe fails
             total_time = time_module.perf_counter() - event_gen_start
             logger.info(
                 f"[TIMING] event_generator FINISHED in {total_time:.2f}s; "
-                f"turn={turn_id}, session={session_id}, n_chunks={chunks_yielded}",
+                f"task={task_id}, session={session_id}, n_chunks={chunks_yielded}",
                 extra={
                     "json_fields": {
                         **log_meta,
@@ -613,22 +569,17 @@ async def resume_session_stream(
     """
     import asyncio
 
-    active_session, last_message_id = await stream_registry.get_active_session(
+    active_task, _last_id = await stream_registry.get_active_task_for_session(
         session_id, user_id
     )
 
-    if not active_session:
+    if not active_task:
         return Response(status_code=204)
 
-    # Subscribe from the beginning ("0-0") to replay all chunks for this turn.
-    # This is necessary because hydrated messages filter out incomplete tool calls
-    # to avoid "No tool invocation found" errors. The resume stream delivers
-    # those tool calls fresh with proper SDK state.
-    # The AI SDK's deduplication will handle any duplicate chunks.
-    subscriber_queue = await stream_registry.subscribe_to_session(
-        session_id=session_id,
+    subscriber_queue = await stream_registry.subscribe_to_task(
+        task_id=active_task.task_id,
         user_id=user_id,
-        last_message_id="0-0",
+        last_message_id="0-0",  # Full replay so useChat rebuilds the message
     )
 
     if subscriber_queue is None:
@@ -664,12 +615,12 @@ async def resume_session_stream(
             logger.error(f"Error in resume stream for session {session_id}: {e}")
         finally:
             try:
-                await stream_registry.unsubscribe_from_session(
-                    session_id, subscriber_queue
+                await stream_registry.unsubscribe_from_task(
+                    active_task.task_id, subscriber_queue
                 )
             except Exception as unsub_err:
                 logger.error(
-                    f"Error unsubscribing from session {active_session.session_id}: {unsub_err}",
+                    f"Error unsubscribing from task {active_task.task_id}: {unsub_err}",
                     exc_info=True,
                 )
             logger.info(
@@ -720,6 +671,231 @@ async def session_assign_user(
     return {"status": "ok"}
 
 
+# ========== Task Streaming (SSE Reconnection) ==========
+
+
+@router.get(
+    "/tasks/{task_id}/stream",
+)
+async def stream_task(
+    task_id: str,
+    user_id: str | None = Depends(auth.get_user_id),
+    last_message_id: str = Query(
+        default="0-0",
+        description="Last Redis Stream message ID received (e.g., '1706540123456-0'). Use '0-0' for full replay.",
+    ),
+):
+    """
+    Reconnect to a long-running task's SSE stream.
+
+    When a long-running operation (like agent generation) starts, the client
+    receives a task_id. If the connection drops, the client can reconnect
+    using this endpoint to resume receiving updates.
+
+    Args:
+        task_id: The task ID from the operation_started response.
+        user_id: Authenticated user ID for ownership validation.
+        last_message_id: Last Redis Stream message ID received ("0-0" for full replay).
+
+    Returns:
+        StreamingResponse: SSE-formatted response chunks starting after last_message_id.
+
+    Raises:
+        HTTPException: 404 if task not found, 410 if task expired, 403 if access denied.
+    """
+    # Check task existence and expiry before subscribing
+    task, error_code = await stream_registry.get_task_with_expiry_info(task_id)
+
+    if error_code == "TASK_EXPIRED":
+        raise HTTPException(
+            status_code=410,
+            detail={
+                "code": "TASK_EXPIRED",
+                "message": "This operation has expired. Please try again.",
+            },
+        )
+
+    if error_code == "TASK_NOT_FOUND":
+        raise HTTPException(
+            status_code=404,
+            detail={
+                "code": "TASK_NOT_FOUND",
+                "message": f"Task {task_id} not found.",
+            },
+        )
+
+    # Validate ownership if task has an owner
+    if task and task.user_id and user_id != task.user_id:
+        raise HTTPException(
+            status_code=403,
+            detail={
+                "code": "ACCESS_DENIED",
+                "message": "You do not have access to this task.",
+            },
+        )
+
+    # Get subscriber queue from stream registry
+    subscriber_queue = await stream_registry.subscribe_to_task(
+        task_id=task_id,
+        user_id=user_id,
+        last_message_id=last_message_id,
+    )
+
+    if subscriber_queue is None:
+        raise HTTPException(
+            status_code=404,
+            detail={
+                "code": "TASK_NOT_FOUND",
+                "message": f"Task {task_id} not found or access denied.",
+            },
+        )
+
+    async def event_generator() -> AsyncGenerator[str, None]:
+        import asyncio
+
+        heartbeat_interval = 15.0  # Send heartbeat every 15 seconds
+        try:
+            while True:
+                try:
+                    # Wait for next chunk with timeout for heartbeats
+                    chunk = await asyncio.wait_for(
+                        subscriber_queue.get(), timeout=heartbeat_interval
+                    )
+                    yield chunk.to_sse()
+
+                    # Check for finish signal
+                    if isinstance(chunk, StreamFinish):
+                        break
+                except asyncio.TimeoutError:
+                    # Send heartbeat to keep connection alive
+                    yield StreamHeartbeat().to_sse()
+        except Exception as e:
+            logger.error(f"Error in task stream {task_id}: {e}", exc_info=True)
+        finally:
+            # Unsubscribe when client disconnects or stream ends
+            try:
+                await stream_registry.unsubscribe_from_task(task_id, subscriber_queue)
+            except Exception as unsub_err:
+                logger.error(
+                    f"Error unsubscribing from task {task_id}: {unsub_err}",
+                    exc_info=True,
+                )
+            # AI SDK protocol termination - always yield even if unsubscribe fails
+            yield "data: [DONE]\n\n"
+
+    return StreamingResponse(
+        event_generator(),
+        media_type="text/event-stream",
+        headers={
+            "Cache-Control": "no-cache",
+            "Connection": "keep-alive",
+            "X-Accel-Buffering": "no",
+            "x-vercel-ai-ui-message-stream": "v1",
+        },
+    )
+
+
+@router.get(
+    "/tasks/{task_id}",
+)
+async def get_task_status(
+    task_id: str,
+    user_id: str | None = Depends(auth.get_user_id),
+) -> dict:
+    """
+    Get the status of a long-running task.
+
+    Args:
+        task_id: The task ID to check.
+        user_id: Authenticated user ID for ownership validation.
+
+    Returns:
+        dict: Task status including task_id, status, tool_name, and operation_id.
+
+    Raises:
+        NotFoundError: If task_id is not found or user doesn't have access.
+    """
+    task = await stream_registry.get_task(task_id)
+
+    if task is None:
+        raise NotFoundError(f"Task {task_id} not found.")
+
+    # Validate ownership - if task has an owner, requester must match
+    if task.user_id and user_id != task.user_id:
+        raise NotFoundError(f"Task {task_id} not found.")
+
+    return {
+        "task_id": task.task_id,
+        "session_id": task.session_id,
+        "status": task.status,
+        "tool_name": task.tool_name,
+        "operation_id": task.operation_id,
+        "created_at": task.created_at.isoformat(),
+    }
+
+
+# ========== External Completion Webhook ==========
+
+
+@router.post(
+    "/operations/{operation_id}/complete",
+    status_code=200,
+)
+async def complete_operation(
+    operation_id: str,
+    request: OperationCompleteRequest,
+    x_api_key: str | None = Header(default=None),
+) -> dict:
+    """
+    External completion webhook for long-running operations.
+
+    Called by Agent Generator (or other services) when an operation completes.
+    This triggers the stream registry to publish completion and continue LLM generation.
+
+    Args:
+        operation_id: The operation ID to complete.
+        request: Completion payload with success status and result/error.
+        x_api_key: Internal API key for authentication.
+
+    Returns:
+        dict: Status of the completion.
+
+    Raises:
+        HTTPException: If API key is invalid or operation not found.
+    """
+    # Validate internal API key - reject if not configured or invalid
+    if not config.internal_api_key:
+        logger.error(
+            "Operation complete webhook rejected: CHAT_INTERNAL_API_KEY not configured"
+        )
+        raise HTTPException(
+            status_code=503,
+            detail="Webhook not available: internal API key not configured",
+        )
+    if x_api_key != config.internal_api_key:
+        raise HTTPException(status_code=401, detail="Invalid API key")
+
+    # Find task by operation_id
+    task = await stream_registry.find_task_by_operation_id(operation_id)
+    if task is None:
+        raise HTTPException(
+            status_code=404,
+            detail=f"Operation {operation_id} not found",
+        )
+
+    logger.info(
+        f"Received completion webhook for operation {operation_id} "
+        f"(task_id={task.task_id}, success={request.success})"
+    )
+
+    if request.success:
+        await process_operation_success(task, request.result)
+    else:
+        await process_operation_failure(task, request.error)
+
+    return {"status": "ok", "task_id": task.task_id}
+
+
 # ========== Configuration ==========
 
 
@@ -794,12 +970,13 @@ ToolResponseUnion = (
     | AgentPreviewResponse
     | AgentSavedResponse
     | ClarificationNeededResponse
-    | SuggestedGoalResponse
     | BlockListResponse
-    | BlockDetailsResponse
     | BlockOutputResponse
     | DocSearchResultsResponse
     | DocPageResponse
+    | OperationStartedResponse
+    | OperationPendingResponse
+    | OperationInProgressResponse
 )
 
 

autogpt_platform/backend/backend/api/features/chat/service_test.py

@@ -0,0 +1,82 @@
+import logging
+from os import getenv
+
+import pytest
+
+from . import service as chat_service
+from .model import create_chat_session, get_chat_session, upsert_chat_session
+from .response_model import (
+    StreamError,
+    StreamFinish,
+    StreamTextDelta,
+    StreamToolOutputAvailable,
+)
+
+logger = logging.getLogger(__name__)
+
+
+@pytest.mark.asyncio(loop_scope="session")
+async def test_stream_chat_completion(setup_test_user, test_user_id):
+    """
+    Test the stream_chat_completion function.
+    """
+    api_key: str | None = getenv("OPEN_ROUTER_API_KEY")
+    if not api_key:
+        return pytest.skip("OPEN_ROUTER_API_KEY is not set, skipping test")
+
+    session = await create_chat_session(test_user_id)
+
+    has_errors = False
+    has_ended = False
+    assistant_message = ""
+    async for chunk in chat_service.stream_chat_completion(
+        session.session_id, "Hello, how are you?", user_id=session.user_id
+    ):
+        logger.info(chunk)
+        if isinstance(chunk, StreamError):
+            has_errors = True
+        if isinstance(chunk, StreamTextDelta):
+            assistant_message += chunk.delta
+        if isinstance(chunk, StreamFinish):
+            has_ended = True
+
+    assert has_ended, "Chat completion did not end"
+    assert not has_errors, "Error occurred while streaming chat completion"
+    assert assistant_message, "Assistant message is empty"
+
+
+@pytest.mark.asyncio(loop_scope="session")
+async def test_stream_chat_completion_with_tool_calls(setup_test_user, test_user_id):
+    """
+    Test the stream_chat_completion function.
+    """
+    api_key: str | None = getenv("OPEN_ROUTER_API_KEY")
+    if not api_key:
+        return pytest.skip("OPEN_ROUTER_API_KEY is not set, skipping test")
+
+    session = await create_chat_session(test_user_id)
+    session = await upsert_chat_session(session)
+
+    has_errors = False
+    has_ended = False
+    had_tool_calls = False
+    async for chunk in chat_service.stream_chat_completion(
+        session.session_id,
+        "Please find me an agent that can help me with my business. Use the query 'moneny printing agent'",
+        user_id=session.user_id,
+    ):
+        logger.info(chunk)
+        if isinstance(chunk, StreamError):
+            has_errors = True
+
+        if isinstance(chunk, StreamFinish):
+            has_ended = True
+        if isinstance(chunk, StreamToolOutputAvailable):
+            had_tool_calls = True
+
+    assert has_ended, "Chat completion did not end"
+    assert not has_errors, "Error occurred while streaming chat completion"
+    assert had_tool_calls, "Tool calls did not occur"
+    session = await get_chat_session(session.session_id)
+    assert session, "Session not found"
+    assert session.usage, "Usage is empty"

autogpt_platform/backend/backend/api/features/chat/stream_registry.py

@@ -1,12 +1,12 @@
 """Stream registry for managing reconnectable SSE streams.
 
-This module provides a registry for tracking active streaming sessions and their
+This module provides a registry for tracking active streaming tasks and their
 messages. It uses Redis for all state management (no in-memory state), making
 pods stateless and horizontally scalable.
 
 Architecture:
 - Redis Stream: Persists all messages for replay and real-time delivery
-- Redis Hash: Session metadata (status, session_id, etc.)
+- Redis Hash: Task metadata (status, session_id, etc.)
 
 Subscribers:
 1. Replay missed messages from Redis Stream (XREAD)
@@ -16,7 +16,6 @@ Subscribers:
 
 import asyncio
 import logging
-import time
 from dataclasses import dataclass, field
 from datetime import datetime, timezone
 from typing import Any, Literal
@@ -26,25 +25,17 @@ import orjson
 from backend.data.redis_client import get_redis_async
 
 from .config import ChatConfig
-from .executor.utils import COPILOT_CONSUMER_TIMEOUT_SECONDS
-from .response_model import (
-    StreamBaseResponse,
-    StreamError,
-    StreamFinish,
-    StreamHeartbeat,
-    StreamTextDelta,
-    StreamTextStart,
-)
+from .response_model import StreamBaseResponse, StreamError, StreamFinish
 
 logger = logging.getLogger(__name__)
 config = ChatConfig()
 
 # Track background tasks for this pod (just the asyncio.Task reference, not subscribers)
-_local_sessions: dict[str, asyncio.Task] = {}
+_local_tasks: dict[str, asyncio.Task] = {}
 
 # Track listener tasks per subscriber queue for cleanup
-# Maps queue id() to (session_id, asyncio.Task) for proper cleanup on unsubscribe
-_listener_sessions: dict[int, tuple[str, asyncio.Task]] = {}
+# Maps queue id() to (task_id, asyncio.Task) for proper cleanup on unsubscribe
+_listener_tasks: dict[int, tuple[str, asyncio.Task]] = {}
 
 # Timeout for putting chunks into subscriber queues (seconds)
 # If the queue is full and doesn't drain within this time, send an overflow error
@@ -52,7 +43,7 @@ QUEUE_PUT_TIMEOUT = 5.0
 
 # Lua script for atomic compare-and-swap status update (idempotent completion)
 # Returns 1 if status was updated, 0 if already completed/failed
-COMPLETE_SESSION_SCRIPT = """
+COMPLETE_TASK_SCRIPT = """
 local current = redis.call("HGET", KEYS[1], "status")
 if current == "running" then
     redis.call("HSET", KEYS[1], "status", ARGV[1])
@@ -63,92 +54,81 @@ return 0
 
 
 @dataclass
-class ActiveSession:
-    """Represents an active streaming session (metadata only, no in-memory queues)."""
+class ActiveTask:
+    """Represents an active streaming task (metadata only, no in-memory queues)."""
 
+    task_id: str
     session_id: str
     user_id: str | None
     tool_call_id: str
     tool_name: str
-    turn_id: str = ""
-    blocking: bool = False  # If True, HTTP request is waiting for completion
+    operation_id: str
     status: Literal["running", "completed", "failed"] = "running"
     created_at: datetime = field(default_factory=lambda: datetime.now(timezone.utc))
     asyncio_task: asyncio.Task | None = None
 
 
-def _get_session_meta_key(session_id: str) -> str:
-    """Get Redis key for session metadata (keyed by session_id)."""
-    return f"{config.session_meta_prefix}{session_id}"
+def _get_task_meta_key(task_id: str) -> str:
+    """Get Redis key for task metadata."""
+    return f"{config.task_meta_prefix}{task_id}"
 
 
-def _get_turn_stream_key(turn_id: str) -> str:
-    """Get Redis key for turn message stream (keyed by turn_id for per-turn isolation)."""
-    return f"{config.turn_stream_prefix}{turn_id}"
+def _get_task_stream_key(task_id: str) -> str:
+    """Get Redis key for task message stream."""
+    return f"{config.task_stream_prefix}{task_id}"
 
 
-def _parse_session_meta(meta: dict[Any, Any], session_id: str = "") -> ActiveSession:
-    """Parse a raw Redis hash into a typed ActiveSession.
-
-    Centralises the ``meta.get(...)`` boilerplate so callers don't repeat it.
-    ``session_id`` is used as a fallback for ``turn_id`` when the meta hash
-    pre-dates the turn_id field (backward compat for in-flight sessions).
-    """
-    return ActiveSession(
-        session_id=meta.get("session_id", "") or session_id,
-        user_id=meta.get("user_id", "") or None,
-        tool_call_id=meta.get("tool_call_id", ""),
-        tool_name=meta.get("tool_name", ""),
-        turn_id=meta.get("turn_id", "") or session_id,
-        blocking=meta.get("blocking") == "1",
-        status=meta.get("status", "running"),  # type: ignore[arg-type]
-    )
+def _get_operation_mapping_key(operation_id: str) -> str:
+    """Get Redis key for operation_id to task_id mapping."""
+    return f"{config.task_op_prefix}{operation_id}"
 
 
-async def create_session_task(
+async def create_task(
+    task_id: str,
     session_id: str,
     user_id: str | None,
     tool_call_id: str,
     tool_name: str,
-    turn_id: str = "",
-    blocking: bool = False,
-) -> ActiveSession:
-    """Create a new streaming session in Redis (keyed by session_id).
+    operation_id: str,
+) -> ActiveTask:
+    """Create a new streaming task in Redis.
 
     Args:
-        session_id: Chat session ID (used as session identifier)
+        task_id: Unique identifier for the task
+        session_id: Chat session ID
         user_id: User ID (may be None for anonymous)
         tool_call_id: Tool call ID from the LLM
         tool_name: Name of the tool being executed
-        turn_id: Unique per-turn UUID for stream isolation
-        blocking: If True, HTTP request is waiting for completion
+        operation_id: Operation ID for webhook callbacks
 
     Returns:
-        The created ActiveSession instance (metadata only)
+        The created ActiveTask instance (metadata only)
     """
+    import time
+
     start_time = time.perf_counter()
 
     # Build log metadata for structured logging
     log_meta = {
         "component": "StreamRegistry",
+        "task_id": task_id,
         "session_id": session_id,
     }
     if user_id:
         log_meta["user_id"] = user_id
 
     logger.info(
-        f"[TIMING] create_session_task STARTED, session={session_id}, user={user_id}, turn_id={turn_id}",
+        f"[TIMING] create_task STARTED, task={task_id}, session={session_id}, user={user_id}",
         extra={"json_fields": log_meta},
     )
 
-    # Create session
-    session = ActiveSession(
+    task = ActiveTask(
+        task_id=task_id,
         session_id=session_id,
         user_id=user_id,
         tool_call_id=tool_call_id,
         tool_name=tool_name,
-        turn_id=turn_id,
-        blocking=blocking,
+        operation_id=operation_id,
     )
 
     # Store metadata in Redis
@@ -160,21 +140,21 @@ async def create_session_task(
         extra={"json_fields": {**log_meta, "duration_ms": redis_time}},
     )
 
-    meta_key = _get_session_meta_key(session_id)
-    # No need to delete old stream — each turn_id is a fresh UUID
+    meta_key = _get_task_meta_key(task_id)
+    op_key = _get_operation_mapping_key(operation_id)
 
     hset_start = time.perf_counter()
     await redis.hset(  # type: ignore[misc]
         meta_key,
         mapping={
+            "task_id": task_id,
             "session_id": session_id,
             "user_id": user_id or "",
             "tool_call_id": tool_call_id,
             "tool_name": tool_name,
-            "turn_id": turn_id,
-            "blocking": "1" if blocking else "0",
-            "status": session.status,
-            "created_at": session.created_at.isoformat(),
+            "operation_id": operation_id,
+            "status": task.status,
+            "created_at": task.created_at.isoformat(),
         },
     )
     hset_time = (time.perf_counter() - hset_start) * 1000
@@ -185,17 +165,20 @@ async def create_session_task(
 
     await redis.expire(meta_key, config.stream_ttl)
 
+    # Create operation_id -> task_id mapping for webhook lookups
+    await redis.set(op_key, task_id, ex=config.stream_ttl)
+
     total_time = (time.perf_counter() - start_time) * 1000
     logger.info(
-        f"[TIMING] create_session_task COMPLETED in {total_time:.1f}ms; session={session_id}",
+        f"[TIMING] create_task COMPLETED in {total_time:.1f}ms; task={task_id}, session={session_id}",
         extra={"json_fields": {**log_meta, "total_time_ms": total_time}},
     )
 
-    return session
+    return task
 
 
 async def publish_chunk(
-    turn_id: str,
+    task_id: str,
     chunk: StreamBaseResponse,
 ) -> str:
     """Publish a chunk to Redis Stream.
@@ -203,12 +186,14 @@ async def publish_chunk(
     All delivery is via Redis Streams - no in-memory state.
 
     Args:
-        turn_id: Turn ID (per-turn UUID) identifying the stream
+        task_id: Task ID to publish to
         chunk: The stream response chunk to publish
 
     Returns:
         The Redis Stream message ID
     """
+    import time
+
     start_time = time.perf_counter()
     chunk_type = type(chunk).__name__
     chunk_json = chunk.model_dump_json()
@@ -217,13 +202,13 @@ async def publish_chunk(
     # Build log metadata
     log_meta = {
         "component": "StreamRegistry",
-        "turn_id": turn_id,
+        "task_id": task_id,
         "chunk_type": chunk_type,
     }
 
     try:
         redis = await get_redis_async()
-        stream_key = _get_turn_stream_key(turn_id)
+        stream_key = _get_task_stream_key(task_id)
 
         # Write to Redis Stream for persistence and real-time delivery
         xadd_start = time.perf_counter()
@@ -235,21 +220,14 @@ async def publish_chunk(
         xadd_time = (time.perf_counter() - xadd_start) * 1000
         message_id = raw_id if isinstance(raw_id, str) else raw_id.decode()
 
-        # Set TTL on stream to match session metadata TTL
+        # Set TTL on stream to match task metadata TTL
         await redis.expire(stream_key, config.stream_ttl)
 
         total_time = (time.perf_counter() - start_time) * 1000
         # Only log timing for significant chunks or slow operations
         if (
             chunk_type
-            in (
-                "StreamStart",
-                "StreamFinish",
-                "StreamTextStart",
-                "StreamTextEnd",
-                "StreamToolInputAvailable",
-                "StreamToolOutputAvailable",
-            )
+            in ("StreamStart", "StreamFinish", "StreamTextStart", "StreamTextEnd")
             or total_time > 50
         ):
             logger.info(
@@ -274,39 +252,41 @@ async def publish_chunk(
     return message_id
 
 
-async def subscribe_to_session(
-    session_id: str,
+async def subscribe_to_task(
+    task_id: str,
     user_id: str | None,
     last_message_id: str = "0-0",
 ) -> asyncio.Queue[StreamBaseResponse] | None:
-    """Subscribe to a session's stream with replay of missed messages.
+    """Subscribe to a task's stream with replay of missed messages.
 
     This is fully stateless - uses Redis Stream for replay and pub/sub for live updates.
 
     Args:
-        session_id: Session ID to subscribe to
+        task_id: Task ID to subscribe to
         user_id: User ID for ownership validation
         last_message_id: Last Redis Stream message ID received ("0-0" for full replay)
 
     Returns:
-        An asyncio Queue that will receive stream chunks, or None if session not found
+        An asyncio Queue that will receive stream chunks, or None if task not found
         or user doesn't have access
     """
+    import time
+
     start_time = time.perf_counter()
 
     # Build log metadata
-    log_meta = {"component": "StreamRegistry", "session_id": session_id}
+    log_meta = {"component": "StreamRegistry", "task_id": task_id}
     if user_id:
         log_meta["user_id"] = user_id
 
     logger.info(
-        f"[TIMING] subscribe_to_session STARTED, session={session_id}, user={user_id}, last_msg={last_message_id}",
+        f"[TIMING] subscribe_to_task STARTED, task={task_id}, user={user_id}, last_msg={last_message_id}",
         extra={"json_fields": {**log_meta, "last_message_id": last_message_id}},
     )
 
     redis_start = time.perf_counter()
     redis = await get_redis_async()
-    meta_key = _get_session_meta_key(session_id)
+    meta_key = _get_task_meta_key(task_id)
     meta: dict[Any, Any] = await redis.hgetall(meta_key)  # type: ignore[misc]
     hgetall_time = (time.perf_counter() - redis_start) * 1000
     logger.info(
@@ -314,69 +294,54 @@ async def subscribe_to_session(
         extra={"json_fields": {**log_meta, "duration_ms": hgetall_time}},
     )
 
-    # RACE CONDITION FIX: If session not found, retry once after small delay
-    # This handles the case where subscribe_to_session is called immediately
-    # after create_session_task but before Redis propagates the write
     if not meta:
-        logger.warning(
-            "[TIMING] Session not found on first attempt, retrying after 50ms delay",
-            extra={"json_fields": {**log_meta}},
-        )
-        await asyncio.sleep(0.05)  # 50ms
-        meta = await redis.hgetall(meta_key)  # type: ignore[misc]
-        if not meta:
-            elapsed = (time.perf_counter() - start_time) * 1000
-            logger.info(
-                f"[TIMING] Session still not found in Redis after retry ({elapsed:.1f}ms total)",
-                extra={
-                    "json_fields": {
-                        **log_meta,
-                        "elapsed_ms": elapsed,
-                        "reason": "session_not_found_after_retry",
-                    }
-                },
-            )
-            return None
+        elapsed = (time.perf_counter() - start_time) * 1000
         logger.info(
-            "[TIMING] Session found after retry",
-            extra={"json_fields": {**log_meta}},
+            f"[TIMING] Task not found in Redis after {elapsed:.1f}ms",
+            extra={
+                "json_fields": {
+                    **log_meta,
+                    "elapsed_ms": elapsed,
+                    "reason": "task_not_found",
+                }
+            },
         )
+        return None
 
     # Note: Redis client uses decode_responses=True, so keys are strings
-    session_status = meta.get("status", "")
-    session_user_id = meta.get("user_id", "") or None
+    task_status = meta.get("status", "")
+    task_user_id = meta.get("user_id", "") or None
     log_meta["session_id"] = meta.get("session_id", "")
 
-    # Validate ownership - if session has an owner, requester must match
-    if session_user_id:
-        if user_id != session_user_id:
+    # Validate ownership - if task has an owner, requester must match
+    if task_user_id:
+        if user_id != task_user_id:
             logger.warning(
-                f"[TIMING] Access denied: user {user_id} tried to access session owned by {session_user_id}",
+                f"[TIMING] Access denied: user {user_id} tried to access task owned by {task_user_id}",
                 extra={
                     "json_fields": {
                         **log_meta,
-                        "session_owner": session_user_id,
+                        "task_owner": task_user_id,
                         "reason": "access_denied",
                     }
                 },
             )
             return None
 
-    session = _parse_session_meta(meta, session_id)
     subscriber_queue: asyncio.Queue[StreamBaseResponse] = asyncio.Queue()
-    stream_key = _get_turn_stream_key(session.turn_id)
+    stream_key = _get_task_stream_key(task_id)
 
     # Step 1: Replay messages from Redis Stream
     xread_start = time.perf_counter()
-    messages = await redis.xread({stream_key: last_message_id}, block=None, count=1000)
+    messages = await redis.xread({stream_key: last_message_id}, block=0, count=1000)
     xread_time = (time.perf_counter() - xread_start) * 1000
     logger.info(
-        f"[TIMING] Redis xread (replay) took {xread_time:.1f}ms, status={session_status}",
+        f"[TIMING] Redis xread (replay) took {xread_time:.1f}ms, status={task_status}",
         extra={
             "json_fields": {
                 **log_meta,
                 "duration_ms": xread_time,
-                "session_status": session_status,
+                "task_status": task_status,
             }
         },
     )
@@ -409,30 +374,28 @@ async def subscribe_to_session(
         },
     )
 
-    # Step 2: If session is still running, start stream listener for live updates
-    if session_status == "running":
+    # Step 2: If task is still running, start stream listener for live updates
+    if task_status == "running":
         logger.info(
-            "[TIMING] Session still running, starting _stream_listener",
-            extra={"json_fields": {**log_meta, "session_status": session_status}},
+            "[TIMING] Task still running, starting _stream_listener",
+            extra={"json_fields": {**log_meta, "task_status": task_status}},
         )
         listener_task = asyncio.create_task(
-            _stream_listener(
-                session_id, subscriber_queue, replay_last_id, log_meta, session.turn_id
-            )
+            _stream_listener(task_id, subscriber_queue, replay_last_id, log_meta)
         )
         # Track listener task for cleanup on unsubscribe
-        _listener_sessions[id(subscriber_queue)] = (session_id, listener_task)
+        _listener_tasks[id(subscriber_queue)] = (task_id, listener_task)
     else:
-        # Session is completed/failed - add finish marker
+        # Task is completed/failed - add finish marker
         logger.info(
-            f"[TIMING] Session already {session_status}, adding StreamFinish",
-            extra={"json_fields": {**log_meta, "session_status": session_status}},
+            f"[TIMING] Task already {task_status}, adding StreamFinish",
+            extra={"json_fields": {**log_meta, "task_status": task_status}},
         )
         await subscriber_queue.put(StreamFinish())
 
     total_time = (time.perf_counter() - start_time) * 1000
     logger.info(
-        f"[TIMING] subscribe_to_session COMPLETED in {total_time:.1f}ms; session={session_id}, "
+        f"[TIMING] subscribe_to_task COMPLETED in {total_time:.1f}ms; task={task_id}, "
         f"n_messages_replayed={replayed_count}",
         extra={
             "json_fields": {
@@ -446,11 +409,10 @@ async def subscribe_to_session(
 
 
 async def _stream_listener(
-    session_id: str,
+    task_id: str,
     subscriber_queue: asyncio.Queue[StreamBaseResponse],
     last_replayed_id: str,
     log_meta: dict | None = None,
-    turn_id: str = "",
 ) -> None:
     """Listen to Redis Stream for new messages using blocking XREAD.
 
@@ -458,20 +420,21 @@ async def _stream_listener(
     when messages are published during the gap between replay and subscription.
 
     Args:
-        session_id: Session ID to listen for
+        task_id: Task ID to listen for
         subscriber_queue: Queue to deliver messages to
         last_replayed_id: Last message ID from replay (continue from here)
         log_meta: Structured logging metadata
-        turn_id: Per-turn UUID for stream key resolution
     """
+    import time
+
     start_time = time.perf_counter()
 
     # Use provided log_meta or build minimal one
     if log_meta is None:
-        log_meta = {"component": "StreamRegistry", "session_id": session_id}
+        log_meta = {"component": "StreamRegistry", "task_id": task_id}
 
     logger.info(
-        f"[TIMING] _stream_listener STARTED, session={session_id}, last_id={last_replayed_id}",
+        f"[TIMING] _stream_listener STARTED, task={task_id}, last_id={last_replayed_id}",
         extra={"json_fields": {**log_meta, "last_replayed_id": last_replayed_id}},
     )
 
@@ -484,17 +447,16 @@ async def _stream_listener(
 
     try:
         redis = await get_redis_async()
-        stream_key = _get_turn_stream_key(turn_id)
+        stream_key = _get_task_stream_key(task_id)
         current_id = last_replayed_id
 
         while True:
-            # Block for up to 5 seconds waiting for new messages
-            # This allows periodic checking if session is still running
-            # Short timeout prevents frontend timeout (12s) while waiting for heartbeats (15s)
+            # Block for up to 30 seconds waiting for new messages
+            # This allows periodic checking if task is still running
             xread_start = time.perf_counter()
             xread_count += 1
             messages = await redis.xread(
-                {stream_key: current_id}, block=5000, count=100
+                {stream_key: current_id}, block=30000, count=100
             )
             xread_time = (time.perf_counter() - xread_start) * 1000
 
@@ -526,8 +488,8 @@ async def _stream_listener(
                 )
 
             if not messages:
-                # Timeout - check if session is still running
-                meta_key = _get_session_meta_key(session_id)
+                # Timeout - check if task is still running
+                meta_key = _get_task_meta_key(task_id)
                 status = await redis.hget(meta_key, "status")  # type: ignore[misc]
                 if status and status != "running":
                     try:
@@ -537,20 +499,9 @@ async def _stream_listener(
                         )
                     except asyncio.TimeoutError:
                         logger.warning(
-                            f"Timeout delivering finish event for session {session_id}"
+                            f"Timeout delivering finish event for task {task_id}"
                         )
                     break
-                # Session still running - send heartbeat to keep connection alive
-                # This prevents frontend timeout (12s) during long-running operations
-                try:
-                    await asyncio.wait_for(
-                        subscriber_queue.put(StreamHeartbeat()),
-                        timeout=QUEUE_PUT_TIMEOUT,
-                    )
-                except asyncio.TimeoutError:
-                    logger.warning(
-                        f"Timeout delivering heartbeat for session {session_id}"
-                    )
                 continue
 
             for _stream_name, stream_messages in messages:
@@ -610,7 +561,7 @@ async def _stream_listener(
                                 except asyncio.QueueFull:
                                     # Queue is completely stuck, nothing more we can do
                                     logger.error(
-                                        f"Cannot deliver overflow error for session {session_id}, "
+                                        f"Cannot deliver overflow error for task {task_id}, "
                                         "queue completely blocked"
                                     )
 
@@ -618,7 +569,7 @@ async def _stream_listener(
                             if isinstance(chunk, StreamFinish):
                                 total_time = (time.perf_counter() - start_time) * 1000
                                 logger.info(
-                                    f"[TIMING] StreamFinish received in {total_time / 1000:.1f}s; delivered={messages_delivered}",
+                                    f"[TIMING] StreamFinish received in {total_time/1000:.1f}s; delivered={messages_delivered}",
                                     extra={
                                         "json_fields": {
                                             **log_meta,
@@ -666,10 +617,10 @@ async def _stream_listener(
                 extra={"json_fields": log_meta},
             )
     finally:
-        # Clean up listener session mapping on exit
+        # Clean up listener task mapping on exit
         total_time = (time.perf_counter() - start_time) * 1000
         logger.info(
-            f"[TIMING] _stream_listener FINISHED in {total_time / 1000:.1f}s; session={session_id}, "
+            f"[TIMING] _stream_listener FINISHED in {total_time/1000:.1f}s; task={task_id}, "
             f"delivered={messages_delivered}, xread_count={xread_count}",
             extra={
                 "json_fields": {
@@ -680,198 +631,217 @@ async def _stream_listener(
                 }
             },
         )
-        _listener_sessions.pop(queue_id, None)
+        _listener_tasks.pop(queue_id, None)
 
 
-async def mark_session_completed(
-    session_id: str,
-    error_message: str | None = None,
+async def mark_task_completed(
+    task_id: str,
+    status: Literal["completed", "failed"] = "completed",
 ) -> bool:
-    """Mark a session as completed, then publish StreamFinish.
-
-    This is the SINGLE place that publishes StreamFinish to the turn stream.
-    Services must NOT yield StreamFinish themselves — the processor intercepts
-    it and calls this function instead, ensuring status is set before
-    StreamFinish reaches the frontend.
+    """Mark a task as completed and publish finish event.
 
+    This is idempotent - calling multiple times with the same task_id is safe.
     Uses atomic compare-and-swap via Lua script to prevent race conditions.
-    Idempotent — calling multiple times is safe (returns False on no-op).
+    Status is updated first (source of truth), then finish event is published (best-effort).
 
     Args:
-        session_id: Session ID to mark as completed
-        error_message: If provided, marks as "failed" and publishes a
-            StreamError before StreamFinish. Otherwise marks as "completed".
+        task_id: Task ID to mark as completed
+        status: Final status ("completed" or "failed")
 
     Returns:
-        True if session was newly marked completed, False if already completed/failed
+        True if task was newly marked completed, False if already completed/failed
     """
-    status: Literal["completed", "failed"] = "failed" if error_message else "completed"
-
     redis = await get_redis_async()
-    meta_key = _get_session_meta_key(session_id)
-
-    # Resolve turn_id for publishing to the correct stream
-    meta: dict[Any, Any] = await redis.hgetall(meta_key)  # type: ignore[misc]
-    turn_id = _parse_session_meta(meta, session_id).turn_id if meta else session_id
+    meta_key = _get_task_meta_key(task_id)
 
     # Atomic compare-and-swap: only update if status is "running"
-    result = await redis.eval(COMPLETE_SESSION_SCRIPT, 1, meta_key, status)  # type: ignore[misc]
+    # This prevents race conditions when multiple callers try to complete simultaneously
+    result = await redis.eval(COMPLETE_TASK_SCRIPT, 1, meta_key, status)  # type: ignore[misc]
 
     if result == 0:
-        logger.debug(f"Session {session_id} already completed/failed, skipping")
+        logger.debug(f"Task {task_id} already completed/failed, skipping")
         return False
 
-    if error_message:
-        try:
-            await publish_chunk(turn_id, StreamError(errorText=error_message))
-        except Exception as e:
-            logger.warning(
-                f"Failed to publish error event for session {session_id}: {e}"
-            )
-
-    # Publish StreamFinish AFTER status is set to "completed"/"failed".
-    # This is the SINGLE place that publishes StreamFinish — services and
-    # the processor must NOT publish it themselves.
+    # THEN publish finish event (best-effort - listeners can detect via status polling)
     try:
-        await publish_chunk(turn_id, StreamFinish())
+        await publish_chunk(task_id, StreamFinish())
     except Exception as e:
         logger.error(
-            f"Failed to publish StreamFinish for session {session_id}: {e}. "
-            "The _stream_listener will detect completion via status polling."
+            f"Failed to publish finish event for task {task_id}: {e}. "
+            "Listeners will detect completion via status polling."
         )
 
-    # Clean up local session reference if exists
-    _local_sessions.pop(session_id, None)
+    # Clean up local task reference if exists
+    _local_tasks.pop(task_id, None)
     return True
 
 
-async def get_session(session_id: str) -> ActiveSession | None:
-    """Get a session by its ID from Redis.
+async def find_task_by_operation_id(operation_id: str) -> ActiveTask | None:
+    """Find a task by its operation ID.
+
+    Used by webhook callbacks to locate the task to update.
 
     Args:
-        session_id: Session ID to look up
+        operation_id: Operation ID to search for
 
     Returns:
-        ActiveSession if found, None otherwise
+        ActiveTask if found, None otherwise
     """
     redis = await get_redis_async()
-    meta_key = _get_session_meta_key(session_id)
+    op_key = _get_operation_mapping_key(operation_id)
+    task_id = await redis.get(op_key)
+
+    if not task_id:
+        return None
+
+    task_id_str = task_id.decode() if isinstance(task_id, bytes) else task_id
+    return await get_task(task_id_str)
+
+
+async def get_task(task_id: str) -> ActiveTask | None:
+    """Get a task by its ID from Redis.
+
+    Args:
+        task_id: Task ID to look up
+
+    Returns:
+        ActiveTask if found, None otherwise
+    """
+    redis = await get_redis_async()
+    meta_key = _get_task_meta_key(task_id)
     meta: dict[Any, Any] = await redis.hgetall(meta_key)  # type: ignore[misc]
 
     if not meta:
         return None
 
-    return _parse_session_meta(meta, session_id)
+    # Note: Redis client uses decode_responses=True, so keys/values are strings
+    return ActiveTask(
+        task_id=meta.get("task_id", ""),
+        session_id=meta.get("session_id", ""),
+        user_id=meta.get("user_id", "") or None,
+        tool_call_id=meta.get("tool_call_id", ""),
+        tool_name=meta.get("tool_name", ""),
+        operation_id=meta.get("operation_id", ""),
+        status=meta.get("status", "running"),  # type: ignore[arg-type]
+    )
 
 
-async def get_session_with_expiry_info(
-    session_id: str,
-) -> tuple[ActiveSession | None, str | None]:
-    """Get a session by its ID with expiration detection.
+async def get_task_with_expiry_info(
+    task_id: str,
+) -> tuple[ActiveTask | None, str | None]:
+    """Get a task by its ID with expiration detection.
 
-    Returns (session, error_code) where error_code is:
-    - None if session found
+    Returns (task, error_code) where error_code is:
+    - None if task found
     - "TASK_EXPIRED" if stream exists but metadata is gone (TTL expired)
     - "TASK_NOT_FOUND" if neither exists
 
     Args:
-        session_id: Session ID to look up
+        task_id: Task ID to look up
 
     Returns:
-        Tuple of (ActiveSession or None, error_code or None)
+        Tuple of (ActiveTask or None, error_code or None)
     """
     redis = await get_redis_async()
-    meta_key = _get_session_meta_key(session_id)
+    meta_key = _get_task_meta_key(task_id)
+    stream_key = _get_task_stream_key(task_id)
 
     meta: dict[Any, Any] = await redis.hgetall(meta_key)  # type: ignore[misc]
 
     if not meta:
-        # Metadata expired — we can't resolve turn_id, so check using
-        # session_id as a best-effort fallback for the stream key.
-        stream_key = _get_turn_stream_key(session_id)
+        # Check if stream still has data (metadata expired but stream hasn't)
         stream_len = await redis.xlen(stream_key)
         if stream_len > 0:
             return None, "TASK_EXPIRED"
         return None, "TASK_NOT_FOUND"
 
-    return _parse_session_meta(meta, session_id), None
+    # Note: Redis client uses decode_responses=True, so keys/values are strings
+    return (
+        ActiveTask(
+            task_id=meta.get("task_id", ""),
+            session_id=meta.get("session_id", ""),
+            user_id=meta.get("user_id", "") or None,
+            tool_call_id=meta.get("tool_call_id", ""),
+            tool_name=meta.get("tool_name", ""),
+            operation_id=meta.get("operation_id", ""),
+            status=meta.get("status", "running"),  # type: ignore[arg-type]
+        ),
+        None,
+    )
 
 
-async def get_active_session(
+async def get_active_task_for_session(
     session_id: str,
     user_id: str | None = None,
-) -> tuple[ActiveSession | None, str]:
-    """Get the active (running) session, if any.
+) -> tuple[ActiveTask | None, str]:
+    """Get the active (running) task for a session, if any.
 
-    Direct O(1) lookup by session_id.
+    Scans Redis for tasks matching the session_id with status="running".
 
     Args:
         session_id: Session ID to look up
         user_id: User ID for ownership validation (optional)
 
     Returns:
-        Tuple of (ActiveSession if found and running, last_message_id from Redis Stream)
+        Tuple of (ActiveTask if found and running, last_message_id from Redis Stream)
     """
 
     redis = await get_redis_async()
-    meta_key = _get_session_meta_key(session_id)
-    meta: dict[Any, Any] = await redis.hgetall(meta_key)  # type: ignore[misc]
 
-    if not meta:
-        return None, "0-0"
+    # Scan Redis for task metadata keys
+    cursor = 0
+    tasks_checked = 0
 
-    session_status = meta.get("status", "")
-    session_user_id = meta.get("user_id", "") or None
+    while True:
+        cursor, keys = await redis.scan(
+            cursor, match=f"{config.task_meta_prefix}*", count=100
+        )
 
-    if session_status != "running":
-        return None, "0-0"
+        for key in keys:
+            tasks_checked += 1
+            meta: dict[Any, Any] = await redis.hgetall(key)  # type: ignore[misc]
+            if not meta:
+                continue
 
-    # Validate ownership - if session has an owner, requester must match
-    if session_user_id and user_id != session_user_id:
-        return None, "0-0"
+            # Note: Redis client uses decode_responses=True, so keys/values are strings
+            task_session_id = meta.get("session_id", "")
+            task_status = meta.get("status", "")
+            task_user_id = meta.get("user_id", "") or None
+            task_id = meta.get("task_id", "")
 
-    # Check if session is stale (running beyond tool timeout + buffer)
-    # Auto-complete it to prevent infinite polling loops
-    # Note: Synchronous tools can run up to COPILOT_CONSUMER_TIMEOUT_SECONDS (1 hour)
-    # so we add a 5-minute buffer to avoid false positives during legitimate operations
-    created_at_str = meta.get("created_at")
-    if created_at_str:
-        try:
-            created_at = datetime.fromisoformat(created_at_str)
-            age_seconds = (datetime.now(timezone.utc) - created_at).total_seconds()
-            stale_threshold = (
-                COPILOT_CONSUMER_TIMEOUT_SECONDS + 300
-            )  # + 5 minutes buffer
-            if age_seconds > stale_threshold:
-                logger.warning(
-                    f"[STALE_SESSION] Auto-completing stale session {session_id[:8]}... "
-                    f"(running for {age_seconds:.0f}s, threshold: {stale_threshold}s)"
+            if task_session_id == session_id and task_status == "running":
+                # Validate ownership - if task has an owner, requester must match
+                if task_user_id and user_id != task_user_id:
+                    continue
+
+                # Get the last message ID from Redis Stream
+                stream_key = _get_task_stream_key(task_id)
+                last_id = "0-0"
+                try:
+                    messages = await redis.xrevrange(stream_key, count=1)
+                    if messages:
+                        msg_id = messages[0][0]
+                        last_id = msg_id if isinstance(msg_id, str) else msg_id.decode()
+                except Exception as e:
+                    logger.warning(f"Failed to get last message ID: {e}")
+
+                return (
+                    ActiveTask(
+                        task_id=task_id,
+                        session_id=task_session_id,
+                        user_id=task_user_id,
+                        tool_call_id=meta.get("tool_call_id", ""),
+                        tool_name=meta.get("tool_name", ""),
+                        operation_id=meta.get("operation_id", ""),
+                        status="running",
+                    ),
+                    last_id,
                 )
-                await mark_session_completed(
-                    session_id,
-                    error_message=f"Session timed out after {age_seconds:.0f}s",
-                )
-                return None, "0-0"
-        except (ValueError, TypeError) as e:
-            logger.warning(f"Failed to parse created_at: {e}")
 
-    session = _parse_session_meta(meta, session_id)
-    logger.info(
-        f"[SESSION_LOOKUP] Found running session {session_id[:8]}..., turn_id={session.turn_id[:8]}"
-    )
+        if cursor == 0:
+            break
 
-    # Get the last message ID from Redis Stream (keyed by turn_id)
-    stream_key = _get_turn_stream_key(session.turn_id)
-    last_id = "0-0"
-    try:
-        messages = await redis.xrevrange(stream_key, count=1)
-        if messages:
-            msg_id = messages[0][0]
-            last_id = msg_id if isinstance(msg_id, str) else msg_id.decode()
-    except Exception as e:
-        logger.warning(f"Failed to get last message ID: {e}")
-
-    return session, last_id
+    return None, "0-0"
 
 
 def _reconstruct_chunk(chunk_data: dict) -> StreamBaseResponse | None:
@@ -891,7 +861,9 @@ def _reconstruct_chunk(chunk_data: dict) -> StreamBaseResponse | None:
         StreamHeartbeat,
         StreamStart,
         StreamStartStep,
+        StreamTextDelta,
         StreamTextEnd,
+        StreamTextStart,
         StreamToolInputAvailable,
         StreamToolInputStart,
         StreamToolOutputAvailable,
@@ -929,20 +901,20 @@ def _reconstruct_chunk(chunk_data: dict) -> StreamBaseResponse | None:
         return None
 
 
-async def set_session_asyncio_task(session_id: str, asyncio_task: asyncio.Task) -> None:
-    """Track the asyncio.Task for a session (local reference only).
+async def set_task_asyncio_task(task_id: str, asyncio_task: asyncio.Task) -> None:
+    """Track the asyncio.Task for a task (local reference only).
 
-    This is just for cleanup purposes - the session state is in Redis.
+    This is just for cleanup purposes - the task state is in Redis.
 
     Args:
-        session_id: Session ID
+        task_id: Task ID
         asyncio_task: The asyncio Task to track
     """
-    _local_sessions[session_id] = asyncio_task
+    _local_tasks[task_id] = asyncio_task
 
 
-async def unsubscribe_from_session(
-    session_id: str,
+async def unsubscribe_from_task(
+    task_id: str,
     subscriber_queue: asyncio.Queue[StreamBaseResponse],
 ) -> None:
     """Clean up when a subscriber disconnects.
@@ -951,29 +923,29 @@ async def unsubscribe_from_session(
     to prevent resource leaks.
 
     Args:
-        session_id: Session ID
+        task_id: Task ID
         subscriber_queue: The subscriber's queue used to look up the listener task
     """
     queue_id = id(subscriber_queue)
-    listener_entry = _listener_sessions.pop(queue_id, None)
+    listener_entry = _listener_tasks.pop(queue_id, None)
 
     if listener_entry is None:
         logger.debug(
-            f"No listener task found for session {session_id} queue {queue_id} "
+            f"No listener task found for task {task_id} queue {queue_id} "
             "(may have already completed)"
         )
         return
 
-    stored_session_id, listener_task = listener_entry
+    stored_task_id, listener_task = listener_entry
 
-    if stored_session_id != session_id:
+    if stored_task_id != task_id:
         logger.warning(
-            f"Session ID mismatch in unsubscribe: expected {session_id}, "
-            f"found {stored_session_id}"
+            f"Task ID mismatch in unsubscribe: expected {task_id}, "
+            f"found {stored_task_id}"
         )
 
     if listener_task.done():
-        logger.debug(f"Listener task for session {session_id} already completed")
+        logger.debug(f"Listener task for task {task_id} already completed")
         return
 
     # Cancel the listener task
@@ -987,11 +959,9 @@ async def unsubscribe_from_session(
         pass
     except asyncio.TimeoutError:
         logger.warning(
-            f"Timeout waiting for listener task cancellation for session {session_id}"
+            f"Timeout waiting for listener task cancellation for task {task_id}"
         )
     except Exception as e:
-        logger.error(
-            f"Error during listener task cancellation for session {session_id}: {e}"
-        )
+        logger.error(f"Error during listener task cancellation for task {task_id}: {e}")
 
-    logger.debug(f"Successfully unsubscribed from session {session_id}")
+    logger.debug(f"Successfully unsubscribed from task {task_id}")

autogpt_platform/backend/backend/api/features/chat/tools/__init__.py

@@ -3,17 +3,15 @@ from typing import TYPE_CHECKING, Any
 
 from openai.types.chat import ChatCompletionToolParam
 
-from backend.copilot.model import ChatSession
-from backend.copilot.tracking import track_tool_called
+from backend.api.features.chat.model import ChatSession
+from backend.api.features.chat.tracking import track_tool_called
 
 from .add_understanding import AddUnderstandingTool
 from .agent_output import AgentOutputTool
 from .base import BaseTool
-from .bash_exec import BashExecTool
 from .create_agent import CreateAgentTool
 from .customize_agent import CustomizeAgentTool
 from .edit_agent import EditAgentTool
-from .feature_requests import CreateFeatureRequestTool, SearchFeatureRequestsTool
 from .find_agent import FindAgentTool
 from .find_block import FindBlockTool
 from .find_library_agent import FindLibraryAgentTool
@@ -21,7 +19,6 @@ from .get_doc_page import GetDocPageTool
 from .run_agent import RunAgentTool
 from .run_block import RunBlockTool
 from .search_docs import SearchDocsTool
-from .web_fetch import WebFetchTool
 from .workspace_files import (
     DeleteWorkspaceFileTool,
     ListWorkspaceFilesTool,
@@ -30,7 +27,7 @@ from .workspace_files import (
 )
 
 if TYPE_CHECKING:
-    from backend.copilot.response_model import StreamToolOutputAvailable
+    from backend.api.features.chat.response_model import StreamToolOutputAvailable
 
 logger = logging.getLogger(__name__)
 
@@ -48,14 +45,6 @@ TOOL_REGISTRY: dict[str, BaseTool] = {
     "view_agent_output": AgentOutputTool(),
     "search_docs": SearchDocsTool(),
     "get_doc_page": GetDocPageTool(),
-    # Web fetch for safe URL retrieval
-    "web_fetch": WebFetchTool(),
-    # Sandboxed code execution (bubblewrap)
-    "bash_exec": BashExecTool(),
-    # Persistent workspace tools (cloud storage, survives across sessions)
-    # Feature request tools
-    "search_feature_requests": SearchFeatureRequestsTool(),
-    "create_feature_request": CreateFeatureRequestTool(),
     # Workspace tools for CoPilot file operations
     "list_workspace_files": ListWorkspaceFilesTool(),
     "read_workspace_file": ReadWorkspaceFileTool(),

autogpt_platform/backend/backend/api/features/chat/tools/_test_data.py

@@ -6,11 +6,11 @@ import pytest
 from prisma.types import ProfileCreateInput
 from pydantic import SecretStr
 
+from backend.api.features.chat.model import ChatSession
 from backend.api.features.store import db as store_db
 from backend.blocks.firecrawl.scrape import FirecrawlScrapeBlock
 from backend.blocks.io import AgentInputBlock, AgentOutputBlock
 from backend.blocks.llm import AITextGeneratorBlock
-from backend.copilot.model import ChatSession
 from backend.data.db import prisma
 from backend.data.graph import Graph, Link, Node, create_graph
 from backend.data.model import APIKeyCredentials

autogpt_platform/backend/backend/api/features/chat/tools/add_understanding.py

@@ -3,9 +3,11 @@
 import logging
 from typing import Any
 
-from backend.copilot.model import ChatSession
-from backend.data.db_accessors import understanding_db
-from backend.data.understanding import BusinessUnderstandingInput
+from backend.api.features.chat.model import ChatSession
+from backend.data.understanding import (
+    BusinessUnderstandingInput,
+    upsert_business_understanding,
+)
 
 from .base import BaseTool
 from .models import ErrorResponse, ToolResponseBase, UnderstandingUpdatedResponse
@@ -97,9 +99,7 @@ and automations for the user's specific needs."""
         ]
 
         # Upsert with merge
-        understanding = await understanding_db().upsert_business_understanding(
-            user_id, input_data
-        )
+        understanding = await upsert_business_understanding(user_id, input_data)
 
         # Build current understanding summary (filter out empty values)
         current_understanding = {

autogpt_platform/backend/backend/api/features/chat/tools/agent_generator/__init__.py

@@ -19,7 +19,6 @@ from .core import (
     get_all_relevant_agents_for_generation,
     get_library_agent_by_graph_id,
     get_library_agent_by_id,
-    get_library_agents_by_ids,
     get_library_agents_for_generation,
     graph_to_json,
     json_to_graph,
@@ -50,7 +49,6 @@ __all__ = [
     "get_all_relevant_agents_for_generation",
     "get_library_agent_by_graph_id",
     "get_library_agent_by_id",
-    "get_library_agents_by_ids",
     "get_library_agents_for_generation",
     "get_user_message_for_error",
     "graph_to_json",

autogpt_platform/backend/backend/api/features/chat/tools/agent_generator/core.py

@@ -3,11 +3,11 @@
 import logging
 import re
 import uuid
-from collections.abc import Sequence
 from typing import Any, NotRequired, TypedDict
 
-from backend.data.db_accessors import graph_db, library_db, store_db
-from backend.data.graph import Graph, Link, Node
+from backend.api.features.library import db as library_db
+from backend.api.features.store import db as store_db
+from backend.data.graph import Graph, Link, Node, get_graph, get_store_listed_graphs
 from backend.util.exceptions import DatabaseError, NotFoundError
 
 from .service import (
@@ -79,7 +79,7 @@ AgentSummary = LibraryAgentSummary | MarketplaceAgentSummary | dict[str, Any]
 
 
 def _to_dict_list(
-    agents: Sequence[AgentSummary] | Sequence[dict[str, Any]] | None,
+    agents: list[AgentSummary] | list[dict[str, Any]] | None,
 ) -> list[dict[str, Any]] | None:
     """Convert typed agent summaries to plain dicts for external service calls."""
     if agents is None:
@@ -145,9 +145,8 @@ async def get_library_agent_by_id(
     Returns:
         LibraryAgentSummary if found, None otherwise
     """
-    db = library_db()
     try:
-        agent = await db.get_library_agent_by_graph_id(user_id, agent_id)
+        agent = await library_db.get_library_agent_by_graph_id(user_id, agent_id)
         if agent:
             logger.debug(f"Found library agent by graph_id: {agent.name}")
             return LibraryAgentSummary(
@@ -164,7 +163,7 @@ async def get_library_agent_by_id(
         logger.debug(f"Could not fetch library agent by graph_id {agent_id}: {e}")
 
     try:
-        agent = await db.get_library_agent(agent_id, user_id)
+        agent = await library_db.get_library_agent(agent_id, user_id)
         if agent:
             logger.debug(f"Found library agent by library_id: {agent.name}")
             return LibraryAgentSummary(
@@ -191,36 +190,6 @@ async def get_library_agent_by_id(
 get_library_agent_by_graph_id = get_library_agent_by_id
 
 
-async def get_library_agents_by_ids(
-    user_id: str,
-    agent_ids: list[str],
-) -> list[LibraryAgentSummary]:
-    """Fetch multiple library agents by their IDs.
-
-    Args:
-        user_id: The user ID
-        agent_ids: List of agent IDs (can be graph_ids or library agent IDs)
-
-    Returns:
-        List of LibraryAgentSummary for found agents (silently skips not found)
-    """
-    agents: list[LibraryAgentSummary] = []
-    for agent_id in agent_ids:
-        try:
-            agent = await get_library_agent_by_id(user_id, agent_id)
-            if agent:
-                agents.append(agent)
-                logger.debug(f"Fetched library agent by ID: {agent['name']}")
-            else:
-                logger.warning(f"Library agent not found for ID: {agent_id}")
-        except Exception as e:
-            logger.warning(f"Failed to fetch library agent {agent_id}: {e}")
-            continue
-
-    logger.info(f"Fetched {len(agents)}/{len(agent_ids)} library agents by ID")
-    return agents
-
-
 async def get_library_agents_for_generation(
     user_id: str,
     search_query: str | None = None,
@@ -245,17 +214,10 @@ async def get_library_agents_for_generation(
     Returns:
         List of LibraryAgentSummary with schemas and recent executions for sub-agent composition
     """
-    search_term = search_query.strip() if search_query else None
-    if search_term and len(search_term) > 100:
-        raise ValueError(
-            f"Search query is too long ({len(search_term)} chars, max 100). "
-            f"Please use a shorter, more specific search term."
-        )
-
     try:
-        response = await library_db().list_library_agents(
+        response = await library_db.list_library_agents(
             user_id=user_id,
-            search_term=search_term,
+            search_term=search_query,
             page=1,
             page_size=max_results,
             include_executions=True,
@@ -309,16 +271,9 @@ async def search_marketplace_agents_for_generation(
     Returns:
         List of LibraryAgentSummary with full input/output schemas
     """
-    search_term = search_query.strip()
-    if len(search_term) > 100:
-        raise ValueError(
-            f"Search query is too long ({len(search_term)} chars, max 100). "
-            f"Please use a shorter, more specific search term."
-        )
-
     try:
-        response = await store_db().get_store_agents(
-            search_query=search_term,
+        response = await store_db.get_store_agents(
+            search_query=search_query,
             page=1,
             page_size=max_results,
         )
@@ -331,7 +286,7 @@ async def search_marketplace_agents_for_generation(
             return []
 
         graph_ids = [agent.agent_graph_id for agent in agents_with_graphs]
-        graphs = await graph_db().get_store_listed_graphs(graph_ids)
+        graphs = await get_store_listed_graphs(*graph_ids)
 
         results: list[LibraryAgentSummary] = []
         for agent in agents_with_graphs:
@@ -469,7 +424,7 @@ def extract_search_terms_from_steps(
 async def enrich_library_agents_from_steps(
     user_id: str,
     decomposition_result: DecompositionResult | dict[str, Any],
-    existing_agents: Sequence[AgentSummary] | Sequence[dict[str, Any]],
+    existing_agents: list[AgentSummary] | list[dict[str, Any]],
     exclude_graph_id: str | None = None,
     include_marketplace: bool = True,
     max_additional_results: int = 10,
@@ -493,7 +448,7 @@ async def enrich_library_agents_from_steps(
     search_terms = extract_search_terms_from_steps(decomposition_result)
 
     if not search_terms:
-        return list(existing_agents)
+        return existing_agents
 
     existing_ids: set[str] = set()
     existing_names: set[str] = set()
@@ -556,7 +511,7 @@ async def enrich_library_agents_from_steps(
 async def decompose_goal(
     description: str,
     context: str = "",
-    library_agents: Sequence[AgentSummary] | None = None,
+    library_agents: list[AgentSummary] | None = None,
 ) -> DecompositionResult | None:
     """Break down a goal into steps or return clarifying questions.
 
@@ -584,16 +539,22 @@ async def decompose_goal(
 
 async def generate_agent(
     instructions: DecompositionResult | dict[str, Any],
-    library_agents: Sequence[AgentSummary] | Sequence[dict[str, Any]] | None = None,
+    library_agents: list[AgentSummary] | list[dict[str, Any]] | None = None,
+    operation_id: str | None = None,
+    task_id: str | None = None,
 ) -> dict[str, Any] | None:
     """Generate agent JSON from instructions.
 
     Args:
         instructions: Structured instructions from decompose_goal
         library_agents: User's library agents available for sub-agent composition
+        operation_id: Operation ID for async processing (enables Redis Streams
+            completion notification)
+        task_id: Task ID for async processing (enables Redis Streams persistence
+            and SSE delivery)
 
     Returns:
-        Agent JSON dict, error dict {"type": "error", ...}, or None on error
+        Agent JSON dict, {"status": "accepted"} for async, error dict {"type": "error", ...}, or None on error
 
     Raises:
         AgentGeneratorNotConfiguredError: If the external service is not configured.
@@ -601,9 +562,13 @@ async def generate_agent(
     _check_service_configured()
     logger.info("Calling external Agent Generator service for generate_agent")
     result = await generate_agent_external(
-        dict(instructions), _to_dict_list(library_agents)
+        dict(instructions), _to_dict_list(library_agents), operation_id, task_id
     )
 
+    # Don't modify async response
+    if result and result.get("status") == "accepted":
+        return result
+
     if result:
         if isinstance(result, dict) and result.get("type") == "error":
             return result
@@ -708,10 +673,9 @@ async def save_agent_to_library(
         Tuple of (created Graph, LibraryAgent)
     """
     graph = json_to_graph(agent_json)
-    db = library_db()
     if is_update:
-        return await db.update_graph_in_library(graph, user_id)
-    return await db.create_graph_in_library(graph, user_id)
+        return await library_db.update_graph_in_library(graph, user_id)
+    return await library_db.create_graph_in_library(graph, user_id)
 
 
 def graph_to_json(graph: Graph) -> dict[str, Any]:
@@ -771,14 +735,12 @@ async def get_agent_as_json(
     Returns:
         Agent as JSON dict or None if not found
     """
-    db = graph_db()
-
-    graph = await db.get_graph(agent_id, version=None, user_id=user_id)
+    graph = await get_graph(agent_id, version=None, user_id=user_id)
 
     if not graph and user_id:
         try:
-            library_agent = await library_db().get_library_agent(agent_id, user_id)
-            graph = await db.get_graph(
+            library_agent = await library_db.get_library_agent(agent_id, user_id)
+            graph = await get_graph(
                 library_agent.graph_id, version=None, user_id=user_id
             )
         except NotFoundError:
@@ -793,7 +755,9 @@ async def get_agent_as_json(
 async def generate_agent_patch(
     update_request: str,
     current_agent: dict[str, Any],
-    library_agents: Sequence[AgentSummary] | None = None,
+    library_agents: list[AgentSummary] | None = None,
+    operation_id: str | None = None,
+    task_id: str | None = None,
 ) -> dict[str, Any] | None:
     """Update an existing agent using natural language.
 
@@ -806,10 +770,12 @@ async def generate_agent_patch(
         update_request: Natural language description of changes
         current_agent: Current agent JSON
         library_agents: User's library agents available for sub-agent composition
+        operation_id: Operation ID for async processing (enables Redis Streams callback)
+        task_id: Task ID for async processing (enables Redis Streams callback)
 
     Returns:
         Updated agent JSON, clarifying questions dict {"type": "clarifying_questions", ...},
-        error dict {"type": "error", ...}, or None on error
+        {"status": "accepted"} for async, error dict {"type": "error", ...}, or None on error
 
     Raises:
         AgentGeneratorNotConfiguredError: If the external service is not configured.
@@ -820,6 +786,8 @@ async def generate_agent_patch(
         update_request,
         current_agent,
         _to_dict_list(library_agents),
+        operation_id,
+        task_id,
     )
 
 

autogpt_platform/backend/backend/api/features/chat/tools/agent_generator/service.py

@@ -12,19 +12,8 @@ import httpx
 
 from backend.util.settings import Settings
 
-from .dummy import (
-    customize_template_dummy,
-    decompose_goal_dummy,
-    generate_agent_dummy,
-    generate_agent_patch_dummy,
-    get_blocks_dummy,
-    health_check_dummy,
-)
-
 logger = logging.getLogger(__name__)
 
-_dummy_mode_warned = False
-
 
 def _create_error_response(
     error_message: str,
@@ -101,26 +90,10 @@ def _get_settings() -> Settings:
     return _settings
 
 
-def _is_dummy_mode() -> bool:
-    """Check if dummy mode is enabled for testing."""
-    global _dummy_mode_warned
-    settings = _get_settings()
-    is_dummy = bool(settings.config.agentgenerator_use_dummy)
-    if is_dummy and not _dummy_mode_warned:
-        logger.warning(
-            "Agent Generator running in DUMMY MODE - returning mock responses. "
-            "Do not use in production!"
-        )
-        _dummy_mode_warned = True
-    return is_dummy
-
-
 def is_external_service_configured() -> bool:
-    """Check if external Agent Generator service is configured (or dummy mode)."""
+    """Check if external Agent Generator service is configured."""
     settings = _get_settings()
-    return bool(settings.config.agentgenerator_host) or bool(
-        settings.config.agentgenerator_use_dummy
-    )
+    return bool(settings.config.agentgenerator_host)
 
 
 def _get_base_url() -> str:
@@ -164,9 +137,6 @@ async def decompose_goal_external(
         - {"type": "error", "error": "...", "error_type": "..."} on error
         Or None on unexpected error
     """
-    if _is_dummy_mode():
-        return await decompose_goal_dummy(description, context, library_agents)
-
     client = _get_client()
 
     if context:
@@ -242,28 +212,45 @@ async def decompose_goal_external(
 async def generate_agent_external(
     instructions: dict[str, Any],
     library_agents: list[dict[str, Any]] | None = None,
+    operation_id: str | None = None,
+    task_id: str | None = None,
 ) -> dict[str, Any] | None:
     """Call the external service to generate an agent from instructions.
 
     Args:
         instructions: Structured instructions from decompose_goal
         library_agents: User's library agents available for sub-agent composition
+        operation_id: Operation ID for async processing (enables Redis Streams callback)
+        task_id: Task ID for async processing (enables Redis Streams callback)
 
     Returns:
-        Agent JSON dict or error dict {"type": "error", ...} on error
+        Agent JSON dict, {"status": "accepted"} for async, or error dict {"type": "error", ...} on error
     """
-    if _is_dummy_mode():
-        return await generate_agent_dummy(instructions, library_agents)
-
     client = _get_client()
 
     # Build request payload
     payload: dict[str, Any] = {"instructions": instructions}
     if library_agents:
         payload["library_agents"] = library_agents
+    if operation_id and task_id:
+        payload["operation_id"] = operation_id
+        payload["task_id"] = task_id
 
     try:
         response = await client.post("/api/generate-agent", json=payload)
+
+        # Handle 202 Accepted for async processing
+        if response.status_code == 202:
+            logger.info(
+                f"Agent Generator accepted async request "
+                f"(operation_id={operation_id}, task_id={task_id})"
+            )
+            return {
+                "status": "accepted",
+                "operation_id": operation_id,
+                "task_id": task_id,
+            }
+
         response.raise_for_status()
         data = response.json()
 
@@ -295,6 +282,8 @@ async def generate_agent_patch_external(
     update_request: str,
     current_agent: dict[str, Any],
     library_agents: list[dict[str, Any]] | None = None,
+    operation_id: str | None = None,
+    task_id: str | None = None,
 ) -> dict[str, Any] | None:
     """Call the external service to generate a patch for an existing agent.
 
@@ -303,16 +292,11 @@ async def generate_agent_patch_external(
         current_agent: Current agent JSON
         library_agents: User's library agents available for sub-agent composition
         operation_id: Operation ID for async processing (enables Redis Streams callback)
-        session_id: Session ID for async processing (enables Redis Streams callback)
+        task_id: Task ID for async processing (enables Redis Streams callback)
 
     Returns:
         Updated agent JSON, clarifying questions dict, {"status": "accepted"} for async, or error dict on error
     """
-    if _is_dummy_mode():
-        return await generate_agent_patch_dummy(
-            update_request, current_agent, library_agents
-        )
-
     client = _get_client()
 
     # Build request payload
@@ -322,9 +306,25 @@ async def generate_agent_patch_external(
     }
     if library_agents:
         payload["library_agents"] = library_agents
+    if operation_id and task_id:
+        payload["operation_id"] = operation_id
+        payload["task_id"] = task_id
 
     try:
         response = await client.post("/api/update-agent", json=payload)
+
+        # Handle 202 Accepted for async processing
+        if response.status_code == 202:
+            logger.info(
+                f"Agent Generator accepted async update request "
+                f"(operation_id={operation_id}, task_id={task_id})"
+            )
+            return {
+                "status": "accepted",
+                "operation_id": operation_id,
+                "task_id": task_id,
+            }
+
         response.raise_for_status()
         data = response.json()
 
@@ -379,17 +379,10 @@ async def customize_template_external(
         template_agent: The template agent JSON to customize
         modification_request: Natural language description of customizations
         context: Additional context (e.g., answers to previous questions)
-        operation_id: Operation ID for async processing (enables Redis Streams callback)
-        session_id: Session ID for async processing (enables Redis Streams callback)
 
     Returns:
         Customized agent JSON, clarifying questions dict, or error dict on error
     """
-    if _is_dummy_mode():
-        return await customize_template_dummy(
-            template_agent, modification_request, context
-        )
-
     client = _get_client()
 
     request = modification_request
@@ -452,9 +445,6 @@ async def get_blocks_external() -> list[dict[str, Any]] | None:
     Returns:
         List of block info dicts or None on error
     """
-    if _is_dummy_mode():
-        return await get_blocks_dummy()
-
     client = _get_client()
 
     try:
@@ -488,9 +478,6 @@ async def health_check() -> bool:
     if not is_external_service_configured():
         return False
 
-    if _is_dummy_mode():
-        return await health_check_dummy()
-
     client = _get_client()
 
     try:

autogpt_platform/backend/backend/api/features/chat/tools/agent_output.py

@@ -7,9 +7,10 @@ from typing import Any
 
 from pydantic import BaseModel, field_validator
 
+from backend.api.features.chat.model import ChatSession
+from backend.api.features.library import db as library_db
 from backend.api.features.library.model import LibraryAgent
-from backend.copilot.model import ChatSession
-from backend.data.db_accessors import execution_db, library_db
+from backend.data import execution as execution_db
 from backend.data.execution import ExecutionStatus, GraphExecution, GraphExecutionMeta
 
 from .base import BaseTool
@@ -164,12 +165,10 @@ class AgentOutputTool(BaseTool):
         Resolve agent from provided identifiers.
         Returns (library_agent, error_message).
         """
-        lib_db = library_db()
-
         # Priority 1: Exact library agent ID
         if library_agent_id:
             try:
-                agent = await lib_db.get_library_agent(library_agent_id, user_id)
+                agent = await library_db.get_library_agent(library_agent_id, user_id)
                 return agent, None
             except Exception as e:
                 logger.warning(f"Failed to get library agent by ID: {e}")
@@ -183,7 +182,7 @@ class AgentOutputTool(BaseTool):
                 return None, f"Agent '{store_slug}' not found in marketplace"
 
             # Find in user's library by graph_id
-            agent = await lib_db.get_library_agent_by_graph_id(user_id, graph.id)
+            agent = await library_db.get_library_agent_by_graph_id(user_id, graph.id)
             if not agent:
                 return (
                     None,
@@ -195,7 +194,7 @@ class AgentOutputTool(BaseTool):
         # Priority 3: Fuzzy name search in library
         if agent_name:
             try:
-                response = await lib_db.list_library_agents(
+                response = await library_db.list_library_agents(
                     user_id=user_id,
                     search_term=agent_name,
                     page_size=5,
@@ -229,11 +228,9 @@ class AgentOutputTool(BaseTool):
         Fetch execution(s) based on filters.
         Returns (single_execution, available_executions_meta, error_message).
         """
-        exec_db = execution_db()
-
         # If specific execution_id provided, fetch it directly
         if execution_id:
-            execution = await exec_db.get_graph_execution(
+            execution = await execution_db.get_graph_execution(
                 user_id=user_id,
                 execution_id=execution_id,
                 include_node_executions=False,
@@ -243,7 +240,7 @@ class AgentOutputTool(BaseTool):
             return execution, [], None
 
         # Get completed executions with time filters
-        executions = await exec_db.get_graph_executions(
+        executions = await execution_db.get_graph_executions(
             graph_id=graph_id,
             user_id=user_id,
             statuses=[ExecutionStatus.COMPLETED],
@@ -257,7 +254,7 @@ class AgentOutputTool(BaseTool):
 
         # If only one execution, fetch full details
         if len(executions) == 1:
-            full_execution = await exec_db.get_graph_execution(
+            full_execution = await execution_db.get_graph_execution(
                 user_id=user_id,
                 execution_id=executions[0].id,
                 include_node_executions=False,
@@ -265,7 +262,7 @@ class AgentOutputTool(BaseTool):
             return full_execution, [], None
 
         # Multiple executions - return latest with full details, plus list of available
-        full_execution = await exec_db.get_graph_execution(
+        full_execution = await execution_db.get_graph_execution(
             user_id=user_id,
             execution_id=executions[0].id,
             include_node_executions=False,
@@ -383,7 +380,7 @@ class AgentOutputTool(BaseTool):
             and not input_data.store_slug
         ):
             # Fetch execution directly to get graph_id
-            execution = await execution_db().get_graph_execution(
+            execution = await execution_db.get_graph_execution(
                 user_id=user_id,
                 execution_id=input_data.execution_id,
                 include_node_executions=False,
@@ -395,7 +392,7 @@ class AgentOutputTool(BaseTool):
                 )
 
             # Find library agent by graph_id
-            agent = await library_db().get_library_agent_by_graph_id(
+            agent = await library_db.get_library_agent_by_graph_id(
                 user_id, execution.graph_id
             )
             if not agent:

autogpt_platform/backend/backend/api/features/chat/tools/agent_search.py

@@ -4,7 +4,8 @@ import logging
 import re
 from typing import Literal
 
-from backend.data.db_accessors import library_db, store_db
+from backend.api.features.library import db as library_db
+from backend.api.features.store import db as store_db
 from backend.util.exceptions import DatabaseError, NotFoundError
 
 from .models import (
@@ -44,10 +45,8 @@ async def _get_library_agent_by_id(user_id: str, agent_id: str) -> AgentInfo | N
     Returns:
         AgentInfo if found, None otherwise
     """
-    lib_db = library_db()
-
     try:
-        agent = await lib_db.get_library_agent_by_graph_id(user_id, agent_id)
+        agent = await library_db.get_library_agent_by_graph_id(user_id, agent_id)
         if agent:
             logger.debug(f"Found library agent by graph_id: {agent.name}")
             return AgentInfo(
@@ -72,7 +71,7 @@ async def _get_library_agent_by_id(user_id: str, agent_id: str) -> AgentInfo | N
         )
 
     try:
-        agent = await lib_db.get_library_agent(agent_id, user_id)
+        agent = await library_db.get_library_agent(agent_id, user_id)
         if agent:
             logger.debug(f"Found library agent by library_id: {agent.name}")
             return AgentInfo(
@@ -134,7 +133,7 @@ async def search_agents(
     try:
         if source == "marketplace":
             logger.info(f"Searching marketplace for: {query}")
-            results = await store_db().get_store_agents(search_query=query, page_size=5)
+            results = await store_db.get_store_agents(search_query=query, page_size=5)
             for agent in results.agents:
                 agents.append(
                     AgentInfo(
@@ -160,7 +159,7 @@ async def search_agents(
 
             if not agents:
                 logger.info(f"Searching user library for: {query}")
-                results = await library_db().list_library_agents(
+                results = await library_db.list_library_agents(
                     user_id=user_id,  # type: ignore[arg-type]
                     search_term=query,
                     page_size=10,

autogpt_platform/backend/backend/api/features/chat/tools/base.py

@@ -5,8 +5,8 @@ from typing import Any
 
 from openai.types.chat import ChatCompletionToolParam
 
-from backend.copilot.model import ChatSession
-from backend.copilot.response_model import StreamToolOutputAvailable
+from backend.api.features.chat.model import ChatSession
+from backend.api.features.chat.response_model import StreamToolOutputAvailable
 
 from .models import ErrorResponse, NeedLoginResponse, ToolResponseBase
 
@@ -36,6 +36,16 @@ class BaseTool:
         """Whether this tool requires authentication."""
         return False
 
+    @property
+    def is_long_running(self) -> bool:
+        """Whether this tool is long-running and should execute in background.
+
+        Long-running tools (like agent generation) are executed via background
+        tasks to survive SSE disconnections. The result is persisted to chat
+        history and visible when the user refreshes.
+        """
+        return False
+
     def as_openai_tool(self) -> ChatCompletionToolParam:
         """Convert to OpenAI tool format."""
         return ChatCompletionToolParam(

autogpt_platform/backend/backend/api/features/chat/tools/create_agent.py

@@ -3,13 +3,14 @@
 import logging
 from typing import Any
 
-from backend.copilot.model import ChatSession
+from backend.api.features.chat.model import ChatSession
 
 from .agent_generator import (
     AgentGeneratorNotConfiguredError,
     decompose_goal,
     enrich_library_agents_from_steps,
     generate_agent,
+    get_all_relevant_agents_for_generation,
     get_user_message_for_error,
     save_agent_to_library,
 )
@@ -17,10 +18,10 @@ from .base import BaseTool
 from .models import (
     AgentPreviewResponse,
     AgentSavedResponse,
+    AsyncProcessingResponse,
     ClarificationNeededResponse,
     ClarifyingQuestion,
     ErrorResponse,
-    SuggestedGoalResponse,
     ToolResponseBase,
 )
 
@@ -38,16 +39,17 @@ class CreateAgentTool(BaseTool):
     def description(self) -> str:
         return (
             "Create a new agent workflow from a natural language description. "
-            "First generates a preview, then saves to library if save=true. "
-            "\n\nIMPORTANT: Before calling this tool, search for relevant existing agents "
-            "using find_library_agent that could be used as building blocks. "
-            "Pass their IDs in the library_agent_ids parameter so the generator can compose them."
+            "First generates a preview, then saves to library if save=true."
         )
 
     @property
     def requires_auth(self) -> bool:
         return True
 
+    @property
+    def is_long_running(self) -> bool:
+        return True
+
     @property
     def parameters(self) -> dict[str, Any]:
         return {
@@ -67,15 +69,6 @@ class CreateAgentTool(BaseTool):
                         "Include any preferences or constraints mentioned by the user."
                     ),
                 },
-                "library_agent_ids": {
-                    "type": "array",
-                    "items": {"type": "string"},
-                    "description": (
-                        "List of library agent IDs to use as building blocks. "
-                        "Search for relevant agents using find_library_agent first, "
-                        "then pass their IDs here so they can be composed into the new agent."
-                    ),
-                },
                 "save": {
                     "type": "boolean",
                     "description": (
@@ -103,14 +96,12 @@ class CreateAgentTool(BaseTool):
         """
         description = kwargs.get("description", "").strip()
         context = kwargs.get("context", "")
-        library_agent_ids = kwargs.get("library_agent_ids", [])
         save = kwargs.get("save", True)
         session_id = session.session_id if session else None
 
-        logger.info(
-            f"[AGENT_CREATE_DEBUG] START - description_len={len(description)}, "
-            f"library_agent_ids={library_agent_ids}, save={save}, user_id={user_id}, session_id={session_id}"
-        )
+        # Extract async processing params (passed by long-running tool handler)
+        operation_id = kwargs.get("_operation_id")
+        task_id = kwargs.get("_task_id")
 
         if not description:
             return ErrorResponse(
@@ -119,34 +110,25 @@ class CreateAgentTool(BaseTool):
                 session_id=session_id,
             )
 
-        # Fetch library agents by IDs if provided
         library_agents = None
-        if user_id and library_agent_ids:
+        if user_id:
             try:
-                from .agent_generator import get_library_agents_by_ids
-
-                library_agents = await get_library_agents_by_ids(
+                library_agents = await get_all_relevant_agents_for_generation(
                     user_id=user_id,
-                    agent_ids=library_agent_ids,
+                    search_query=description,
+                    include_marketplace=True,
                 )
                 logger.debug(
-                    f"Fetched {len(library_agents)} library agents by ID for sub-agent composition"
+                    f"Found {len(library_agents)} relevant agents for sub-agent composition"
                 )
             except Exception as e:
-                logger.warning(f"Failed to fetch library agents by IDs: {e}")
+                logger.warning(f"Failed to fetch library agents: {e}")
 
         try:
             decomposition_result = await decompose_goal(
                 description, context, library_agents
             )
-            logger.info(
-                f"[AGENT_CREATE_DEBUG] DECOMPOSE - type={decomposition_result.get('type') if decomposition_result else None}, "
-                f"session_id={session_id}"
-            )
         except AgentGeneratorNotConfiguredError:
-            logger.error(
-                f"[AGENT_CREATE_DEBUG] ERROR - AgentGeneratorNotConfigured, session_id={session_id}"
-            )
             return ErrorResponse(
                 message=(
                     "Agent generation is not available. "
@@ -204,28 +186,26 @@ class CreateAgentTool(BaseTool):
         if decomposition_result.get("type") == "unachievable_goal":
             suggested = decomposition_result.get("suggested_goal", "")
             reason = decomposition_result.get("reason", "")
-            return SuggestedGoalResponse(
+            return ErrorResponse(
                 message=(
-                    f"This goal cannot be accomplished with the available blocks. {reason}"
+                    f"This goal cannot be accomplished with the available blocks. "
+                    f"{reason} "
+                    f"Suggestion: {suggested}"
                 ),
-                suggested_goal=suggested,
-                reason=reason,
-                original_goal=description,
-                goal_type="unachievable",
+                error="unachievable_goal",
+                details={"suggested_goal": suggested, "reason": reason},
                 session_id=session_id,
             )
 
         if decomposition_result.get("type") == "vague_goal":
             suggested = decomposition_result.get("suggested_goal", "")
-            reason = decomposition_result.get(
-                "reason", "The goal needs more specific details"
-            )
-            return SuggestedGoalResponse(
-                message="The goal is too vague to create a specific workflow.",
-                suggested_goal=suggested,
-                reason=reason,
-                original_goal=description,
-                goal_type="vague",
+            return ErrorResponse(
+                message=(
+                    f"The goal is too vague to create a specific workflow. "
+                    f"Suggestion: {suggested}"
+                ),
+                error="vague_goal",
+                details={"suggested_goal": suggested},
                 session_id=session_id,
             )
 
@@ -247,17 +227,10 @@ class CreateAgentTool(BaseTool):
             agent_json = await generate_agent(
                 decomposition_result,
                 library_agents,
-            )
-            logger.info(
-                f"[AGENT_CREATE_DEBUG] GENERATE - "
-                f"success={agent_json is not None}, "
-                f"is_error={isinstance(agent_json, dict) and agent_json.get('type') == 'error'}, "
-                f"session_id={session_id}"
+                operation_id=operation_id,
+                task_id=task_id,
             )
         except AgentGeneratorNotConfiguredError:
-            logger.error(
-                f"[AGENT_CREATE_DEBUG] ERROR - AgentGeneratorNotConfigured during generation, session_id={session_id}"
-            )
             return ErrorResponse(
                 message=(
                     "Agent generation is not available. "
@@ -300,20 +273,25 @@ class CreateAgentTool(BaseTool):
                 session_id=session_id,
             )
 
+        # Check if Agent Generator accepted for async processing
+        if agent_json.get("status") == "accepted":
+            logger.info(
+                f"Agent generation delegated to async processing "
+                f"(operation_id={operation_id}, task_id={task_id})"
+            )
+            return AsyncProcessingResponse(
+                message="Agent generation started. You'll be notified when it's complete.",
+                operation_id=operation_id,
+                task_id=task_id,
+                session_id=session_id,
+            )
+
         agent_name = agent_json.get("name", "Generated Agent")
         agent_description = agent_json.get("description", "")
         node_count = len(agent_json.get("nodes", []))
         link_count = len(agent_json.get("links", []))
 
-        logger.info(
-            f"[AGENT_CREATE_DEBUG] AGENT_JSON - name={agent_name}, "
-            f"nodes={node_count}, links={link_count}, save={save}, session_id={session_id}"
-        )
-
         if not save:
-            logger.info(
-                f"[AGENT_CREATE_DEBUG] RETURN - AgentPreviewResponse, session_id={session_id}"
-            )
             return AgentPreviewResponse(
                 message=(
                     f"I've generated an agent called '{agent_name}' with {node_count} blocks. "
@@ -339,13 +317,6 @@ class CreateAgentTool(BaseTool):
                 agent_json, user_id
             )
 
-            logger.info(
-                f"[AGENT_CREATE_DEBUG] SAVED - graph_id={created_graph.id}, "
-                f"library_agent_id={library_agent.id}, session_id={session_id}"
-            )
-            logger.info(
-                f"[AGENT_CREATE_DEBUG] RETURN - AgentSavedResponse, session_id={session_id}"
-            )
             return AgentSavedResponse(
                 message=f"Agent '{created_graph.name}' has been saved to your library!",
                 agent_id=created_graph.id,
@@ -356,12 +327,6 @@ class CreateAgentTool(BaseTool):
                 session_id=session_id,
             )
         except Exception as e:
-            logger.error(
-                f"[AGENT_CREATE_DEBUG] ERROR - save_failed: {str(e)}, session_id={session_id}"
-            )
-            logger.info(
-                f"[AGENT_CREATE_DEBUG] RETURN - ErrorResponse (save_failed), session_id={session_id}"
-            )
             return ErrorResponse(
                 message=f"Failed to save the agent: {str(e)}",
                 error="save_failed",

autogpt_platform/backend/backend/api/features/chat/tools/customize_agent.py

@@ -3,9 +3,9 @@
 import logging
 from typing import Any
 
+from backend.api.features.chat.model import ChatSession
+from backend.api.features.store import db as store_db
 from backend.api.features.store.exceptions import AgentNotFoundError
-from backend.copilot.model import ChatSession
-from backend.data.db_accessors import store_db as get_store_db
 
 from .agent_generator import (
     AgentGeneratorNotConfiguredError,
@@ -46,6 +46,10 @@ class CustomizeAgentTool(BaseTool):
     def requires_auth(self) -> bool:
         return True
 
+    @property
+    def is_long_running(self) -> bool:
+        return True
+
     @property
     def parameters(self) -> dict[str, Any]:
         return {
@@ -133,8 +137,6 @@ class CustomizeAgentTool(BaseTool):
 
         creator_username, agent_slug = parts
 
-        store_db = get_store_db()
-
         # Fetch the marketplace agent details
         try:
             agent_details = await store_db.get_store_agent_details(

autogpt_platform/backend/backend/api/features/chat/tools/edit_agent.py

@@ -3,12 +3,13 @@
 import logging
 from typing import Any
 
-from backend.copilot.model import ChatSession
+from backend.api.features.chat.model import ChatSession
 
 from .agent_generator import (
     AgentGeneratorNotConfiguredError,
     generate_agent_patch,
     get_agent_as_json,
+    get_all_relevant_agents_for_generation,
     get_user_message_for_error,
     save_agent_to_library,
 )
@@ -16,6 +17,7 @@ from .base import BaseTool
 from .models import (
     AgentPreviewResponse,
     AgentSavedResponse,
+    AsyncProcessingResponse,
     ClarificationNeededResponse,
     ClarifyingQuestion,
     ErrorResponse,
@@ -36,16 +38,17 @@ class EditAgentTool(BaseTool):
     def description(self) -> str:
         return (
             "Edit an existing agent from the user's library using natural language. "
-            "Generates updates to the agent while preserving unchanged parts. "
-            "\n\nIMPORTANT: Before calling this tool, if the changes involve adding new "
-            "functionality, search for relevant existing agents using find_library_agent "
-            "that could be used as building blocks. Pass their IDs in library_agent_ids."
+            "Generates updates to the agent while preserving unchanged parts."
         )
 
     @property
     def requires_auth(self) -> bool:
         return True
 
+    @property
+    def is_long_running(self) -> bool:
+        return True
+
     @property
     def parameters(self) -> dict[str, Any]:
         return {
@@ -71,15 +74,6 @@ class EditAgentTool(BaseTool):
                         "Additional context or answers to previous clarifying questions."
                     ),
                 },
-                "library_agent_ids": {
-                    "type": "array",
-                    "items": {"type": "string"},
-                    "description": (
-                        "List of library agent IDs to use as building blocks for the changes. "
-                        "If adding new functionality, search for relevant agents using "
-                        "find_library_agent first, then pass their IDs here."
-                    ),
-                },
                 "save": {
                     "type": "boolean",
                     "description": (
@@ -108,10 +102,13 @@ class EditAgentTool(BaseTool):
         agent_id = kwargs.get("agent_id", "").strip()
         changes = kwargs.get("changes", "").strip()
         context = kwargs.get("context", "")
-        library_agent_ids = kwargs.get("library_agent_ids", [])
         save = kwargs.get("save", True)
         session_id = session.session_id if session else None
 
+        # Extract async processing params (passed by long-running tool handler)
+        operation_id = kwargs.get("_operation_id")
+        task_id = kwargs.get("_task_id")
+
         if not agent_id:
             return ErrorResponse(
                 message="Please provide the agent ID to edit.",
@@ -135,25 +132,21 @@ class EditAgentTool(BaseTool):
                 session_id=session_id,
             )
 
-        # Fetch library agents by IDs if provided
         library_agents = None
-        if user_id and library_agent_ids:
+        if user_id:
             try:
-                from .agent_generator import get_library_agents_by_ids
-
                 graph_id = current_agent.get("id")
-                # Filter out the current agent being edited
-                filtered_ids = [id for id in library_agent_ids if id != graph_id]
-
-                library_agents = await get_library_agents_by_ids(
+                library_agents = await get_all_relevant_agents_for_generation(
                     user_id=user_id,
-                    agent_ids=filtered_ids,
+                    search_query=changes,
+                    exclude_graph_id=graph_id,
+                    include_marketplace=True,
                 )
                 logger.debug(
-                    f"Fetched {len(library_agents)} library agents by ID for sub-agent composition"
+                    f"Found {len(library_agents)} relevant agents for sub-agent composition"
                 )
             except Exception as e:
-                logger.warning(f"Failed to fetch library agents by IDs: {e}")
+                logger.warning(f"Failed to fetch library agents: {e}")
 
         update_request = changes
         if context:
@@ -164,6 +157,8 @@ class EditAgentTool(BaseTool):
                 update_request,
                 current_agent,
                 library_agents,
+                operation_id=operation_id,
+                task_id=task_id,
             )
         except AgentGeneratorNotConfiguredError:
             return ErrorResponse(
@@ -183,6 +178,19 @@ class EditAgentTool(BaseTool):
                 session_id=session_id,
             )
 
+        # Check if Agent Generator accepted for async processing
+        if result.get("status") == "accepted":
+            logger.info(
+                f"Agent edit delegated to async processing "
+                f"(operation_id={operation_id}, task_id={task_id})"
+            )
+            return AsyncProcessingResponse(
+                message="Agent edit started. You'll be notified when it's complete.",
+                operation_id=operation_id,
+                task_id=task_id,
+                session_id=session_id,
+            )
+
         # Check if the result is an error from the external service
         if isinstance(result, dict) and result.get("type") == "error":
             error_msg = result.get("error", "Unknown error")

autogpt_platform/backend/backend/api/features/chat/tools/find_agent.py

@@ -2,7 +2,7 @@
 
 from typing import Any
 
-from backend.copilot.model import ChatSession
+from backend.api.features.chat.model import ChatSession
 
 from .agent_search import search_agents
 from .base import BaseTool

autogpt_platform/backend/backend/api/features/chat/tools/find_block.py

@@ -3,18 +3,17 @@ from typing import Any
 
 from prisma.enums import ContentType
 
-from backend.blocks import get_block
-from backend.blocks._base import BlockType
-from backend.copilot.model import ChatSession
-from backend.data.db_accessors import search
-
-from .base import BaseTool, ToolResponseBase
-from .models import (
+from backend.api.features.chat.model import ChatSession
+from backend.api.features.chat.tools.base import BaseTool, ToolResponseBase
+from backend.api.features.chat.tools.models import (
     BlockInfoSummary,
+    BlockInputFieldInfo,
     BlockListResponse,
     ErrorResponse,
     NoResultsResponse,
 )
+from backend.api.features.store.hybrid_search import unified_hybrid_search
+from backend.data.block import BlockType, get_block
 
 logger = logging.getLogger(__name__)
 
@@ -55,8 +54,7 @@ class FindBlockTool(BaseTool):
             "Blocks are reusable components that perform specific tasks like "
             "sending emails, making API calls, processing text, etc. "
             "IMPORTANT: Use this tool FIRST to get the block's 'id' before calling run_block. "
-            "The response includes each block's id, name, and description. "
-            "Call run_block with the block's id **with no inputs** to see detailed inputs/outputs and execute it."
+            "The response includes each block's id, required_inputs, and input_schema."
         )
 
     @property
@@ -108,7 +106,7 @@ class FindBlockTool(BaseTool):
 
         try:
             # Search for blocks using hybrid search
-            results, total = await search().unified_hybrid_search(
+            results, total = await unified_hybrid_search(
                 query=query,
                 content_types=[ContentType.BLOCK],
                 page=1,
@@ -125,7 +123,7 @@ class FindBlockTool(BaseTool):
                     session_id=session_id,
                 )
 
-            # Enrich results with block information
+            # Enrich results with full block information
             blocks: list[BlockInfoSummary] = []
             for result in results:
                 block_id = result["content_id"]
@@ -142,12 +140,65 @@ class FindBlockTool(BaseTool):
                 ):
                     continue
 
+                # Get input/output schemas
+                input_schema = {}
+                output_schema = {}
+                try:
+                    input_schema = block.input_schema.jsonschema()
+                except Exception as e:
+                    logger.debug(
+                        "Failed to generate input schema for block %s: %s",
+                        block_id,
+                        e,
+                    )
+                try:
+                    output_schema = block.output_schema.jsonschema()
+                except Exception as e:
+                    logger.debug(
+                        "Failed to generate output schema for block %s: %s",
+                        block_id,
+                        e,
+                    )
+
+                # Get categories from block instance
+                categories = []
+                if hasattr(block, "categories") and block.categories:
+                    categories = [cat.value for cat in block.categories]
+
+                # Extract required inputs for easier use
+                required_inputs: list[BlockInputFieldInfo] = []
+                if input_schema:
+                    properties = input_schema.get("properties", {})
+                    required_fields = set(input_schema.get("required", []))
+                    # Get credential field names to exclude from required inputs
+                    credentials_fields = set(
+                        block.input_schema.get_credentials_fields().keys()
+                    )
+
+                    for field_name, field_schema in properties.items():
+                        # Skip credential fields - they're handled separately
+                        if field_name in credentials_fields:
+                            continue
+
+                        required_inputs.append(
+                            BlockInputFieldInfo(
+                                name=field_name,
+                                type=field_schema.get("type", "string"),
+                                description=field_schema.get("description", ""),
+                                required=field_name in required_fields,
+                                default=field_schema.get("default"),
+                            )
+                        )
+
                 blocks.append(
                     BlockInfoSummary(
                         id=block_id,
                         name=block.name,
                         description=block.description or "",
-                        categories=[c.value for c in block.categories],
+                        categories=categories,
+                        input_schema=input_schema,
+                        output_schema=output_schema,
+                        required_inputs=required_inputs,
                     )
                 )
 
@@ -176,7 +227,8 @@ class FindBlockTool(BaseTool):
             return BlockListResponse(
                 message=(
                     f"Found {len(blocks)} block(s) matching '{query}'. "
-                    "To see a block's inputs/outputs and execute it, use run_block with the block's 'id' - providing no inputs."
+                    "To execute a block, use run_block with the block's 'id' field "
+                    "and provide 'input_data' matching the block's input_schema."
                 ),
                 blocks=blocks,
                 count=len(blocks),

autogpt_platform/backend/backend/api/features/chat/tools/find_block_test.py

@@ -0,0 +1,139 @@
+"""Tests for block filtering in FindBlockTool."""
+
+from unittest.mock import AsyncMock, MagicMock, patch
+
+import pytest
+
+from backend.api.features.chat.tools.find_block import (
+    COPILOT_EXCLUDED_BLOCK_IDS,
+    COPILOT_EXCLUDED_BLOCK_TYPES,
+    FindBlockTool,
+)
+from backend.api.features.chat.tools.models import BlockListResponse
+from backend.data.block import BlockType
+
+from ._test_data import make_session
+
+_TEST_USER_ID = "test-user-find-block"
+
+
+def make_mock_block(
+    block_id: str, name: str, block_type: BlockType, disabled: bool = False
+):
+    """Create a mock block for testing."""
+    mock = MagicMock()
+    mock.id = block_id
+    mock.name = name
+    mock.description = f"{name} description"
+    mock.block_type = block_type
+    mock.disabled = disabled
+    mock.input_schema = MagicMock()
+    mock.input_schema.jsonschema.return_value = {"properties": {}, "required": []}
+    mock.input_schema.get_credentials_fields.return_value = {}
+    mock.output_schema = MagicMock()
+    mock.output_schema.jsonschema.return_value = {}
+    mock.categories = []
+    return mock
+
+
+class TestFindBlockFiltering:
+    """Tests for block filtering in FindBlockTool."""
+
+    def test_excluded_block_types_contains_expected_types(self):
+        """Verify COPILOT_EXCLUDED_BLOCK_TYPES contains all graph-only types."""
+        assert BlockType.INPUT in COPILOT_EXCLUDED_BLOCK_TYPES
+        assert BlockType.OUTPUT in COPILOT_EXCLUDED_BLOCK_TYPES
+        assert BlockType.WEBHOOK in COPILOT_EXCLUDED_BLOCK_TYPES
+        assert BlockType.WEBHOOK_MANUAL in COPILOT_EXCLUDED_BLOCK_TYPES
+        assert BlockType.NOTE in COPILOT_EXCLUDED_BLOCK_TYPES
+        assert BlockType.HUMAN_IN_THE_LOOP in COPILOT_EXCLUDED_BLOCK_TYPES
+        assert BlockType.AGENT in COPILOT_EXCLUDED_BLOCK_TYPES
+
+    def test_excluded_block_ids_contains_smart_decision_maker(self):
+        """Verify SmartDecisionMakerBlock is in COPILOT_EXCLUDED_BLOCK_IDS."""
+        assert "3b191d9f-356f-482d-8238-ba04b6d18381" in COPILOT_EXCLUDED_BLOCK_IDS
+
+    @pytest.mark.asyncio(loop_scope="session")
+    async def test_excluded_block_type_filtered_from_results(self):
+        """Verify blocks with excluded BlockTypes are filtered from search results."""
+        session = make_session(user_id=_TEST_USER_ID)
+
+        # Mock search returns an INPUT block (excluded) and a STANDARD block (included)
+        search_results = [
+            {"content_id": "input-block-id", "score": 0.9},
+            {"content_id": "standard-block-id", "score": 0.8},
+        ]
+
+        input_block = make_mock_block("input-block-id", "Input Block", BlockType.INPUT)
+        standard_block = make_mock_block(
+            "standard-block-id", "HTTP Request", BlockType.STANDARD
+        )
+
+        def mock_get_block(block_id):
+            return {
+                "input-block-id": input_block,
+                "standard-block-id": standard_block,
+            }.get(block_id)
+
+        with patch(
+            "backend.api.features.chat.tools.find_block.unified_hybrid_search",
+            new_callable=AsyncMock,
+            return_value=(search_results, 2),
+        ):
+            with patch(
+                "backend.api.features.chat.tools.find_block.get_block",
+                side_effect=mock_get_block,
+            ):
+                tool = FindBlockTool()
+                response = await tool._execute(
+                    user_id=_TEST_USER_ID, session=session, query="test"
+                )
+
+        # Should only return the standard block, not the INPUT block
+        assert isinstance(response, BlockListResponse)
+        assert len(response.blocks) == 1
+        assert response.blocks[0].id == "standard-block-id"
+
+    @pytest.mark.asyncio(loop_scope="session")
+    async def test_excluded_block_id_filtered_from_results(self):
+        """Verify SmartDecisionMakerBlock is filtered from search results."""
+        session = make_session(user_id=_TEST_USER_ID)
+
+        smart_decision_id = "3b191d9f-356f-482d-8238-ba04b6d18381"
+        search_results = [
+            {"content_id": smart_decision_id, "score": 0.9},
+            {"content_id": "normal-block-id", "score": 0.8},
+        ]
+
+        # SmartDecisionMakerBlock has STANDARD type but is excluded by ID
+        smart_block = make_mock_block(
+            smart_decision_id, "Smart Decision Maker", BlockType.STANDARD
+        )
+        normal_block = make_mock_block(
+            "normal-block-id", "Normal Block", BlockType.STANDARD
+        )
+
+        def mock_get_block(block_id):
+            return {
+                smart_decision_id: smart_block,
+                "normal-block-id": normal_block,
+            }.get(block_id)
+
+        with patch(
+            "backend.api.features.chat.tools.find_block.unified_hybrid_search",
+            new_callable=AsyncMock,
+            return_value=(search_results, 2),
+        ):
+            with patch(
+                "backend.api.features.chat.tools.find_block.get_block",
+                side_effect=mock_get_block,
+            ):
+                tool = FindBlockTool()
+                response = await tool._execute(
+                    user_id=_TEST_USER_ID, session=session, query="decision"
+                )
+
+        # Should only return normal block, not SmartDecisionMakerBlock
+        assert isinstance(response, BlockListResponse)
+        assert len(response.blocks) == 1
+        assert response.blocks[0].id == "normal-block-id"

autogpt_platform/backend/backend/api/features/chat/tools/find_library_agent.py

@@ -2,7 +2,7 @@
 
 from typing import Any
 
-from backend.copilot.model import ChatSession
+from backend.api.features.chat.model import ChatSession
 
 from .agent_search import search_agents
 from .base import BaseTool

autogpt_platform/backend/backend/api/features/chat/tools/get_doc_page.py

@@ -4,10 +4,13 @@ import logging
 from pathlib import Path
 from typing import Any
 
-from backend.copilot.model import ChatSession
-
-from .base import BaseTool
-from .models import DocPageResponse, ErrorResponse, ToolResponseBase
+from backend.api.features.chat.model import ChatSession
+from backend.api.features.chat.tools.base import BaseTool
+from backend.api.features.chat.tools.models import (
+    DocPageResponse,
+    ErrorResponse,
+    ToolResponseBase,
+)
 
 logger = logging.getLogger(__name__)
 

autogpt_platform/backend/backend/api/features/chat/tools/models.py

@@ -2,7 +2,7 @@
 
 from datetime import datetime
 from enum import Enum
-from typing import Any, Literal
+from typing import Any
 
 from pydantic import BaseModel, Field
 
@@ -25,7 +25,6 @@ class ResponseType(str, Enum):
     AGENT_SAVED = "agent_saved"
     CLARIFICATION_NEEDED = "clarification_needed"
     BLOCK_LIST = "block_list"
-    BLOCK_DETAILS = "block_details"
     BLOCK_OUTPUT = "block_output"
     DOC_SEARCH_RESULTS = "doc_search_results"
     DOC_PAGE = "doc_page"
@@ -36,18 +35,11 @@ class ResponseType(str, Enum):
     WORKSPACE_FILE_WRITTEN = "workspace_file_written"
     WORKSPACE_FILE_DELETED = "workspace_file_deleted"
     # Long-running operation types
+    OPERATION_STARTED = "operation_started"
+    OPERATION_PENDING = "operation_pending"
     OPERATION_IN_PROGRESS = "operation_in_progress"
     # Input validation
     INPUT_VALIDATION_ERROR = "input_validation_error"
-    # Web fetch
-    WEB_FETCH = "web_fetch"
-    # Code execution
-    BASH_EXEC = "bash_exec"
-    # Feature request types
-    FEATURE_REQUEST_SEARCH = "feature_request_search"
-    FEATURE_REQUEST_CREATED = "feature_request_created"
-    # Goal refinement
-    SUGGESTED_GOAL = "suggested_goal"
 
 
 # Base response model
@@ -294,22 +286,6 @@ class ClarificationNeededResponse(ToolResponseBase):
     questions: list[ClarifyingQuestion] = Field(default_factory=list)
 
 
-class SuggestedGoalResponse(ToolResponseBase):
-    """Response when the goal needs refinement with a suggested alternative."""
-
-    type: ResponseType = ResponseType.SUGGESTED_GOAL
-    suggested_goal: str = Field(description="The suggested alternative goal")
-    reason: str = Field(
-        default="", description="Why the original goal needs refinement"
-    )
-    original_goal: str = Field(
-        default="", description="The user's original goal for context"
-    )
-    goal_type: Literal["vague", "unachievable"] = Field(
-        default="vague", description="Type: 'vague' or 'unachievable'"
-    )
-
-
 # Documentation search models
 class DocSearchResult(BaseModel):
     """A single documentation search result."""
@@ -359,17 +335,11 @@ class BlockInfoSummary(BaseModel):
     name: str
     description: str
     categories: list[str]
-    input_schema: dict[str, Any] = Field(
-        default_factory=dict,
-        description="Full JSON schema for block inputs",
-    )
-    output_schema: dict[str, Any] = Field(
-        default_factory=dict,
-        description="Full JSON schema for block outputs",
-    )
+    input_schema: dict[str, Any]
+    output_schema: dict[str, Any]
     required_inputs: list[BlockInputFieldInfo] = Field(
         default_factory=list,
-        description="List of input fields for this block",
+        description="List of required input fields for this block",
     )
 
 
@@ -382,29 +352,10 @@ class BlockListResponse(ToolResponseBase):
     query: str
     usage_hint: str = Field(
         default="To execute a block, call run_block with block_id set to the block's "
-        "'id' field and input_data containing the fields listed in required_inputs."
+        "'id' field and input_data containing the required fields from input_schema."
     )
 
 
-class BlockDetails(BaseModel):
-    """Detailed block information."""
-
-    id: str
-    name: str
-    description: str
-    inputs: dict[str, Any] = {}
-    outputs: dict[str, Any] = {}
-    credentials: list[CredentialsMetaInput] = []
-
-
-class BlockDetailsResponse(ToolResponseBase):
-    """Response for block details (first run_block attempt)."""
-
-    type: ResponseType = ResponseType.BLOCK_DETAILS
-    block: BlockDetails
-    user_authenticated: bool = False
-
-
 class BlockOutputResponse(ToolResponseBase):
     """Response for run_block tool."""
 
@@ -416,6 +367,34 @@ class BlockOutputResponse(ToolResponseBase):
 
 
 # Long-running operation models
+class OperationStartedResponse(ToolResponseBase):
+    """Response when a long-running operation has been started in the background.
+
+    This is returned immediately to the client while the operation continues
+    to execute. The user can close the tab and check back later.
+
+    The task_id can be used to reconnect to the SSE stream via
+    GET /chat/tasks/{task_id}/stream?last_idx=0
+    """
+
+    type: ResponseType = ResponseType.OPERATION_STARTED
+    operation_id: str
+    tool_name: str
+    task_id: str | None = None  # For SSE reconnection
+
+
+class OperationPendingResponse(ToolResponseBase):
+    """Response stored in chat history while a long-running operation is executing.
+
+    This is persisted to the database so users see a pending state when they
+    refresh before the operation completes.
+    """
+
+    type: ResponseType = ResponseType.OPERATION_PENDING
+    operation_id: str
+    tool_name: str
+
+
 class OperationInProgressResponse(ToolResponseBase):
     """Response when an operation is already in progress.
 
@@ -427,52 +406,18 @@ class OperationInProgressResponse(ToolResponseBase):
     tool_call_id: str
 
 
-class WebFetchResponse(ToolResponseBase):
-    """Response for web_fetch tool."""
+class AsyncProcessingResponse(ToolResponseBase):
+    """Response when an operation has been delegated to async processing.
 
-    type: ResponseType = ResponseType.WEB_FETCH
-    url: str
-    status_code: int
-    content_type: str
-    content: str
-    truncated: bool = False
+    This is returned by tools when the external service accepts the request
+    for async processing (HTTP 202 Accepted). The Redis Streams completion
+    consumer will handle the result when the external service completes.
 
+    The status field is specifically "accepted" to allow the long-running tool
+    handler to detect this response and skip LLM continuation.
+    """
 
-class BashExecResponse(ToolResponseBase):
-    """Response for bash_exec tool."""
-
-    type: ResponseType = ResponseType.BASH_EXEC
-    stdout: str
-    stderr: str
-    exit_code: int
-    timed_out: bool = False
-
-
-# Feature request models
-class FeatureRequestInfo(BaseModel):
-    """Information about a feature request issue."""
-
-    id: str
-    identifier: str
-    title: str
-
-
-class FeatureRequestSearchResponse(ToolResponseBase):
-    """Response for search_feature_requests tool."""
-
-    type: ResponseType = ResponseType.FEATURE_REQUEST_SEARCH
-    results: list[FeatureRequestInfo]
-    count: int
-    query: str
-
-
-class FeatureRequestCreatedResponse(ToolResponseBase):
-    """Response for create_feature_request tool."""
-
-    type: ResponseType = ResponseType.FEATURE_REQUEST_CREATED
-    issue_id: str
-    issue_identifier: str
-    issue_title: str
-    issue_url: str
-    is_new_issue: bool  # False if added to existing
-    customer_name: str
+    type: ResponseType = ResponseType.OPERATION_STARTED
+    status: str = "accepted"  # Must be "accepted" for detection
+    operation_id: str | None = None
+    task_id: str | None = None

autogpt_platform/backend/backend/api/features/chat/tools/run_agent.py

@@ -5,12 +5,16 @@ from typing import Any
 
 from pydantic import BaseModel, Field, field_validator
 
-from backend.copilot.config import ChatConfig
-from backend.copilot.model import ChatSession
-from backend.copilot.tracking import track_agent_run_success, track_agent_scheduled
-from backend.data.db_accessors import graph_db, library_db, user_db
+from backend.api.features.chat.config import ChatConfig
+from backend.api.features.chat.model import ChatSession
+from backend.api.features.chat.tracking import (
+    track_agent_run_success,
+    track_agent_scheduled,
+)
+from backend.api.features.library import db as library_db
 from backend.data.graph import GraphModel
 from backend.data.model import CredentialsMetaInput
+from backend.data.user import get_user_by_id
 from backend.executor import utils as execution_utils
 from backend.util.clients import get_scheduler_client
 from backend.util.exceptions import DatabaseError, NotFoundError
@@ -196,7 +200,7 @@ class RunAgentTool(BaseTool):
 
             # Priority: library_agent_id if provided
             if has_library_id:
-                library_agent = await library_db().get_library_agent(
+                library_agent = await library_db.get_library_agent(
                     params.library_agent_id, user_id
                 )
                 if not library_agent:
@@ -205,7 +209,9 @@ class RunAgentTool(BaseTool):
                         session_id=session_id,
                     )
                 # Get the graph from the library agent
-                graph = await graph_db().get_graph(
+                from backend.data.graph import get_graph
+
+                graph = await get_graph(
                     library_agent.graph_id,
                     library_agent.graph_version,
                     user_id=user_id,
@@ -516,7 +522,7 @@ class RunAgentTool(BaseTool):
         library_agent = await get_or_create_library_agent(graph, user_id)
 
         # Get user timezone
-        user = await user_db().get_user_by_id(user_id)
+        user = await get_user_by_id(user_id)
         user_timezone = get_user_timezone_or_utc(user.timezone if user else timezone)
 
         # Create schedule

autogpt_platform/backend/backend/api/features/chat/tools/run_block.py

@@ -7,24 +7,23 @@ from typing import Any
 
 from pydantic_core import PydanticUndefined
 
-from backend.blocks import get_block
-from backend.blocks._base import AnyBlockSchema
-from backend.copilot.model import ChatSession
-from backend.data.db_accessors import workspace_db
+from backend.api.features.chat.model import ChatSession
+from backend.api.features.chat.tools.find_block import (
+    COPILOT_EXCLUDED_BLOCK_IDS,
+    COPILOT_EXCLUDED_BLOCK_TYPES,
+)
+from backend.data.block import AnyBlockSchema, get_block
 from backend.data.execution import ExecutionContext
 from backend.data.model import CredentialsFieldInfo, CredentialsMetaInput
+from backend.data.workspace import get_or_create_workspace
 from backend.integrations.creds_manager import IntegrationCredentialsManager
 from backend.util.exceptions import BlockError
 
 from .base import BaseTool
-from .find_block import COPILOT_EXCLUDED_BLOCK_IDS, COPILOT_EXCLUDED_BLOCK_TYPES
 from .helpers import get_inputs_from_schema
 from .models import (
-    BlockDetails,
-    BlockDetailsResponse,
     BlockOutputResponse,
     ErrorResponse,
-    InputValidationErrorResponse,
     SetupInfo,
     SetupRequirementsResponse,
     ToolResponseBase,
@@ -51,8 +50,8 @@ class RunBlockTool(BaseTool):
             "Execute a specific block with the provided input data. "
             "IMPORTANT: You MUST call find_block first to get the block's 'id' - "
             "do NOT guess or make up block IDs. "
-            "On first attempt (without input_data), returns detailed schema showing "
-            "required inputs and outputs. Then call again with proper input_data to execute."
+            "Use the 'id' from find_block results and provide input_data "
+            "matching the block's required_inputs."
         )
 
     @property
@@ -67,19 +66,11 @@ class RunBlockTool(BaseTool):
                         "NEVER guess this - always get it from find_block first."
                     ),
                 },
-                "block_name": {
-                    "type": "string",
-                    "description": (
-                        "The block's human-readable name from find_block results. "
-                        "Used for display purposes in the UI."
-                    ),
-                },
                 "input_data": {
                     "type": "object",
                     "description": (
-                        "Input values for the block. "
-                        "First call with empty {} to see the block's schema, "
-                        "then call again with proper values to execute."
+                        "Input values for the block. Use the 'required_inputs' field "
+                        "from find_block to see what fields are needed."
                     ),
                 },
             },
@@ -160,38 +151,9 @@ class RunBlockTool(BaseTool):
         logger.info(f"Executing block {block.name} ({block_id}) for user {user_id}")
 
         creds_manager = IntegrationCredentialsManager()
-        (
-            matched_credentials,
-            missing_credentials,
-        ) = await self._resolve_block_credentials(user_id, block, input_data)
-
-        # Get block schemas for details/validation
-        try:
-            input_schema: dict[str, Any] = block.input_schema.jsonschema()
-        except Exception as e:
-            logger.warning(
-                "Failed to generate input schema for block %s: %s",
-                block_id,
-                e,
-            )
-            return ErrorResponse(
-                message=f"Block '{block.name}' has an invalid input schema",
-                error=str(e),
-                session_id=session_id,
-            )
-        try:
-            output_schema: dict[str, Any] = block.output_schema.jsonschema()
-        except Exception as e:
-            logger.warning(
-                "Failed to generate output schema for block %s: %s",
-                block_id,
-                e,
-            )
-            return ErrorResponse(
-                message=f"Block '{block.name}' has an invalid output schema",
-                error=str(e),
-                session_id=session_id,
-            )
+        matched_credentials, missing_credentials = (
+            await self._resolve_block_credentials(user_id, block, input_data)
+        )
 
         if missing_credentials:
             # Return setup requirements response with missing credentials
@@ -225,56 +187,9 @@ class RunBlockTool(BaseTool):
                 graph_version=None,
             )
 
-        # Check if this is a first attempt (required inputs missing)
-        # Return block details so user can see what inputs are needed
-        credentials_fields = set(block.input_schema.get_credentials_fields().keys())
-        required_keys = set(input_schema.get("required", []))
-        required_non_credential_keys = required_keys - credentials_fields
-        provided_input_keys = set(input_data.keys()) - credentials_fields
-
-        # Check for unknown input fields
-        valid_fields = (
-            set(input_schema.get("properties", {}).keys()) - credentials_fields
-        )
-        unrecognized_fields = provided_input_keys - valid_fields
-        if unrecognized_fields:
-            return InputValidationErrorResponse(
-                message=(
-                    f"Unknown input field(s) provided: {', '.join(sorted(unrecognized_fields))}. "
-                    f"Block was not executed. Please use the correct field names from the schema."
-                ),
-                session_id=session_id,
-                unrecognized_fields=sorted(unrecognized_fields),
-                inputs=input_schema,
-            )
-
-        # Show details when not all required non-credential inputs are provided
-        if not (required_non_credential_keys <= provided_input_keys):
-            # Get credentials info for the response
-            credentials_meta = []
-            for field_name, cred_meta in matched_credentials.items():
-                credentials_meta.append(cred_meta)
-
-            return BlockDetailsResponse(
-                message=(
-                    f"Block '{block.name}' details. "
-                    "Provide input_data matching the inputs schema to execute the block."
-                ),
-                session_id=session_id,
-                block=BlockDetails(
-                    id=block_id,
-                    name=block.name,
-                    description=block.description or "",
-                    inputs=input_schema,
-                    outputs=output_schema,
-                    credentials=credentials_meta,
-                ),
-                user_authenticated=True,
-            )
-
         try:
             # Get or create user's workspace for CoPilot file operations
-            workspace = await workspace_db().get_or_create_workspace(user_id)
+            workspace = await get_or_create_workspace(user_id)
 
             # Generate synthetic IDs for CoPilot context
             # Each chat session is treated as its own agent with one continuous run

autogpt_platform/backend/backend/api/features/chat/tools/run_block_test.py

@@ -0,0 +1,106 @@
+"""Tests for block execution guards in RunBlockTool."""
+
+from unittest.mock import MagicMock, patch
+
+import pytest
+
+from backend.api.features.chat.tools.models import ErrorResponse
+from backend.api.features.chat.tools.run_block import RunBlockTool
+from backend.data.block import BlockType
+
+from ._test_data import make_session
+
+_TEST_USER_ID = "test-user-run-block"
+
+
+def make_mock_block(
+    block_id: str, name: str, block_type: BlockType, disabled: bool = False
+):
+    """Create a mock block for testing."""
+    mock = MagicMock()
+    mock.id = block_id
+    mock.name = name
+    mock.block_type = block_type
+    mock.disabled = disabled
+    mock.input_schema = MagicMock()
+    mock.input_schema.jsonschema.return_value = {"properties": {}, "required": []}
+    mock.input_schema.get_credentials_fields_info.return_value = []
+    return mock
+
+
+class TestRunBlockFiltering:
+    """Tests for block execution guards in RunBlockTool."""
+
+    @pytest.mark.asyncio(loop_scope="session")
+    async def test_excluded_block_type_returns_error(self):
+        """Attempting to execute a block with excluded BlockType returns error."""
+        session = make_session(user_id=_TEST_USER_ID)
+
+        input_block = make_mock_block("input-block-id", "Input Block", BlockType.INPUT)
+
+        with patch(
+            "backend.api.features.chat.tools.run_block.get_block",
+            return_value=input_block,
+        ):
+            tool = RunBlockTool()
+            response = await tool._execute(
+                user_id=_TEST_USER_ID,
+                session=session,
+                block_id="input-block-id",
+                input_data={},
+            )
+
+        assert isinstance(response, ErrorResponse)
+        assert "cannot be run directly in CoPilot" in response.message
+        assert "designed for use within graphs only" in response.message
+
+    @pytest.mark.asyncio(loop_scope="session")
+    async def test_excluded_block_id_returns_error(self):
+        """Attempting to execute SmartDecisionMakerBlock returns error."""
+        session = make_session(user_id=_TEST_USER_ID)
+
+        smart_decision_id = "3b191d9f-356f-482d-8238-ba04b6d18381"
+        smart_block = make_mock_block(
+            smart_decision_id, "Smart Decision Maker", BlockType.STANDARD
+        )
+
+        with patch(
+            "backend.api.features.chat.tools.run_block.get_block",
+            return_value=smart_block,
+        ):
+            tool = RunBlockTool()
+            response = await tool._execute(
+                user_id=_TEST_USER_ID,
+                session=session,
+                block_id=smart_decision_id,
+                input_data={},
+            )
+
+        assert isinstance(response, ErrorResponse)
+        assert "cannot be run directly in CoPilot" in response.message
+
+    @pytest.mark.asyncio(loop_scope="session")
+    async def test_non_excluded_block_passes_guard(self):
+        """Non-excluded blocks pass the filtering guard (may fail later for other reasons)."""
+        session = make_session(user_id=_TEST_USER_ID)
+
+        standard_block = make_mock_block(
+            "standard-id", "HTTP Request", BlockType.STANDARD
+        )
+
+        with patch(
+            "backend.api.features.chat.tools.run_block.get_block",
+            return_value=standard_block,
+        ):
+            tool = RunBlockTool()
+            response = await tool._execute(
+                user_id=_TEST_USER_ID,
+                session=session,
+                block_id="standard-id",
+                input_data={},
+            )
+
+        # Should NOT be an ErrorResponse about CoPilot exclusion
+        # (may be other errors like missing credentials, but not the exclusion guard)
+        if isinstance(response, ErrorResponse):
+            assert "cannot be run directly in CoPilot" not in response.message

autogpt_platform/backend/backend/api/features/chat/tools/search_docs.py

@@ -5,17 +5,16 @@ from typing import Any
 
 from prisma.enums import ContentType
 
-from backend.copilot.model import ChatSession
-from backend.data.db_accessors import search
-
-from .base import BaseTool
-from .models import (
+from backend.api.features.chat.model import ChatSession
+from backend.api.features.chat.tools.base import BaseTool
+from backend.api.features.chat.tools.models import (
     DocSearchResult,
     DocSearchResultsResponse,
     ErrorResponse,
     NoResultsResponse,
     ToolResponseBase,
 )
+from backend.api.features.store.hybrid_search import unified_hybrid_search
 
 logger = logging.getLogger(__name__)
 
@@ -118,7 +117,7 @@ class SearchDocsTool(BaseTool):
 
         try:
             # Search using hybrid search for DOCUMENTATION content type only
-            results, total = await search().unified_hybrid_search(
+            results, total = await unified_hybrid_search(
                 query=query,
                 content_types=[ContentType.DOCUMENTATION],
                 page=1,

autogpt_platform/backend/backend/api/features/chat/tools/utils.py

@@ -3,8 +3,9 @@
 import logging
 from typing import Any
 
+from backend.api.features.library import db as library_db
 from backend.api.features.library import model as library_model
-from backend.data.db_accessors import library_db, store_db
+from backend.api.features.store import db as store_db
 from backend.data.graph import GraphModel
 from backend.data.model import (
     Credentials,
@@ -14,7 +15,6 @@ from backend.data.model import (
     OAuth2Credentials,
 )
 from backend.integrations.creds_manager import IntegrationCredentialsManager
-from backend.integrations.providers import ProviderName
 from backend.util.exceptions import NotFoundError
 
 logger = logging.getLogger(__name__)
@@ -38,14 +38,13 @@ async def fetch_graph_from_store_slug(
     Raises:
         DatabaseError: If there's a database error during lookup.
     """
-    sdb = store_db()
     try:
-        store_agent = await sdb.get_store_agent_details(username, agent_name)
+        store_agent = await store_db.get_store_agent_details(username, agent_name)
     except NotFoundError:
         return None, None
 
     # Get the graph from store listing version
-    graph = await sdb.get_available_graph(
+    graph = await store_db.get_available_graph(
         store_agent.store_listing_version_id, hide_nodes=False
     )
     return graph, store_agent
@@ -210,13 +209,13 @@ async def get_or_create_library_agent(
     Returns:
         LibraryAgent instance
     """
-    existing = await library_db().get_library_agent_by_graph_id(
+    existing = await library_db.get_library_agent_by_graph_id(
         graph_id=graph.id, user_id=user_id
     )
     if existing:
         return existing
 
-    library_agents = await library_db().create_library_agent(
+    library_agents = await library_db.create_library_agent(
         graph=graph,
         user_id=user_id,
         create_library_agents_for_sub_graphs=False,
@@ -360,7 +359,7 @@ async def match_user_credentials_to_graph(
         _,
         _,
     ) in aggregated_creds.items():
-        # Find first matching credential by provider, type, scopes, and host/URL
+        # Find first matching credential by provider, type, and scopes
         matching_cred = next(
             (
                 cred
@@ -375,10 +374,6 @@ async def match_user_credentials_to_graph(
                     cred.type != "host_scoped"
                     or _credential_is_for_host(cred, credential_requirements)
                 )
-                and (
-                    cred.provider != ProviderName.MCP
-                    or _credential_is_for_mcp_server(cred, credential_requirements)
-                )
             ),
             None,
         )
@@ -449,22 +444,6 @@ def _credential_is_for_host(
     return credential.matches_url(list(requirements.discriminator_values)[0])
 
 
-def _credential_is_for_mcp_server(
-    credential: Credentials,
-    requirements: CredentialsFieldInfo,
-) -> bool:
-    """Check if an MCP OAuth credential matches the required server URL."""
-    if not requirements.discriminator_values:
-        return True
-
-    server_url = (
-        credential.metadata.get("mcp_server_url")
-        if isinstance(credential, OAuth2Credentials)
-        else None
-    )
-    return server_url in requirements.discriminator_values if server_url else False
-
-
 async def check_user_has_required_credentials(
     user_id: str,
     required_credentials: list[CredentialsMetaInput],

autogpt_platform/backend/backend/api/features/chat/tools/workspace_files.py

@@ -0,0 +1,620 @@
+"""CoPilot tools for workspace file operations."""
+
+import base64
+import logging
+from typing import Any, Optional
+
+from pydantic import BaseModel
+
+from backend.api.features.chat.model import ChatSession
+from backend.data.workspace import get_or_create_workspace
+from backend.util.settings import Config
+from backend.util.virus_scanner import scan_content_safe
+from backend.util.workspace import WorkspaceManager
+
+from .base import BaseTool
+from .models import ErrorResponse, ResponseType, ToolResponseBase
+
+logger = logging.getLogger(__name__)
+
+
+class WorkspaceFileInfoData(BaseModel):
+    """Data model for workspace file information (not a response itself)."""
+
+    file_id: str
+    name: str
+    path: str
+    mime_type: str
+    size_bytes: int
+
+
+class WorkspaceFileListResponse(ToolResponseBase):
+    """Response containing list of workspace files."""
+
+    type: ResponseType = ResponseType.WORKSPACE_FILE_LIST
+    files: list[WorkspaceFileInfoData]
+    total_count: int
+
+
+class WorkspaceFileContentResponse(ToolResponseBase):
+    """Response containing workspace file content (legacy, for small text files)."""
+
+    type: ResponseType = ResponseType.WORKSPACE_FILE_CONTENT
+    file_id: str
+    name: str
+    path: str
+    mime_type: str
+    content_base64: str
+
+
+class WorkspaceFileMetadataResponse(ToolResponseBase):
+    """Response containing workspace file metadata and download URL (prevents context bloat)."""
+
+    type: ResponseType = ResponseType.WORKSPACE_FILE_METADATA
+    file_id: str
+    name: str
+    path: str
+    mime_type: str
+    size_bytes: int
+    download_url: str
+    preview: str | None = None  # First 500 chars for text files
+
+
+class WorkspaceWriteResponse(ToolResponseBase):
+    """Response after writing a file to workspace."""
+
+    type: ResponseType = ResponseType.WORKSPACE_FILE_WRITTEN
+    file_id: str
+    name: str
+    path: str
+    size_bytes: int
+
+
+class WorkspaceDeleteResponse(ToolResponseBase):
+    """Response after deleting a file from workspace."""
+
+    type: ResponseType = ResponseType.WORKSPACE_FILE_DELETED
+    file_id: str
+    success: bool
+
+
+class ListWorkspaceFilesTool(BaseTool):
+    """Tool for listing files in user's workspace."""
+
+    @property
+    def name(self) -> str:
+        return "list_workspace_files"
+
+    @property
+    def description(self) -> str:
+        return (
+            "List files in the user's workspace. "
+            "Returns file names, paths, sizes, and metadata. "
+            "Optionally filter by path prefix."
+        )
+
+    @property
+    def parameters(self) -> dict[str, Any]:
+        return {
+            "type": "object",
+            "properties": {
+                "path_prefix": {
+                    "type": "string",
+                    "description": (
+                        "Optional path prefix to filter files "
+                        "(e.g., '/documents/' to list only files in documents folder). "
+                        "By default, only files from the current session are listed."
+                    ),
+                },
+                "limit": {
+                    "type": "integer",
+                    "description": "Maximum number of files to return (default 50, max 100)",
+                    "minimum": 1,
+                    "maximum": 100,
+                },
+                "include_all_sessions": {
+                    "type": "boolean",
+                    "description": (
+                        "If true, list files from all sessions. "
+                        "Default is false (only current session's files)."
+                    ),
+                },
+            },
+            "required": [],
+        }
+
+    @property
+    def requires_auth(self) -> bool:
+        return True
+
+    async def _execute(
+        self,
+        user_id: str | None,
+        session: ChatSession,
+        **kwargs,
+    ) -> ToolResponseBase:
+        session_id = session.session_id
+
+        if not user_id:
+            return ErrorResponse(
+                message="Authentication required",
+                session_id=session_id,
+            )
+
+        path_prefix: Optional[str] = kwargs.get("path_prefix")
+        limit = min(kwargs.get("limit", 50), 100)
+        include_all_sessions: bool = kwargs.get("include_all_sessions", False)
+
+        try:
+            workspace = await get_or_create_workspace(user_id)
+            # Pass session_id for session-scoped file access
+            manager = WorkspaceManager(user_id, workspace.id, session_id)
+
+            files = await manager.list_files(
+                path=path_prefix,
+                limit=limit,
+                include_all_sessions=include_all_sessions,
+            )
+            total = await manager.get_file_count(
+                path=path_prefix,
+                include_all_sessions=include_all_sessions,
+            )
+
+            file_infos = [
+                WorkspaceFileInfoData(
+                    file_id=f.id,
+                    name=f.name,
+                    path=f.path,
+                    mime_type=f.mimeType,
+                    size_bytes=f.sizeBytes,
+                )
+                for f in files
+            ]
+
+            scope_msg = "all sessions" if include_all_sessions else "current session"
+            return WorkspaceFileListResponse(
+                files=file_infos,
+                total_count=total,
+                message=f"Found {len(files)} files in workspace ({scope_msg})",
+                session_id=session_id,
+            )
+
+        except Exception as e:
+            logger.error(f"Error listing workspace files: {e}", exc_info=True)
+            return ErrorResponse(
+                message=f"Failed to list workspace files: {str(e)}",
+                error=str(e),
+                session_id=session_id,
+            )
+
+
+class ReadWorkspaceFileTool(BaseTool):
+    """Tool for reading file content from workspace."""
+
+    # Size threshold for returning full content vs metadata+URL
+    # Files larger than this return metadata with download URL to prevent context bloat
+    MAX_INLINE_SIZE_BYTES = 32 * 1024  # 32KB
+    # Preview size for text files
+    PREVIEW_SIZE = 500
+
+    @property
+    def name(self) -> str:
+        return "read_workspace_file"
+
+    @property
+    def description(self) -> str:
+        return (
+            "Read a file from the user's workspace. "
+            "Specify either file_id or path to identify the file. "
+            "For small text files, returns content directly. "
+            "For large or binary files, returns metadata and a download URL. "
+            "Paths are scoped to the current session by default. "
+            "Use /sessions/<session_id>/... for cross-session access."
+        )
+
+    @property
+    def parameters(self) -> dict[str, Any]:
+        return {
+            "type": "object",
+            "properties": {
+                "file_id": {
+                    "type": "string",
+                    "description": "The file's unique ID (from list_workspace_files)",
+                },
+                "path": {
+                    "type": "string",
+                    "description": (
+                        "The virtual file path (e.g., '/documents/report.pdf'). "
+                        "Scoped to current session by default."
+                    ),
+                },
+                "force_download_url": {
+                    "type": "boolean",
+                    "description": (
+                        "If true, always return metadata+URL instead of inline content. "
+                        "Default is false (auto-selects based on file size/type)."
+                    ),
+                },
+            },
+            "required": [],  # At least one must be provided
+        }
+
+    @property
+    def requires_auth(self) -> bool:
+        return True
+
+    def _is_text_mime_type(self, mime_type: str) -> bool:
+        """Check if the MIME type is a text-based type."""
+        text_types = [
+            "text/",
+            "application/json",
+            "application/xml",
+            "application/javascript",
+            "application/x-python",
+            "application/x-sh",
+        ]
+        return any(mime_type.startswith(t) for t in text_types)
+
+    async def _execute(
+        self,
+        user_id: str | None,
+        session: ChatSession,
+        **kwargs,
+    ) -> ToolResponseBase:
+        session_id = session.session_id
+
+        if not user_id:
+            return ErrorResponse(
+                message="Authentication required",
+                session_id=session_id,
+            )
+
+        file_id: Optional[str] = kwargs.get("file_id")
+        path: Optional[str] = kwargs.get("path")
+        force_download_url: bool = kwargs.get("force_download_url", False)
+
+        if not file_id and not path:
+            return ErrorResponse(
+                message="Please provide either file_id or path",
+                session_id=session_id,
+            )
+
+        try:
+            workspace = await get_or_create_workspace(user_id)
+            # Pass session_id for session-scoped file access
+            manager = WorkspaceManager(user_id, workspace.id, session_id)
+
+            # Get file info
+            if file_id:
+                file_info = await manager.get_file_info(file_id)
+                if file_info is None:
+                    return ErrorResponse(
+                        message=f"File not found: {file_id}",
+                        session_id=session_id,
+                    )
+                target_file_id = file_id
+            else:
+                # path is guaranteed to be non-None here due to the check above
+                assert path is not None
+                file_info = await manager.get_file_info_by_path(path)
+                if file_info is None:
+                    return ErrorResponse(
+                        message=f"File not found at path: {path}",
+                        session_id=session_id,
+                    )
+                target_file_id = file_info.id
+
+            # Decide whether to return inline content or metadata+URL
+            is_small_file = file_info.sizeBytes <= self.MAX_INLINE_SIZE_BYTES
+            is_text_file = self._is_text_mime_type(file_info.mimeType)
+
+            # Return inline content for small text files (unless force_download_url)
+            if is_small_file and is_text_file and not force_download_url:
+                content = await manager.read_file_by_id(target_file_id)
+                content_b64 = base64.b64encode(content).decode("utf-8")
+
+                return WorkspaceFileContentResponse(
+                    file_id=file_info.id,
+                    name=file_info.name,
+                    path=file_info.path,
+                    mime_type=file_info.mimeType,
+                    content_base64=content_b64,
+                    message=f"Successfully read file: {file_info.name}",
+                    session_id=session_id,
+                )
+
+            # Return metadata + workspace:// reference for large or binary files
+            # This prevents context bloat (100KB file = ~133KB as base64)
+            # Use workspace:// format so frontend urlTransform can add proxy prefix
+            download_url = f"workspace://{target_file_id}"
+
+            # Generate preview for text files
+            preview: str | None = None
+            if is_text_file:
+                try:
+                    content = await manager.read_file_by_id(target_file_id)
+                    preview_text = content[: self.PREVIEW_SIZE].decode(
+                        "utf-8", errors="replace"
+                    )
+                    if len(content) > self.PREVIEW_SIZE:
+                        preview_text += "..."
+                    preview = preview_text
+                except Exception:
+                    pass  # Preview is optional
+
+            return WorkspaceFileMetadataResponse(
+                file_id=file_info.id,
+                name=file_info.name,
+                path=file_info.path,
+                mime_type=file_info.mimeType,
+                size_bytes=file_info.sizeBytes,
+                download_url=download_url,
+                preview=preview,
+                message=f"File: {file_info.name} ({file_info.sizeBytes} bytes). Use download_url to retrieve content.",
+                session_id=session_id,
+            )
+
+        except FileNotFoundError as e:
+            return ErrorResponse(
+                message=str(e),
+                session_id=session_id,
+            )
+        except Exception as e:
+            logger.error(f"Error reading workspace file: {e}", exc_info=True)
+            return ErrorResponse(
+                message=f"Failed to read workspace file: {str(e)}",
+                error=str(e),
+                session_id=session_id,
+            )
+
+
+class WriteWorkspaceFileTool(BaseTool):
+    """Tool for writing files to workspace."""
+
+    @property
+    def name(self) -> str:
+        return "write_workspace_file"
+
+    @property
+    def description(self) -> str:
+        return (
+            "Write or create a file in the user's workspace. "
+            "Provide the content as a base64-encoded string. "
+            f"Maximum file size is {Config().max_file_size_mb}MB. "
+            "Files are saved to the current session's folder by default. "
+            "Use /sessions/<session_id>/... for cross-session access."
+        )
+
+    @property
+    def parameters(self) -> dict[str, Any]:
+        return {
+            "type": "object",
+            "properties": {
+                "filename": {
+                    "type": "string",
+                    "description": "Name for the file (e.g., 'report.pdf')",
+                },
+                "content_base64": {
+                    "type": "string",
+                    "description": "Base64-encoded file content",
+                },
+                "path": {
+                    "type": "string",
+                    "description": (
+                        "Optional virtual path where to save the file "
+                        "(e.g., '/documents/report.pdf'). "
+                        "Defaults to '/{filename}'. Scoped to current session."
+                    ),
+                },
+                "mime_type": {
+                    "type": "string",
+                    "description": (
+                        "Optional MIME type of the file. "
+                        "Auto-detected from filename if not provided."
+                    ),
+                },
+                "overwrite": {
+                    "type": "boolean",
+                    "description": "Whether to overwrite if file exists at path (default: false)",
+                },
+            },
+            "required": ["filename", "content_base64"],
+        }
+
+    @property
+    def requires_auth(self) -> bool:
+        return True
+
+    async def _execute(
+        self,
+        user_id: str | None,
+        session: ChatSession,
+        **kwargs,
+    ) -> ToolResponseBase:
+        session_id = session.session_id
+
+        if not user_id:
+            return ErrorResponse(
+                message="Authentication required",
+                session_id=session_id,
+            )
+
+        filename: str = kwargs.get("filename", "")
+        content_b64: str = kwargs.get("content_base64", "")
+        path: Optional[str] = kwargs.get("path")
+        mime_type: Optional[str] = kwargs.get("mime_type")
+        overwrite: bool = kwargs.get("overwrite", False)
+
+        if not filename:
+            return ErrorResponse(
+                message="Please provide a filename",
+                session_id=session_id,
+            )
+
+        if not content_b64:
+            return ErrorResponse(
+                message="Please provide content_base64",
+                session_id=session_id,
+            )
+
+        # Decode content
+        try:
+            content = base64.b64decode(content_b64)
+        except Exception:
+            return ErrorResponse(
+                message="Invalid base64-encoded content",
+                session_id=session_id,
+            )
+
+        # Check size
+        max_file_size = Config().max_file_size_mb * 1024 * 1024
+        if len(content) > max_file_size:
+            return ErrorResponse(
+                message=f"File too large. Maximum size is {Config().max_file_size_mb}MB",
+                session_id=session_id,
+            )
+
+        try:
+            # Virus scan
+            await scan_content_safe(content, filename=filename)
+
+            workspace = await get_or_create_workspace(user_id)
+            # Pass session_id for session-scoped file access
+            manager = WorkspaceManager(user_id, workspace.id, session_id)
+
+            file_record = await manager.write_file(
+                content=content,
+                filename=filename,
+                path=path,
+                mime_type=mime_type,
+                overwrite=overwrite,
+            )
+
+            return WorkspaceWriteResponse(
+                file_id=file_record.id,
+                name=file_record.name,
+                path=file_record.path,
+                size_bytes=file_record.sizeBytes,
+                message=f"Successfully wrote file: {file_record.name}",
+                session_id=session_id,
+            )
+
+        except ValueError as e:
+            return ErrorResponse(
+                message=str(e),
+                session_id=session_id,
+            )
+        except Exception as e:
+            logger.error(f"Error writing workspace file: {e}", exc_info=True)
+            return ErrorResponse(
+                message=f"Failed to write workspace file: {str(e)}",
+                error=str(e),
+                session_id=session_id,
+            )
+
+
+class DeleteWorkspaceFileTool(BaseTool):
+    """Tool for deleting files from workspace."""
+
+    @property
+    def name(self) -> str:
+        return "delete_workspace_file"
+
+    @property
+    def description(self) -> str:
+        return (
+            "Delete a file from the user's workspace. "
+            "Specify either file_id or path to identify the file. "
+            "Paths are scoped to the current session by default. "
+            "Use /sessions/<session_id>/... for cross-session access."
+        )
+
+    @property
+    def parameters(self) -> dict[str, Any]:
+        return {
+            "type": "object",
+            "properties": {
+                "file_id": {
+                    "type": "string",
+                    "description": "The file's unique ID (from list_workspace_files)",
+                },
+                "path": {
+                    "type": "string",
+                    "description": (
+                        "The virtual file path (e.g., '/documents/report.pdf'). "
+                        "Scoped to current session by default."
+                    ),
+                },
+            },
+            "required": [],  # At least one must be provided
+        }
+
+    @property
+    def requires_auth(self) -> bool:
+        return True
+
+    async def _execute(
+        self,
+        user_id: str | None,
+        session: ChatSession,
+        **kwargs,
+    ) -> ToolResponseBase:
+        session_id = session.session_id
+
+        if not user_id:
+            return ErrorResponse(
+                message="Authentication required",
+                session_id=session_id,
+            )
+
+        file_id: Optional[str] = kwargs.get("file_id")
+        path: Optional[str] = kwargs.get("path")
+
+        if not file_id and not path:
+            return ErrorResponse(
+                message="Please provide either file_id or path",
+                session_id=session_id,
+            )
+
+        try:
+            workspace = await get_or_create_workspace(user_id)
+            # Pass session_id for session-scoped file access
+            manager = WorkspaceManager(user_id, workspace.id, session_id)
+
+            # Determine the file_id to delete
+            target_file_id: str
+            if file_id:
+                target_file_id = file_id
+            else:
+                # path is guaranteed to be non-None here due to the check above
+                assert path is not None
+                file_info = await manager.get_file_info_by_path(path)
+                if file_info is None:
+                    return ErrorResponse(
+                        message=f"File not found at path: {path}",
+                        session_id=session_id,
+                    )
+                target_file_id = file_info.id
+
+            success = await manager.delete_file(target_file_id)
+
+            if not success:
+                return ErrorResponse(
+                    message=f"File not found: {target_file_id}",
+                    session_id=session_id,
+                )
+
+            return WorkspaceDeleteResponse(
+                file_id=target_file_id,
+                success=True,
+                message="File deleted successfully",
+                session_id=session_id,
+            )
+
+        except Exception as e:
+            logger.error(f"Error deleting workspace file: {e}", exc_info=True)
+            return ErrorResponse(
+                message=f"Failed to delete workspace file: {str(e)}",
+                error=str(e),
+                session_id=session_id,
+            )

autogpt_platform/backend/backend/api/features/integrations/router.py

@@ -1,7 +1,7 @@
 import asyncio
 import logging
 from datetime import datetime, timedelta, timezone
-from typing import TYPE_CHECKING, Annotated, Any, List, Literal
+from typing import TYPE_CHECKING, Annotated, List, Literal
 
 from autogpt_libs.auth import get_user_id
 from fastapi import (
@@ -14,7 +14,7 @@ from fastapi import (
     Security,
     status,
 )
-from pydantic import BaseModel, Field, SecretStr, model_validator
+from pydantic import BaseModel, Field, SecretStr
 from starlette.status import HTTP_500_INTERNAL_SERVER_ERROR, HTTP_502_BAD_GATEWAY
 
 from backend.api.features.library.db import set_preset_webhook, update_preset
@@ -39,11 +39,7 @@ from backend.data.onboarding import OnboardingStep, complete_onboarding_step
 from backend.data.user import get_user_integrations
 from backend.executor.utils import add_graph_execution
 from backend.integrations.ayrshare import AyrshareClient, SocialPlatform
-from backend.integrations.credentials_store import provider_matches
-from backend.integrations.creds_manager import (
-    IntegrationCredentialsManager,
-    create_mcp_oauth_handler,
-)
+from backend.integrations.creds_manager import IntegrationCredentialsManager
 from backend.integrations.oauth import CREDENTIALS_BY_PROVIDER, HANDLERS_BY_NAME
 from backend.integrations.providers import ProviderName
 from backend.integrations.webhooks import get_webhook_manager
@@ -106,37 +102,9 @@ class CredentialsMetaResponse(BaseModel):
     scopes: list[str] | None
     username: str | None
     host: str | None = Field(
-        default=None,
-        description="Host pattern for host-scoped or MCP server URL for MCP credentials",
+        default=None, description="Host pattern for host-scoped credentials"
     )
 
-    @model_validator(mode="before")
-    @classmethod
-    def _normalize_provider(cls, data: Any) -> Any:
-        """Fix ``ProviderName.X`` format from Python 3.13 ``str(Enum)`` bug."""
-        if isinstance(data, dict):
-            prov = data.get("provider", "")
-            if isinstance(prov, str) and prov.startswith("ProviderName."):
-                member = prov.removeprefix("ProviderName.")
-                try:
-                    data = {**data, "provider": ProviderName[member].value}
-                except KeyError:
-                    pass
-        return data
-
-    @staticmethod
-    def get_host(cred: Credentials) -> str | None:
-        """Extract host from credential: HostScoped host or MCP server URL."""
-        if isinstance(cred, HostScopedCredentials):
-            return cred.host
-        if isinstance(cred, OAuth2Credentials) and cred.provider in (
-            ProviderName.MCP,
-            ProviderName.MCP.value,
-            "ProviderName.MCP",
-        ):
-            return (cred.metadata or {}).get("mcp_server_url")
-        return None
-
 
 @router.post("/{provider}/callback", summary="Exchange OAuth code for tokens")
 async def callback(
@@ -211,7 +179,9 @@ async def callback(
         title=credentials.title,
         scopes=credentials.scopes,
         username=credentials.username,
-        host=(CredentialsMetaResponse.get_host(credentials)),
+        host=(
+            credentials.host if isinstance(credentials, HostScopedCredentials) else None
+        ),
     )
 
 
@@ -229,7 +199,7 @@ async def list_credentials(
             title=cred.title,
             scopes=cred.scopes if isinstance(cred, OAuth2Credentials) else None,
             username=cred.username if isinstance(cred, OAuth2Credentials) else None,
-            host=CredentialsMetaResponse.get_host(cred),
+            host=cred.host if isinstance(cred, HostScopedCredentials) else None,
         )
         for cred in credentials
     ]
@@ -252,7 +222,7 @@ async def list_credentials_by_provider(
             title=cred.title,
             scopes=cred.scopes if isinstance(cred, OAuth2Credentials) else None,
             username=cred.username if isinstance(cred, OAuth2Credentials) else None,
-            host=CredentialsMetaResponse.get_host(cred),
+            host=cred.host if isinstance(cred, HostScopedCredentials) else None,
         )
         for cred in credentials
     ]
@@ -352,11 +322,7 @@ async def delete_credentials(
 
     tokens_revoked = None
     if isinstance(creds, OAuth2Credentials):
-        if provider_matches(provider.value, ProviderName.MCP.value):
-            # MCP uses dynamic per-server OAuth — create handler from metadata
-            handler = create_mcp_oauth_handler(creds)
-        else:
-            handler = _get_provider_oauth_handler(request, provider)
+        handler = _get_provider_oauth_handler(request, provider)
         tokens_revoked = await handler.revoke_tokens(creds)
 
     return CredentialsDeletionResponse(revoked=tokens_revoked)

autogpt_platform/backend/backend/api/features/library/db.py

@@ -12,11 +12,12 @@ import backend.api.features.store.image_gen as store_image_gen
 import backend.api.features.store.media as store_media
 import backend.data.graph as graph_db
 import backend.data.integrations as integrations_db
+from backend.data.block import BlockInput
 from backend.data.db import transaction
 from backend.data.execution import get_graph_execution
 from backend.data.graph import GraphSettings
 from backend.data.includes import AGENT_PRESET_INCLUDE, library_agent_include
-from backend.data.model import CredentialsMetaInput, GraphInput
+from backend.data.model import CredentialsMetaInput
 from backend.integrations.creds_manager import IntegrationCredentialsManager
 from backend.integrations.webhooks.graph_lifecycle_hooks import (
     on_graph_activate,
@@ -1129,7 +1130,7 @@ async def create_preset_from_graph_execution(
 async def update_preset(
     user_id: str,
     preset_id: str,
-    inputs: Optional[GraphInput] = None,
+    inputs: Optional[BlockInput] = None,
     credentials: Optional[dict[str, CredentialsMetaInput]] = None,
     name: Optional[str] = None,
     description: Optional[str] = None,

autogpt_platform/backend/backend/api/features/library/model.py

@@ -6,12 +6,9 @@ import prisma.enums
 import prisma.models
 import pydantic
 
+from backend.data.block import BlockInput
 from backend.data.graph import GraphModel, GraphSettings, GraphTriggerInfo
-from backend.data.model import (
-    CredentialsMetaInput,
-    GraphInput,
-    is_credentials_field_name,
-)
+from backend.data.model import CredentialsMetaInput, is_credentials_field_name
 from backend.util.json import loads as json_loads
 from backend.util.models import Pagination
 
@@ -326,7 +323,7 @@ class LibraryAgentPresetCreatable(pydantic.BaseModel):
     graph_id: str
     graph_version: int
 
-    inputs: GraphInput
+    inputs: BlockInput
     credentials: dict[str, CredentialsMetaInput]
 
     name: str
@@ -355,7 +352,7 @@ class LibraryAgentPresetUpdatable(pydantic.BaseModel):
     Request model used when updating a preset for a library agent.
     """
 
-    inputs: Optional[GraphInput] = None
+    inputs: Optional[BlockInput] = None
     credentials: Optional[dict[str, CredentialsMetaInput]] = None
 
     name: Optional[str] = None
@@ -398,7 +395,7 @@ class LibraryAgentPreset(LibraryAgentPresetCreatable):
                 "Webhook must be included in AgentPreset query when webhookId is set"
             )
 
-        input_data: GraphInput = {}
+        input_data: BlockInput = {}
         input_credentials: dict[str, CredentialsMetaInput] = {}
 
         for preset_input in preset.InputPresets:

autogpt_platform/backend/backend/api/features/mcp/routes.py

@@ -1,404 +0,0 @@
-"""
-MCP (Model Context Protocol) API routes.
-
-Provides endpoints for MCP tool discovery and OAuth authentication so the
-frontend can list available tools on an MCP server before placing a block.
-"""
-
-import logging
-from typing import Annotated, Any
-from urllib.parse import urlparse
-
-import fastapi
-from autogpt_libs.auth import get_user_id
-from fastapi import Security
-from pydantic import BaseModel, Field
-
-from backend.api.features.integrations.router import CredentialsMetaResponse
-from backend.blocks.mcp.client import MCPClient, MCPClientError
-from backend.blocks.mcp.oauth import MCPOAuthHandler
-from backend.data.model import OAuth2Credentials
-from backend.integrations.creds_manager import IntegrationCredentialsManager
-from backend.integrations.providers import ProviderName
-from backend.util.request import HTTPClientError, Requests
-from backend.util.settings import Settings
-
-logger = logging.getLogger(__name__)
-
-settings = Settings()
-router = fastapi.APIRouter(tags=["mcp"])
-creds_manager = IntegrationCredentialsManager()
-
-
-# ====================== Tool Discovery ====================== #
-
-
-class DiscoverToolsRequest(BaseModel):
-    """Request to discover tools on an MCP server."""
-
-    server_url: str = Field(description="URL of the MCP server")
-    auth_token: str | None = Field(
-        default=None,
-        description="Optional Bearer token for authenticated MCP servers",
-    )
-
-
-class MCPToolResponse(BaseModel):
-    """A single MCP tool returned by discovery."""
-
-    name: str
-    description: str
-    input_schema: dict[str, Any]
-
-
-class DiscoverToolsResponse(BaseModel):
-    """Response containing the list of tools available on an MCP server."""
-
-    tools: list[MCPToolResponse]
-    server_name: str | None = None
-    protocol_version: str | None = None
-
-
-@router.post(
-    "/discover-tools",
-    summary="Discover available tools on an MCP server",
-    response_model=DiscoverToolsResponse,
-)
-async def discover_tools(
-    request: DiscoverToolsRequest,
-    user_id: Annotated[str, Security(get_user_id)],
-) -> DiscoverToolsResponse:
-    """
-    Connect to an MCP server and return its available tools.
-
-    If the user has a stored MCP credential for this server URL, it will be
-    used automatically — no need to pass an explicit auth token.
-    """
-    auth_token = request.auth_token
-
-    # Auto-use stored MCP credential when no explicit token is provided.
-    if not auth_token:
-        mcp_creds = await creds_manager.store.get_creds_by_provider(
-            user_id, ProviderName.MCP.value
-        )
-        # Find the freshest credential for this server URL
-        best_cred: OAuth2Credentials | None = None
-        for cred in mcp_creds:
-            if (
-                isinstance(cred, OAuth2Credentials)
-                and (cred.metadata or {}).get("mcp_server_url") == request.server_url
-            ):
-                if best_cred is None or (
-                    (cred.access_token_expires_at or 0)
-                    > (best_cred.access_token_expires_at or 0)
-                ):
-                    best_cred = cred
-        if best_cred:
-            # Refresh the token if expired before using it
-            best_cred = await creds_manager.refresh_if_needed(user_id, best_cred)
-            logger.info(
-                f"Using MCP credential {best_cred.id} for {request.server_url}, "
-                f"expires_at={best_cred.access_token_expires_at}"
-            )
-            auth_token = best_cred.access_token.get_secret_value()
-
-    client = MCPClient(request.server_url, auth_token=auth_token)
-
-    try:
-        init_result = await client.initialize()
-        tools = await client.list_tools()
-    except HTTPClientError as e:
-        if e.status_code in (401, 403):
-            raise fastapi.HTTPException(
-                status_code=401,
-                detail="This MCP server requires authentication. "
-                "Please provide a valid auth token.",
-            )
-        raise fastapi.HTTPException(status_code=502, detail=str(e))
-    except MCPClientError as e:
-        raise fastapi.HTTPException(status_code=502, detail=str(e))
-    except Exception as e:
-        raise fastapi.HTTPException(
-            status_code=502,
-            detail=f"Failed to connect to MCP server: {e}",
-        )
-
-    return DiscoverToolsResponse(
-        tools=[
-            MCPToolResponse(
-                name=t.name,
-                description=t.description,
-                input_schema=t.input_schema,
-            )
-            for t in tools
-        ],
-        server_name=(
-            init_result.get("serverInfo", {}).get("name")
-            or urlparse(request.server_url).hostname
-            or "MCP"
-        ),
-        protocol_version=init_result.get("protocolVersion"),
-    )
-
-
-# ======================== OAuth Flow ======================== #
-
-
-class MCPOAuthLoginRequest(BaseModel):
-    """Request to start an OAuth flow for an MCP server."""
-
-    server_url: str = Field(description="URL of the MCP server that requires OAuth")
-
-
-class MCPOAuthLoginResponse(BaseModel):
-    """Response with the OAuth login URL for the user to authenticate."""
-
-    login_url: str
-    state_token: str
-
-
-@router.post(
-    "/oauth/login",
-    summary="Initiate OAuth login for an MCP server",
-)
-async def mcp_oauth_login(
-    request: MCPOAuthLoginRequest,
-    user_id: Annotated[str, Security(get_user_id)],
-) -> MCPOAuthLoginResponse:
-    """
-    Discover OAuth metadata from the MCP server and return a login URL.
-
-    1. Discovers the protected-resource metadata (RFC 9728)
-    2. Fetches the authorization server metadata (RFC 8414)
-    3. Performs Dynamic Client Registration (RFC 7591) if available
-    4. Returns the authorization URL for the frontend to open in a popup
-    """
-    client = MCPClient(request.server_url)
-
-    # Step 1: Discover protected-resource metadata (RFC 9728)
-    protected_resource = await client.discover_auth()
-
-    metadata: dict[str, Any] | None = None
-
-    if protected_resource and protected_resource.get("authorization_servers"):
-        auth_server_url = protected_resource["authorization_servers"][0]
-        resource_url = protected_resource.get("resource", request.server_url)
-
-        # Step 2a: Discover auth-server metadata (RFC 8414)
-        metadata = await client.discover_auth_server_metadata(auth_server_url)
-    else:
-        # Fallback: Some MCP servers (e.g. Linear) are their own auth server
-        # and serve OAuth metadata directly without protected-resource metadata.
-        # Don't assume a resource_url — omitting it lets the auth server choose
-        # the correct audience for the token (RFC 8707 resource is optional).
-        resource_url = None
-        metadata = await client.discover_auth_server_metadata(request.server_url)
-
-    if (
-        not metadata
-        or "authorization_endpoint" not in metadata
-        or "token_endpoint" not in metadata
-    ):
-        raise fastapi.HTTPException(
-            status_code=400,
-            detail="This MCP server does not advertise OAuth support. "
-            "You may need to provide an auth token manually.",
-        )
-
-    authorize_url = metadata["authorization_endpoint"]
-    token_url = metadata["token_endpoint"]
-    registration_endpoint = metadata.get("registration_endpoint")
-    revoke_url = metadata.get("revocation_endpoint")
-
-    # Step 3: Dynamic Client Registration (RFC 7591) if available
-    frontend_base_url = settings.config.frontend_base_url
-    if not frontend_base_url:
-        raise fastapi.HTTPException(
-            status_code=500,
-            detail="Frontend base URL is not configured.",
-        )
-    redirect_uri = f"{frontend_base_url}/auth/integrations/mcp_callback"
-
-    client_id = ""
-    client_secret = ""
-    if registration_endpoint:
-        reg_result = await _register_mcp_client(
-            registration_endpoint, redirect_uri, request.server_url
-        )
-        if reg_result:
-            client_id = reg_result.get("client_id", "")
-            client_secret = reg_result.get("client_secret", "")
-
-    if not client_id:
-        client_id = "autogpt-platform"
-
-    # Step 4: Store state token with OAuth metadata for the callback
-    scopes = (protected_resource or {}).get("scopes_supported") or metadata.get(
-        "scopes_supported", []
-    )
-    state_token, code_challenge = await creds_manager.store.store_state_token(
-        user_id,
-        ProviderName.MCP.value,
-        scopes,
-        state_metadata={
-            "authorize_url": authorize_url,
-            "token_url": token_url,
-            "revoke_url": revoke_url,
-            "resource_url": resource_url,
-            "server_url": request.server_url,
-            "client_id": client_id,
-            "client_secret": client_secret,
-        },
-    )
-
-    # Step 5: Build and return the login URL
-    handler = MCPOAuthHandler(
-        client_id=client_id,
-        client_secret=client_secret,
-        redirect_uri=redirect_uri,
-        authorize_url=authorize_url,
-        token_url=token_url,
-        resource_url=resource_url,
-    )
-    login_url = handler.get_login_url(
-        scopes, state_token, code_challenge=code_challenge
-    )
-
-    return MCPOAuthLoginResponse(login_url=login_url, state_token=state_token)
-
-
-class MCPOAuthCallbackRequest(BaseModel):
-    """Request to exchange an OAuth code for tokens."""
-
-    code: str = Field(description="Authorization code from OAuth callback")
-    state_token: str = Field(description="State token for CSRF verification")
-
-
-class MCPOAuthCallbackResponse(BaseModel):
-    """Response after successfully storing OAuth credentials."""
-
-    credential_id: str
-
-
-@router.post(
-    "/oauth/callback",
-    summary="Exchange OAuth code for MCP tokens",
-)
-async def mcp_oauth_callback(
-    request: MCPOAuthCallbackRequest,
-    user_id: Annotated[str, Security(get_user_id)],
-) -> CredentialsMetaResponse:
-    """
-    Exchange the authorization code for tokens and store the credential.
-
-    The frontend calls this after receiving the OAuth code from the popup.
-    On success, subsequent ``/discover-tools`` calls for the same server URL
-    will automatically use the stored credential.
-    """
-    valid_state = await creds_manager.store.verify_state_token(
-        user_id, request.state_token, ProviderName.MCP.value
-    )
-    if not valid_state:
-        raise fastapi.HTTPException(
-            status_code=400,
-            detail="Invalid or expired state token.",
-        )
-
-    meta = valid_state.state_metadata
-    frontend_base_url = settings.config.frontend_base_url
-    if not frontend_base_url:
-        raise fastapi.HTTPException(
-            status_code=500,
-            detail="Frontend base URL is not configured.",
-        )
-    redirect_uri = f"{frontend_base_url}/auth/integrations/mcp_callback"
-
-    handler = MCPOAuthHandler(
-        client_id=meta["client_id"],
-        client_secret=meta.get("client_secret", ""),
-        redirect_uri=redirect_uri,
-        authorize_url=meta["authorize_url"],
-        token_url=meta["token_url"],
-        revoke_url=meta.get("revoke_url"),
-        resource_url=meta.get("resource_url"),
-    )
-
-    try:
-        credentials = await handler.exchange_code_for_tokens(
-            request.code, valid_state.scopes, valid_state.code_verifier
-        )
-    except Exception as e:
-        raise fastapi.HTTPException(
-            status_code=400,
-            detail=f"OAuth token exchange failed: {e}",
-        )
-
-    # Enrich credential metadata for future lookup and token refresh
-    if credentials.metadata is None:
-        credentials.metadata = {}
-    credentials.metadata["mcp_server_url"] = meta["server_url"]
-    credentials.metadata["mcp_client_id"] = meta["client_id"]
-    credentials.metadata["mcp_client_secret"] = meta.get("client_secret", "")
-    credentials.metadata["mcp_token_url"] = meta["token_url"]
-    credentials.metadata["mcp_resource_url"] = meta.get("resource_url", "")
-
-    hostname = urlparse(meta["server_url"]).hostname or meta["server_url"]
-    credentials.title = f"MCP: {hostname}"
-
-    # Remove old MCP credentials for the same server to prevent stale token buildup.
-    try:
-        old_creds = await creds_manager.store.get_creds_by_provider(
-            user_id, ProviderName.MCP.value
-        )
-        for old in old_creds:
-            if (
-                isinstance(old, OAuth2Credentials)
-                and (old.metadata or {}).get("mcp_server_url") == meta["server_url"]
-            ):
-                await creds_manager.store.delete_creds_by_id(user_id, old.id)
-                logger.info(
-                    f"Removed old MCP credential {old.id} for {meta['server_url']}"
-                )
-    except Exception:
-        logger.debug("Could not clean up old MCP credentials", exc_info=True)
-
-    await creds_manager.create(user_id, credentials)
-
-    return CredentialsMetaResponse(
-        id=credentials.id,
-        provider=credentials.provider,
-        type=credentials.type,
-        title=credentials.title,
-        scopes=credentials.scopes,
-        username=credentials.username,
-        host=credentials.metadata.get("mcp_server_url"),
-    )
-
-
-# ======================== Helpers ======================== #
-
-
-async def _register_mcp_client(
-    registration_endpoint: str,
-    redirect_uri: str,
-    server_url: str,
-) -> dict[str, Any] | None:
-    """Attempt Dynamic Client Registration (RFC 7591) with an MCP auth server."""
-    try:
-        response = await Requests(raise_for_status=True).post(
-            registration_endpoint,
-            json={
-                "client_name": "AutoGPT Platform",
-                "redirect_uris": [redirect_uri],
-                "grant_types": ["authorization_code"],
-                "response_types": ["code"],
-                "token_endpoint_auth_method": "client_secret_post",
-            },
-        )
-        data = response.json()
-        if isinstance(data, dict) and "client_id" in data:
-            return data
-        return None
-    except Exception as e:
-        logger.warning(f"Dynamic client registration failed for {server_url}: {e}")
-        return None

autogpt_platform/backend/backend/api/features/mcp/test_routes.py

@@ -1,436 +0,0 @@
-"""Tests for MCP API routes.
-
-Uses httpx.AsyncClient with ASGITransport instead of fastapi.testclient.TestClient
-to avoid creating blocking portals that can corrupt pytest-asyncio's session event loop.
-"""
-
-from unittest.mock import AsyncMock, patch
-
-import fastapi
-import httpx
-import pytest
-import pytest_asyncio
-from autogpt_libs.auth import get_user_id
-
-from backend.api.features.mcp.routes import router
-from backend.blocks.mcp.client import MCPClientError, MCPTool
-from backend.util.request import HTTPClientError
-
-app = fastapi.FastAPI()
-app.include_router(router)
-app.dependency_overrides[get_user_id] = lambda: "test-user-id"
-
-
-@pytest_asyncio.fixture(scope="module")
-async def client():
-    transport = httpx.ASGITransport(app=app)
-    async with httpx.AsyncClient(transport=transport, base_url="http://test") as c:
-        yield c
-
-
-class TestDiscoverTools:
-    @pytest.mark.asyncio(loop_scope="session")
-    async def test_discover_tools_success(self, client):
-        mock_tools = [
-            MCPTool(
-                name="get_weather",
-                description="Get weather for a city",
-                input_schema={
-                    "type": "object",
-                    "properties": {"city": {"type": "string"}},
-                    "required": ["city"],
-                },
-            ),
-            MCPTool(
-                name="add_numbers",
-                description="Add two numbers",
-                input_schema={
-                    "type": "object",
-                    "properties": {
-                        "a": {"type": "number"},
-                        "b": {"type": "number"},
-                    },
-                },
-            ),
-        ]
-
-        with (
-            patch("backend.api.features.mcp.routes.MCPClient") as MockClient,
-            patch("backend.api.features.mcp.routes.creds_manager") as mock_cm,
-        ):
-            mock_cm.store.get_creds_by_provider = AsyncMock(return_value=[])
-            instance = MockClient.return_value
-            instance.initialize = AsyncMock(
-                return_value={
-                    "protocolVersion": "2025-03-26",
-                    "serverInfo": {"name": "test-server"},
-                }
-            )
-            instance.list_tools = AsyncMock(return_value=mock_tools)
-
-            response = await client.post(
-                "/discover-tools",
-                json={"server_url": "https://mcp.example.com/mcp"},
-            )
-
-        assert response.status_code == 200
-        data = response.json()
-        assert len(data["tools"]) == 2
-        assert data["tools"][0]["name"] == "get_weather"
-        assert data["tools"][1]["name"] == "add_numbers"
-        assert data["server_name"] == "test-server"
-        assert data["protocol_version"] == "2025-03-26"
-
-    @pytest.mark.asyncio(loop_scope="session")
-    async def test_discover_tools_with_auth_token(self, client):
-        with patch("backend.api.features.mcp.routes.MCPClient") as MockClient:
-            instance = MockClient.return_value
-            instance.initialize = AsyncMock(
-                return_value={"serverInfo": {}, "protocolVersion": "2025-03-26"}
-            )
-            instance.list_tools = AsyncMock(return_value=[])
-
-            response = await client.post(
-                "/discover-tools",
-                json={
-                    "server_url": "https://mcp.example.com/mcp",
-                    "auth_token": "my-secret-token",
-                },
-            )
-
-        assert response.status_code == 200
-        MockClient.assert_called_once_with(
-            "https://mcp.example.com/mcp",
-            auth_token="my-secret-token",
-        )
-
-    @pytest.mark.asyncio(loop_scope="session")
-    async def test_discover_tools_auto_uses_stored_credential(self, client):
-        """When no explicit token is given, stored MCP credentials are used."""
-        from pydantic import SecretStr
-
-        from backend.data.model import OAuth2Credentials
-
-        stored_cred = OAuth2Credentials(
-            provider="mcp",
-            title="MCP: example.com",
-            access_token=SecretStr("stored-token-123"),
-            refresh_token=None,
-            access_token_expires_at=None,
-            refresh_token_expires_at=None,
-            scopes=[],
-            metadata={"mcp_server_url": "https://mcp.example.com/mcp"},
-        )
-
-        with (
-            patch("backend.api.features.mcp.routes.MCPClient") as MockClient,
-            patch("backend.api.features.mcp.routes.creds_manager") as mock_cm,
-        ):
-            mock_cm.store.get_creds_by_provider = AsyncMock(return_value=[stored_cred])
-            mock_cm.refresh_if_needed = AsyncMock(return_value=stored_cred)
-            instance = MockClient.return_value
-            instance.initialize = AsyncMock(
-                return_value={"serverInfo": {}, "protocolVersion": "2025-03-26"}
-            )
-            instance.list_tools = AsyncMock(return_value=[])
-
-            response = await client.post(
-                "/discover-tools",
-                json={"server_url": "https://mcp.example.com/mcp"},
-            )
-
-        assert response.status_code == 200
-        MockClient.assert_called_once_with(
-            "https://mcp.example.com/mcp",
-            auth_token="stored-token-123",
-        )
-
-    @pytest.mark.asyncio(loop_scope="session")
-    async def test_discover_tools_mcp_error(self, client):
-        with (
-            patch("backend.api.features.mcp.routes.MCPClient") as MockClient,
-            patch("backend.api.features.mcp.routes.creds_manager") as mock_cm,
-        ):
-            mock_cm.store.get_creds_by_provider = AsyncMock(return_value=[])
-            instance = MockClient.return_value
-            instance.initialize = AsyncMock(
-                side_effect=MCPClientError("Connection refused")
-            )
-
-            response = await client.post(
-                "/discover-tools",
-                json={"server_url": "https://bad-server.example.com/mcp"},
-            )
-
-        assert response.status_code == 502
-        assert "Connection refused" in response.json()["detail"]
-
-    @pytest.mark.asyncio(loop_scope="session")
-    async def test_discover_tools_generic_error(self, client):
-        with (
-            patch("backend.api.features.mcp.routes.MCPClient") as MockClient,
-            patch("backend.api.features.mcp.routes.creds_manager") as mock_cm,
-        ):
-            mock_cm.store.get_creds_by_provider = AsyncMock(return_value=[])
-            instance = MockClient.return_value
-            instance.initialize = AsyncMock(side_effect=Exception("Network timeout"))
-
-            response = await client.post(
-                "/discover-tools",
-                json={"server_url": "https://timeout.example.com/mcp"},
-            )
-
-        assert response.status_code == 502
-        assert "Failed to connect" in response.json()["detail"]
-
-    @pytest.mark.asyncio(loop_scope="session")
-    async def test_discover_tools_auth_required(self, client):
-        with (
-            patch("backend.api.features.mcp.routes.MCPClient") as MockClient,
-            patch("backend.api.features.mcp.routes.creds_manager") as mock_cm,
-        ):
-            mock_cm.store.get_creds_by_provider = AsyncMock(return_value=[])
-            instance = MockClient.return_value
-            instance.initialize = AsyncMock(
-                side_effect=HTTPClientError("HTTP 401 Error: Unauthorized", 401)
-            )
-
-            response = await client.post(
-                "/discover-tools",
-                json={"server_url": "https://auth-server.example.com/mcp"},
-            )
-
-        assert response.status_code == 401
-        assert "requires authentication" in response.json()["detail"]
-
-    @pytest.mark.asyncio(loop_scope="session")
-    async def test_discover_tools_forbidden(self, client):
-        with (
-            patch("backend.api.features.mcp.routes.MCPClient") as MockClient,
-            patch("backend.api.features.mcp.routes.creds_manager") as mock_cm,
-        ):
-            mock_cm.store.get_creds_by_provider = AsyncMock(return_value=[])
-            instance = MockClient.return_value
-            instance.initialize = AsyncMock(
-                side_effect=HTTPClientError("HTTP 403 Error: Forbidden", 403)
-            )
-
-            response = await client.post(
-                "/discover-tools",
-                json={"server_url": "https://auth-server.example.com/mcp"},
-            )
-
-        assert response.status_code == 401
-        assert "requires authentication" in response.json()["detail"]
-
-    @pytest.mark.asyncio(loop_scope="session")
-    async def test_discover_tools_missing_url(self, client):
-        response = await client.post("/discover-tools", json={})
-        assert response.status_code == 422
-
-
-class TestOAuthLogin:
-    @pytest.mark.asyncio(loop_scope="session")
-    async def test_oauth_login_success(self, client):
-        with (
-            patch("backend.api.features.mcp.routes.MCPClient") as MockClient,
-            patch("backend.api.features.mcp.routes.creds_manager") as mock_cm,
-            patch("backend.api.features.mcp.routes.settings") as mock_settings,
-            patch(
-                "backend.api.features.mcp.routes._register_mcp_client"
-            ) as mock_register,
-        ):
-            instance = MockClient.return_value
-            instance.discover_auth = AsyncMock(
-                return_value={
-                    "authorization_servers": ["https://auth.sentry.io"],
-                    "resource": "https://mcp.sentry.dev/mcp",
-                    "scopes_supported": ["openid"],
-                }
-            )
-            instance.discover_auth_server_metadata = AsyncMock(
-                return_value={
-                    "authorization_endpoint": "https://auth.sentry.io/authorize",
-                    "token_endpoint": "https://auth.sentry.io/token",
-                    "registration_endpoint": "https://auth.sentry.io/register",
-                }
-            )
-            mock_register.return_value = {
-                "client_id": "registered-client-id",
-                "client_secret": "registered-secret",
-            }
-            mock_cm.store.store_state_token = AsyncMock(
-                return_value=("state-token-123", "code-challenge-abc")
-            )
-            mock_settings.config.frontend_base_url = "http://localhost:3000"
-
-            response = await client.post(
-                "/oauth/login",
-                json={"server_url": "https://mcp.sentry.dev/mcp"},
-            )
-
-        assert response.status_code == 200
-        data = response.json()
-        assert "login_url" in data
-        assert data["state_token"] == "state-token-123"
-        assert "auth.sentry.io/authorize" in data["login_url"]
-        assert "registered-client-id" in data["login_url"]
-
-    @pytest.mark.asyncio(loop_scope="session")
-    async def test_oauth_login_no_oauth_support(self, client):
-        with patch("backend.api.features.mcp.routes.MCPClient") as MockClient:
-            instance = MockClient.return_value
-            instance.discover_auth = AsyncMock(return_value=None)
-            instance.discover_auth_server_metadata = AsyncMock(return_value=None)
-
-            response = await client.post(
-                "/oauth/login",
-                json={"server_url": "https://simple-server.example.com/mcp"},
-            )
-
-        assert response.status_code == 400
-        assert "does not advertise OAuth" in response.json()["detail"]
-
-    @pytest.mark.asyncio(loop_scope="session")
-    async def test_oauth_login_fallback_to_public_client(self, client):
-        """When DCR is unavailable, falls back to default public client ID."""
-        with (
-            patch("backend.api.features.mcp.routes.MCPClient") as MockClient,
-            patch("backend.api.features.mcp.routes.creds_manager") as mock_cm,
-            patch("backend.api.features.mcp.routes.settings") as mock_settings,
-        ):
-            instance = MockClient.return_value
-            instance.discover_auth = AsyncMock(
-                return_value={
-                    "authorization_servers": ["https://auth.example.com"],
-                    "resource": "https://mcp.example.com/mcp",
-                }
-            )
-            instance.discover_auth_server_metadata = AsyncMock(
-                return_value={
-                    "authorization_endpoint": "https://auth.example.com/authorize",
-                    "token_endpoint": "https://auth.example.com/token",
-                    # No registration_endpoint
-                }
-            )
-            mock_cm.store.store_state_token = AsyncMock(
-                return_value=("state-abc", "challenge-xyz")
-            )
-            mock_settings.config.frontend_base_url = "http://localhost:3000"
-
-            response = await client.post(
-                "/oauth/login",
-                json={"server_url": "https://mcp.example.com/mcp"},
-            )
-
-        assert response.status_code == 200
-        data = response.json()
-        assert "autogpt-platform" in data["login_url"]
-
-
-class TestOAuthCallback:
-    @pytest.mark.asyncio(loop_scope="session")
-    async def test_oauth_callback_success(self, client):
-        from pydantic import SecretStr
-
-        from backend.data.model import OAuth2Credentials
-
-        mock_creds = OAuth2Credentials(
-            provider="mcp",
-            title=None,
-            access_token=SecretStr("access-token-xyz"),
-            refresh_token=None,
-            access_token_expires_at=None,
-            refresh_token_expires_at=None,
-            scopes=[],
-            metadata={
-                "mcp_token_url": "https://auth.sentry.io/token",
-                "mcp_resource_url": "https://mcp.sentry.dev/mcp",
-            },
-        )
-
-        with (
-            patch("backend.api.features.mcp.routes.creds_manager") as mock_cm,
-            patch("backend.api.features.mcp.routes.settings") as mock_settings,
-            patch("backend.api.features.mcp.routes.MCPOAuthHandler") as MockHandler,
-        ):
-            mock_settings.config.frontend_base_url = "http://localhost:3000"
-
-            # Mock state verification
-            mock_state = AsyncMock()
-            mock_state.state_metadata = {
-                "authorize_url": "https://auth.sentry.io/authorize",
-                "token_url": "https://auth.sentry.io/token",
-                "client_id": "test-client-id",
-                "client_secret": "test-secret",
-                "server_url": "https://mcp.sentry.dev/mcp",
-            }
-            mock_state.scopes = ["openid"]
-            mock_state.code_verifier = "verifier-123"
-            mock_cm.store.verify_state_token = AsyncMock(return_value=mock_state)
-            mock_cm.create = AsyncMock()
-
-            handler_instance = MockHandler.return_value
-            handler_instance.exchange_code_for_tokens = AsyncMock(
-                return_value=mock_creds
-            )
-
-            # Mock old credential cleanup
-            mock_cm.store.get_creds_by_provider = AsyncMock(return_value=[])
-
-            response = await client.post(
-                "/oauth/callback",
-                json={"code": "auth-code-abc", "state_token": "state-token-123"},
-            )
-
-        assert response.status_code == 200
-        data = response.json()
-        assert "id" in data
-        assert data["provider"] == "mcp"
-        assert data["type"] == "oauth2"
-        mock_cm.create.assert_called_once()
-
-    @pytest.mark.asyncio(loop_scope="session")
-    async def test_oauth_callback_invalid_state(self, client):
-        with patch("backend.api.features.mcp.routes.creds_manager") as mock_cm:
-            mock_cm.store.verify_state_token = AsyncMock(return_value=None)
-
-            response = await client.post(
-                "/oauth/callback",
-                json={"code": "auth-code", "state_token": "bad-state"},
-            )
-
-        assert response.status_code == 400
-        assert "Invalid or expired" in response.json()["detail"]
-
-    @pytest.mark.asyncio(loop_scope="session")
-    async def test_oauth_callback_token_exchange_fails(self, client):
-        with (
-            patch("backend.api.features.mcp.routes.creds_manager") as mock_cm,
-            patch("backend.api.features.mcp.routes.settings") as mock_settings,
-            patch("backend.api.features.mcp.routes.MCPOAuthHandler") as MockHandler,
-        ):
-            mock_settings.config.frontend_base_url = "http://localhost:3000"
-            mock_state = AsyncMock()
-            mock_state.state_metadata = {
-                "authorize_url": "https://auth.example.com/authorize",
-                "token_url": "https://auth.example.com/token",
-                "client_id": "cid",
-                "server_url": "https://mcp.example.com/mcp",
-            }
-            mock_state.scopes = []
-            mock_state.code_verifier = "v"
-            mock_cm.store.verify_state_token = AsyncMock(return_value=mock_state)
-
-            handler_instance = MockHandler.return_value
-            handler_instance.exchange_code_for_tokens = AsyncMock(
-                side_effect=RuntimeError("Token exchange failed")
-            )
-
-            response = await client.post(
-                "/oauth/callback",
-                json={"code": "bad-code", "state_token": "state"},
-            )
-
-        assert response.status_code == 400
-        assert "token exchange failed" in response.json()["detail"].lower()

autogpt_platform/backend/backend/api/features/otto/service.py

@@ -5,8 +5,8 @@ from typing import Optional
 import aiohttp
 from fastapi import HTTPException
 
-from backend.blocks import get_block
 from backend.data import graph as graph_db
+from backend.data.block import get_block
 from backend.util.settings import Settings
 
 from .models import ApiResponse, ChatRequest, GraphData

autogpt_platform/backend/backend/api/features/store/content_handlers.py

@@ -152,7 +152,7 @@ class BlockHandler(ContentHandler):
 
     async def get_missing_items(self, batch_size: int) -> list[ContentItem]:
         """Fetch blocks without embeddings."""
-        from backend.blocks import get_blocks
+        from backend.data.block import get_blocks
 
         # Get all available blocks
         all_blocks = get_blocks()
@@ -249,7 +249,7 @@ class BlockHandler(ContentHandler):
 
     async def get_stats(self) -> dict[str, int]:
         """Get statistics about block embedding coverage."""
-        from backend.blocks import get_blocks
+        from backend.data.block import get_blocks
 
         all_blocks = get_blocks()
 

autogpt_platform/backend/backend/api/features/store/content_handlers_test.py

@@ -93,7 +93,7 @@ async def test_block_handler_get_missing_items(mocker):
     mock_existing = []
 
     with patch(
-        "backend.blocks.get_blocks",
+        "backend.data.block.get_blocks",
         return_value=mock_blocks,
     ):
         with patch(
@@ -135,7 +135,7 @@ async def test_block_handler_get_stats(mocker):
     mock_embedded = [{"count": 2}]
 
     with patch(
-        "backend.blocks.get_blocks",
+        "backend.data.block.get_blocks",
         return_value=mock_blocks,
     ):
         with patch(
@@ -327,7 +327,7 @@ async def test_block_handler_handles_missing_attributes():
     mock_blocks = {"block-minimal": mock_block_class}
 
     with patch(
-        "backend.blocks.get_blocks",
+        "backend.data.block.get_blocks",
         return_value=mock_blocks,
     ):
         with patch(
@@ -360,7 +360,7 @@ async def test_block_handler_skips_failed_blocks():
     mock_blocks = {"good-block": good_block, "bad-block": bad_block}
 
     with patch(
-        "backend.blocks.get_blocks",
+        "backend.data.block.get_blocks",
         return_value=mock_blocks,
     ):
         with patch(

autogpt_platform/backend/backend/api/features/store/embeddings.py

@@ -662,7 +662,7 @@ async def cleanup_orphaned_embeddings() -> dict[str, Any]:
                 )
                 current_ids = {row["id"] for row in valid_agents}
             elif content_type == ContentType.BLOCK:
-                from backend.blocks import get_blocks
+                from backend.data.block import get_blocks
 
                 current_ids = set(get_blocks().keys())
             elif content_type == ContentType.DOCUMENTATION:

autogpt_platform/backend/backend/api/features/store/image_gen.py

@@ -7,6 +7,15 @@ from replicate.client import Client as ReplicateClient
 from replicate.exceptions import ReplicateError
 from replicate.helpers import FileOutput
 
+from backend.blocks.ideogram import (
+    AspectRatio,
+    ColorPalettePreset,
+    IdeogramModelBlock,
+    IdeogramModelName,
+    MagicPromptOption,
+    StyleType,
+    UpscaleOption,
+)
 from backend.data.graph import GraphBaseMeta
 from backend.data.model import CredentialsMetaInput, ProviderName
 from backend.integrations.credentials_store import ideogram_credentials
@@ -41,16 +50,6 @@ async def generate_agent_image_v2(graph: GraphBaseMeta | AgentGraph) -> io.Bytes
     if not ideogram_credentials.api_key:
         raise ValueError("Missing Ideogram API key")
 
-    from backend.blocks.ideogram import (
-        AspectRatio,
-        ColorPalettePreset,
-        IdeogramModelBlock,
-        IdeogramModelName,
-        MagicPromptOption,
-        StyleType,
-        UpscaleOption,
-    )
-
     name = graph.name
     description = f"{name} ({graph.description})" if graph.description else name
 

autogpt_platform/backend/backend/api/features/v1.py

@@ -40,11 +40,10 @@ from backend.api.model import (
     UpdateTimezoneRequest,
     UploadFileResponse,
 )
-from backend.blocks import get_block, get_blocks
 from backend.data import execution as execution_db
 from backend.data import graph as graph_db
 from backend.data.auth import api_key as api_key_db
-from backend.data.block import BlockInput, CompletedBlockOutput
+from backend.data.block import BlockInput, CompletedBlockOutput, get_block, get_blocks
 from backend.data.credit import (
     AutoTopUpConfig,
     RefundRequest,

autogpt_platform/backend/backend/api/features/workspace/routes.py

@@ -11,7 +11,7 @@ import fastapi
 from autogpt_libs.auth.dependencies import get_user_id, requires_user
 from fastapi.responses import Response
 
-from backend.data.workspace import WorkspaceFile, get_workspace, get_workspace_file
+from backend.data.workspace import get_workspace, get_workspace_file
 from backend.util.workspace_storage import get_workspace_storage
 
 
@@ -44,11 +44,11 @@ router = fastapi.APIRouter(
 )
 
 
-def _create_streaming_response(content: bytes, file: WorkspaceFile) -> Response:
+def _create_streaming_response(content: bytes, file) -> Response:
     """Create a streaming response for file content."""
     return Response(
         content=content,
-        media_type=file.mime_type,
+        media_type=file.mimeType,
         headers={
             "Content-Disposition": _sanitize_filename_for_header(file.name),
             "Content-Length": str(len(content)),
@@ -56,7 +56,7 @@ def _create_streaming_response(content: bytes, file: WorkspaceFile) -> Response:
     )
 
 
-async def _create_file_download_response(file: WorkspaceFile) -> Response:
+async def _create_file_download_response(file) -> Response:
     """
     Create a download response for a workspace file.
 
@@ -66,33 +66,33 @@ async def _create_file_download_response(file: WorkspaceFile) -> Response:
     storage = await get_workspace_storage()
 
     # For local storage, stream the file directly
-    if file.storage_path.startswith("local://"):
-        content = await storage.retrieve(file.storage_path)
+    if file.storagePath.startswith("local://"):
+        content = await storage.retrieve(file.storagePath)
         return _create_streaming_response(content, file)
 
     # For GCS, try to redirect to signed URL, fall back to streaming
     try:
-        url = await storage.get_download_url(file.storage_path, expires_in=300)
+        url = await storage.get_download_url(file.storagePath, expires_in=300)
         # If we got back an API path (fallback), stream directly instead
         if url.startswith("/api/"):
-            content = await storage.retrieve(file.storage_path)
+            content = await storage.retrieve(file.storagePath)
             return _create_streaming_response(content, file)
         return fastapi.responses.RedirectResponse(url=url, status_code=302)
     except Exception as e:
         # Log the signed URL failure with context
         logger.error(
             f"Failed to get signed URL for file {file.id} "
-            f"(storagePath={file.storage_path}): {e}",
+            f"(storagePath={file.storagePath}): {e}",
             exc_info=True,
         )
         # Fall back to streaming directly from GCS
         try:
-            content = await storage.retrieve(file.storage_path)
+            content = await storage.retrieve(file.storagePath)
             return _create_streaming_response(content, file)
         except Exception as fallback_error:
             logger.error(
                 f"Fallback streaming also failed for file {file.id} "
-                f"(storagePath={file.storage_path}): {fallback_error}",
+                f"(storagePath={file.storagePath}): {fallback_error}",
                 exc_info=True,
             )
             raise

autogpt_platform/backend/backend/api/rest_api.py

@@ -26,7 +26,6 @@ import backend.api.features.executions.review.routes
 import backend.api.features.library.db
 import backend.api.features.library.model
 import backend.api.features.library.routes
-import backend.api.features.mcp.routes as mcp_routes
 import backend.api.features.oauth
 import backend.api.features.otto.routes
 import backend.api.features.postmark.postmark
@@ -41,6 +40,10 @@ import backend.data.user
 import backend.integrations.webhooks.utils
 import backend.util.service
 import backend.util.settings
+from backend.api.features.chat.completion_consumer import (
+    start_completion_consumer,
+    stop_completion_consumer,
+)
 from backend.blocks.llm import DEFAULT_LLM_MODEL
 from backend.data.model import Credentials
 from backend.integrations.providers import ProviderName
@@ -119,9 +122,21 @@ async def lifespan_context(app: fastapi.FastAPI):
     await backend.data.graph.migrate_llm_models(DEFAULT_LLM_MODEL)
     await backend.integrations.webhooks.utils.migrate_legacy_triggered_graphs()
 
+    # Start chat completion consumer for Redis Streams notifications
+    try:
+        await start_completion_consumer()
+    except Exception as e:
+        logger.warning(f"Could not start chat completion consumer: {e}")
+
     with launch_darkly_context():
         yield
 
+    # Stop chat completion consumer
+    try:
+        await stop_completion_consumer()
+    except Exception as e:
+        logger.warning(f"Error stopping chat completion consumer: {e}")
+
     try:
         await shutdown_cloud_storage_handler()
     except Exception as e:
@@ -328,11 +343,6 @@ app.include_router(
     tags=["workspace"],
     prefix="/api/workspace",
 )
-app.include_router(
-    mcp_routes.router,
-    tags=["v2", "mcp"],
-    prefix="/api/mcp",
-)
 app.include_router(
     backend.api.features.oauth.router,
     tags=["oauth"],

autogpt_platform/backend/backend/app.py

@@ -24,7 +24,7 @@ def run_processes(*processes: "AppProcess", **kwargs):
         # Run the last process in the foreground.
         processes[-1].start(background=False, **kwargs)
     finally:
-        for process in reversed(processes):
+        for process in processes:
             try:
                 process.stop()
             except Exception as e:
@@ -38,9 +38,7 @@ def main(**kwargs):
 
     from backend.api.rest_api import AgentServer
     from backend.api.ws_api import WebsocketServer
-    from backend.copilot.executor.manager import CoPilotExecutor
-    from backend.data.db_manager import DatabaseManager
-    from backend.executor import ExecutionManager, Scheduler
+    from backend.executor import DatabaseManager, ExecutionManager, Scheduler
     from backend.notifications import NotificationManager
 
     run_processes(
@@ -50,7 +48,6 @@ def main(**kwargs):
         WebsocketServer(),
         AgentServer(),
         ExecutionManager(),
-        CoPilotExecutor(),
         **kwargs,
     )
 

autogpt_platform/backend/backend/blocks/__init__.py

@@ -3,19 +3,22 @@ import logging
 import os
 import re
 from pathlib import Path
-from typing import Sequence, Type, TypeVar
+from typing import TYPE_CHECKING, TypeVar
 
-from backend.blocks._base import AnyBlockSchema, BlockType
 from backend.util.cache import cached
 
 logger = logging.getLogger(__name__)
 
+
+if TYPE_CHECKING:
+    from backend.data.block import Block
+
 T = TypeVar("T")
 
 
 @cached(ttl_seconds=3600)
-def load_all_blocks() -> dict[str, type["AnyBlockSchema"]]:
-    from backend.blocks._base import Block
+def load_all_blocks() -> dict[str, type["Block"]]:
+    from backend.data.block import Block
     from backend.util.settings import Config
 
     # Check if example blocks should be loaded from settings
@@ -47,8 +50,8 @@ def load_all_blocks() -> dict[str, type["AnyBlockSchema"]]:
         importlib.import_module(f".{module}", package=__name__)
 
     # Load all Block instances from the available modules
-    available_blocks: dict[str, type["AnyBlockSchema"]] = {}
-    for block_cls in _all_subclasses(Block):
+    available_blocks: dict[str, type["Block"]] = {}
+    for block_cls in all_subclasses(Block):
         class_name = block_cls.__name__
 
         if class_name.endswith("Base"):
@@ -61,7 +64,7 @@ def load_all_blocks() -> dict[str, type["AnyBlockSchema"]]:
                 "please name the class with 'Base' at the end"
             )
 
-        block = block_cls()  # pyright: ignore[reportAbstractUsage]
+        block = block_cls.create()
 
         if not isinstance(block.id, str) or len(block.id) != 36:
             raise ValueError(
@@ -102,7 +105,7 @@ def load_all_blocks() -> dict[str, type["AnyBlockSchema"]]:
         available_blocks[block.id] = block_cls
 
     # Filter out blocks with incomplete auth configs, e.g. missing OAuth server secrets
-    from ._utils import is_block_auth_configured
+    from backend.data.block import is_block_auth_configured
 
     filtered_blocks = {}
     for block_id, block_cls in available_blocks.items():
@@ -112,48 +115,11 @@ def load_all_blocks() -> dict[str, type["AnyBlockSchema"]]:
     return filtered_blocks
 
 
-def _all_subclasses(cls: type[T]) -> list[type[T]]:
+__all__ = ["load_all_blocks"]
+
+
+def all_subclasses(cls: type[T]) -> list[type[T]]:
     subclasses = cls.__subclasses__()
     for subclass in subclasses:
-        subclasses += _all_subclasses(subclass)
+        subclasses += all_subclasses(subclass)
     return subclasses
-
-
-# ============== Block access helper functions ============== #
-
-
-def get_blocks() -> dict[str, Type["AnyBlockSchema"]]:
-    return load_all_blocks()
-
-
-# Note on the return type annotation: https://github.com/microsoft/pyright/issues/10281
-def get_block(block_id: str) -> "AnyBlockSchema | None":
-    cls = get_blocks().get(block_id)
-    return cls() if cls else None
-
-
-@cached(ttl_seconds=3600)
-def get_webhook_block_ids() -> Sequence[str]:
-    return [
-        id
-        for id, B in get_blocks().items()
-        if B().block_type in (BlockType.WEBHOOK, BlockType.WEBHOOK_MANUAL)
-    ]
-
-
-@cached(ttl_seconds=3600)
-def get_io_block_ids() -> Sequence[str]:
-    return [
-        id
-        for id, B in get_blocks().items()
-        if B().block_type in (BlockType.INPUT, BlockType.OUTPUT)
-    ]
-
-
-@cached(ttl_seconds=3600)
-def get_human_in_the_loop_block_ids() -> Sequence[str]:
-    return [
-        id
-        for id, B in get_blocks().items()
-        if B().block_type == BlockType.HUMAN_IN_THE_LOOP
-    ]

autogpt_platform/backend/backend/blocks/_base.py

@@ -1,740 +0,0 @@
-import inspect
-import logging
-from abc import ABC, abstractmethod
-from enum import Enum
-from typing import (
-    TYPE_CHECKING,
-    Any,
-    Callable,
-    ClassVar,
-    Generic,
-    Optional,
-    Type,
-    TypeAlias,
-    TypeVar,
-    cast,
-    get_origin,
-)
-
-import jsonref
-import jsonschema
-from pydantic import BaseModel
-
-from backend.data.block import BlockInput, BlockOutput, BlockOutputEntry
-from backend.data.model import (
-    Credentials,
-    CredentialsFieldInfo,
-    CredentialsMetaInput,
-    SchemaField,
-    is_credentials_field_name,
-)
-from backend.integrations.providers import ProviderName
-from backend.util import json
-from backend.util.exceptions import (
-    BlockError,
-    BlockExecutionError,
-    BlockInputError,
-    BlockOutputError,
-    BlockUnknownError,
-)
-from backend.util.settings import Config
-
-logger = logging.getLogger(__name__)
-
-if TYPE_CHECKING:
-    from backend.data.execution import ExecutionContext
-    from backend.data.model import ContributorDetails, NodeExecutionStats
-
-    from ..data.graph import Link
-
-app_config = Config()
-
-
-BlockTestOutput = BlockOutputEntry | tuple[str, Callable[[Any], bool]]
-
-
-class BlockType(Enum):
-    STANDARD = "Standard"
-    INPUT = "Input"
-    OUTPUT = "Output"
-    NOTE = "Note"
-    WEBHOOK = "Webhook"
-    WEBHOOK_MANUAL = "Webhook (manual)"
-    AGENT = "Agent"
-    AI = "AI"
-    AYRSHARE = "Ayrshare"
-    HUMAN_IN_THE_LOOP = "Human In The Loop"
-    MCP_TOOL = "MCP Tool"
-
-
-class BlockCategory(Enum):
-    AI = "Block that leverages AI to perform a task."
-    SOCIAL = "Block that interacts with social media platforms."
-    TEXT = "Block that processes text data."
-    SEARCH = "Block that searches or extracts information from the internet."
-    BASIC = "Block that performs basic operations."
-    INPUT = "Block that interacts with input of the graph."
-    OUTPUT = "Block that interacts with output of the graph."
-    LOGIC = "Programming logic to control the flow of your agent"
-    COMMUNICATION = "Block that interacts with communication platforms."
-    DEVELOPER_TOOLS = "Developer tools such as GitHub blocks."
-    DATA = "Block that interacts with structured data."
-    HARDWARE = "Block that interacts with hardware."
-    AGENT = "Block that interacts with other agents."
-    CRM = "Block that interacts with CRM services."
-    SAFETY = (
-        "Block that provides AI safety mechanisms such as detecting harmful content"
-    )
-    PRODUCTIVITY = "Block that helps with productivity"
-    ISSUE_TRACKING = "Block that helps with issue tracking"
-    MULTIMEDIA = "Block that interacts with multimedia content"
-    MARKETING = "Block that helps with marketing"
-
-    def dict(self) -> dict[str, str]:
-        return {"category": self.name, "description": self.value}
-
-
-class BlockCostType(str, Enum):
-    RUN = "run"  # cost X credits per run
-    BYTE = "byte"  # cost X credits per byte
-    SECOND = "second"  # cost X credits per second
-
-
-class BlockCost(BaseModel):
-    cost_amount: int
-    cost_filter: BlockInput
-    cost_type: BlockCostType
-
-    def __init__(
-        self,
-        cost_amount: int,
-        cost_type: BlockCostType = BlockCostType.RUN,
-        cost_filter: Optional[BlockInput] = None,
-        **data: Any,
-    ) -> None:
-        super().__init__(
-            cost_amount=cost_amount,
-            cost_filter=cost_filter or {},
-            cost_type=cost_type,
-            **data,
-        )
-
-
-class BlockInfo(BaseModel):
-    id: str
-    name: str
-    inputSchema: dict[str, Any]
-    outputSchema: dict[str, Any]
-    costs: list[BlockCost]
-    description: str
-    categories: list[dict[str, str]]
-    contributors: list[dict[str, Any]]
-    staticOutput: bool
-    uiType: str
-
-
-class BlockSchema(BaseModel):
-    cached_jsonschema: ClassVar[dict[str, Any]]
-
-    @classmethod
-    def jsonschema(cls) -> dict[str, Any]:
-        if cls.cached_jsonschema:
-            return cls.cached_jsonschema
-
-        model = jsonref.replace_refs(cls.model_json_schema(), merge_props=True)
-
-        def ref_to_dict(obj):
-            if isinstance(obj, dict):
-                # OpenAPI <3.1 does not support sibling fields that has a $ref key
-                # So sometimes, the schema has an "allOf"/"anyOf"/"oneOf" with 1 item.
-                keys = {"allOf", "anyOf", "oneOf"}
-                one_key = next((k for k in keys if k in obj and len(obj[k]) == 1), None)
-                if one_key:
-                    obj.update(obj[one_key][0])
-
-                return {
-                    key: ref_to_dict(value)
-                    for key, value in obj.items()
-                    if not key.startswith("$") and key != one_key
-                }
-            elif isinstance(obj, list):
-                return [ref_to_dict(item) for item in obj]
-
-            return obj
-
-        cls.cached_jsonschema = cast(dict[str, Any], ref_to_dict(model))
-
-        return cls.cached_jsonschema
-
-    @classmethod
-    def validate_data(cls, data: BlockInput) -> str | None:
-        return json.validate_with_jsonschema(
-            schema=cls.jsonschema(),
-            data={k: v for k, v in data.items() if v is not None},
-        )
-
-    @classmethod
-    def get_mismatch_error(cls, data: BlockInput) -> str | None:
-        return cls.validate_data(data)
-
-    @classmethod
-    def get_field_schema(cls, field_name: str) -> dict[str, Any]:
-        model_schema = cls.jsonschema().get("properties", {})
-        if not model_schema:
-            raise ValueError(f"Invalid model schema {cls}")
-
-        property_schema = model_schema.get(field_name)
-        if not property_schema:
-            raise ValueError(f"Invalid property name {field_name}")
-
-        return property_schema
-
-    @classmethod
-    def validate_field(cls, field_name: str, data: BlockInput) -> str | None:
-        """
-        Validate the data against a specific property (one of the input/output name).
-        Returns the validation error message if the data does not match the schema.
-        """
-        try:
-            property_schema = cls.get_field_schema(field_name)
-            jsonschema.validate(json.to_dict(data), property_schema)
-            return None
-        except jsonschema.ValidationError as e:
-            return str(e)
-
-    @classmethod
-    def get_fields(cls) -> set[str]:
-        return set(cls.model_fields.keys())
-
-    @classmethod
-    def get_required_fields(cls) -> set[str]:
-        return {
-            field
-            for field, field_info in cls.model_fields.items()
-            if field_info.is_required()
-        }
-
-    @classmethod
-    def __pydantic_init_subclass__(cls, **kwargs):
-        """Validates the schema definition. Rules:
-        - Fields with annotation `CredentialsMetaInput` MUST be
-          named `credentials` or `*_credentials`
-        - Fields named `credentials` or `*_credentials` MUST be
-          of type `CredentialsMetaInput`
-        """
-        super().__pydantic_init_subclass__(**kwargs)
-
-        # Reset cached JSON schema to prevent inheriting it from parent class
-        cls.cached_jsonschema = {}
-
-        credentials_fields = cls.get_credentials_fields()
-
-        for field_name in cls.get_fields():
-            if is_credentials_field_name(field_name):
-                if field_name not in credentials_fields:
-                    raise TypeError(
-                        f"Credentials field '{field_name}' on {cls.__qualname__} "
-                        f"is not of type {CredentialsMetaInput.__name__}"
-                    )
-
-                CredentialsMetaInput.validate_credentials_field_schema(
-                    cls.get_field_schema(field_name), field_name
-                )
-
-            elif field_name in credentials_fields:
-                raise KeyError(
-                    f"Credentials field '{field_name}' on {cls.__qualname__} "
-                    "has invalid name: must be 'credentials' or *_credentials"
-                )
-
-    @classmethod
-    def get_credentials_fields(cls) -> dict[str, type[CredentialsMetaInput]]:
-        return {
-            field_name: info.annotation
-            for field_name, info in cls.model_fields.items()
-            if (
-                inspect.isclass(info.annotation)
-                and issubclass(
-                    get_origin(info.annotation) or info.annotation,
-                    CredentialsMetaInput,
-                )
-            )
-        }
-
-    @classmethod
-    def get_auto_credentials_fields(cls) -> dict[str, dict[str, Any]]:
-        """
-        Get fields that have auto_credentials metadata (e.g., GoogleDriveFileInput).
-
-        Returns a dict mapping kwarg_name -> {field_name, auto_credentials_config}
-
-        Raises:
-            ValueError: If multiple fields have the same kwarg_name, as this would
-                cause silent overwriting and only the last field would be processed.
-        """
-        result: dict[str, dict[str, Any]] = {}
-        schema = cls.jsonschema()
-        properties = schema.get("properties", {})
-
-        for field_name, field_schema in properties.items():
-            auto_creds = field_schema.get("auto_credentials")
-            if auto_creds:
-                kwarg_name = auto_creds.get("kwarg_name", "credentials")
-                if kwarg_name in result:
-                    raise ValueError(
-                        f"Duplicate auto_credentials kwarg_name '{kwarg_name}' "
-                        f"in fields '{result[kwarg_name]['field_name']}' and "
-                        f"'{field_name}' on {cls.__qualname__}"
-                    )
-                result[kwarg_name] = {
-                    "field_name": field_name,
-                    "config": auto_creds,
-                }
-        return result
-
-    @classmethod
-    def get_credentials_fields_info(cls) -> dict[str, CredentialsFieldInfo]:
-        result = {}
-
-        # Regular credentials fields
-        for field_name in cls.get_credentials_fields().keys():
-            result[field_name] = CredentialsFieldInfo.model_validate(
-                cls.get_field_schema(field_name), by_alias=True
-            )
-
-        # Auto-generated credentials fields (from GoogleDriveFileInput etc.)
-        for kwarg_name, info in cls.get_auto_credentials_fields().items():
-            config = info["config"]
-            # Build a schema-like dict that CredentialsFieldInfo can parse
-            auto_schema = {
-                "credentials_provider": [config.get("provider", "google")],
-                "credentials_types": [config.get("type", "oauth2")],
-                "credentials_scopes": config.get("scopes"),
-            }
-            result[kwarg_name] = CredentialsFieldInfo.model_validate(
-                auto_schema, by_alias=True
-            )
-
-        return result
-
-    @classmethod
-    def get_input_defaults(cls, data: BlockInput) -> BlockInput:
-        return data  # Return as is, by default.
-
-    @classmethod
-    def get_missing_links(cls, data: BlockInput, links: list["Link"]) -> set[str]:
-        input_fields_from_nodes = {link.sink_name for link in links}
-        return input_fields_from_nodes - set(data)
-
-    @classmethod
-    def get_missing_input(cls, data: BlockInput) -> set[str]:
-        return cls.get_required_fields() - set(data)
-
-
-class BlockSchemaInput(BlockSchema):
-    """
-    Base schema class for block inputs.
-    All block input schemas should extend this class for consistency.
-    """
-
-    pass
-
-
-class BlockSchemaOutput(BlockSchema):
-    """
-    Base schema class for block outputs that includes a standard error field.
-    All block output schemas should extend this class to ensure consistent error handling.
-    """
-
-    error: str = SchemaField(
-        description="Error message if the operation failed", default=""
-    )
-
-
-BlockSchemaInputType = TypeVar("BlockSchemaInputType", bound=BlockSchemaInput)
-BlockSchemaOutputType = TypeVar("BlockSchemaOutputType", bound=BlockSchemaOutput)
-
-
-class EmptyInputSchema(BlockSchemaInput):
-    pass
-
-
-class EmptyOutputSchema(BlockSchemaOutput):
-    pass
-
-
-# For backward compatibility - will be deprecated
-EmptySchema = EmptyOutputSchema
-
-
-# --8<-- [start:BlockWebhookConfig]
-class BlockManualWebhookConfig(BaseModel):
-    """
-    Configuration model for webhook-triggered blocks on which
-    the user has to manually set up the webhook at the provider.
-    """
-
-    provider: ProviderName
-    """The service provider that the webhook connects to"""
-
-    webhook_type: str
-    """
-    Identifier for the webhook type. E.g. GitHub has repo and organization level hooks.
-
-    Only for use in the corresponding `WebhooksManager`.
-    """
-
-    event_filter_input: str = ""
-    """
-    Name of the block's event filter input.
-    Leave empty if the corresponding webhook doesn't have distinct event/payload types.
-    """
-
-    event_format: str = "{event}"
-    """
-    Template string for the event(s) that a block instance subscribes to.
-    Applied individually to each event selected in the event filter input.
-
-    Example: `"pull_request.{event}"` -> `"pull_request.opened"`
-    """
-
-
-class BlockWebhookConfig(BlockManualWebhookConfig):
-    """
-    Configuration model for webhook-triggered blocks for which
-    the webhook can be automatically set up through the provider's API.
-    """
-
-    resource_format: str
-    """
-    Template string for the resource that a block instance subscribes to.
-    Fields will be filled from the block's inputs (except `payload`).
-
-    Example: `f"{repo}/pull_requests"` (note: not how it's actually implemented)
-
-    Only for use in the corresponding `WebhooksManager`.
-    """
-    # --8<-- [end:BlockWebhookConfig]
-
-
-class Block(ABC, Generic[BlockSchemaInputType, BlockSchemaOutputType]):
-    def __init__(
-        self,
-        id: str = "",
-        description: str = "",
-        contributors: list["ContributorDetails"] = [],
-        categories: set[BlockCategory] | None = None,
-        input_schema: Type[BlockSchemaInputType] = EmptyInputSchema,
-        output_schema: Type[BlockSchemaOutputType] = EmptyOutputSchema,
-        test_input: BlockInput | list[BlockInput] | None = None,
-        test_output: BlockTestOutput | list[BlockTestOutput] | None = None,
-        test_mock: dict[str, Any] | None = None,
-        test_credentials: Optional[Credentials | dict[str, Credentials]] = None,
-        disabled: bool = False,
-        static_output: bool = False,
-        block_type: BlockType = BlockType.STANDARD,
-        webhook_config: Optional[BlockWebhookConfig | BlockManualWebhookConfig] = None,
-        is_sensitive_action: bool = False,
-    ):
-        """
-        Initialize the block with the given schema.
-
-        Args:
-            id: The unique identifier for the block, this value will be persisted in the
-                DB. So it should be a unique and constant across the application run.
-                Use the UUID format for the ID.
-            description: The description of the block, explaining what the block does.
-            contributors: The list of contributors who contributed to the block.
-            input_schema: The schema, defined as a Pydantic model, for the input data.
-            output_schema: The schema, defined as a Pydantic model, for the output data.
-            test_input: The list or single sample input data for the block, for testing.
-            test_output: The list or single expected output if the test_input is run.
-            test_mock: function names on the block implementation to mock on test run.
-            disabled: If the block is disabled, it will not be available for execution.
-            static_output: Whether the output links of the block are static by default.
-        """
-        from backend.data.model import NodeExecutionStats
-
-        self.id = id
-        self.input_schema = input_schema
-        self.output_schema = output_schema
-        self.test_input = test_input
-        self.test_output = test_output
-        self.test_mock = test_mock
-        self.test_credentials = test_credentials
-        self.description = description
-        self.categories = categories or set()
-        self.contributors = contributors or set()
-        self.disabled = disabled
-        self.static_output = static_output
-        self.block_type = block_type
-        self.webhook_config = webhook_config
-        self.is_sensitive_action = is_sensitive_action
-        self.execution_stats: "NodeExecutionStats" = NodeExecutionStats()
-
-        if self.webhook_config:
-            if isinstance(self.webhook_config, BlockWebhookConfig):
-                # Enforce presence of credentials field on auto-setup webhook blocks
-                if not (cred_fields := self.input_schema.get_credentials_fields()):
-                    raise TypeError(
-                        "credentials field is required on auto-setup webhook blocks"
-                    )
-                # Disallow multiple credentials inputs on webhook blocks
-                elif len(cred_fields) > 1:
-                    raise ValueError(
-                        "Multiple credentials inputs not supported on webhook blocks"
-                    )
-
-                self.block_type = BlockType.WEBHOOK
-            else:
-                self.block_type = BlockType.WEBHOOK_MANUAL
-
-            # Enforce shape of webhook event filter, if present
-            if self.webhook_config.event_filter_input:
-                event_filter_field = self.input_schema.model_fields[
-                    self.webhook_config.event_filter_input
-                ]
-                if not (
-                    isinstance(event_filter_field.annotation, type)
-                    and issubclass(event_filter_field.annotation, BaseModel)
-                    and all(
-                        field.annotation is bool
-                        for field in event_filter_field.annotation.model_fields.values()
-                    )
-                ):
-                    raise NotImplementedError(
-                        f"{self.name} has an invalid webhook event selector: "
-                        "field must be a BaseModel and all its fields must be boolean"
-                    )
-
-            # Enforce presence of 'payload' input
-            if "payload" not in self.input_schema.model_fields:
-                raise TypeError(
-                    f"{self.name} is webhook-triggered but has no 'payload' input"
-                )
-
-            # Disable webhook-triggered block if webhook functionality not available
-            if not app_config.platform_base_url:
-                self.disabled = True
-
-    @abstractmethod
-    async def run(self, input_data: BlockSchemaInputType, **kwargs) -> BlockOutput:
-        """
-        Run the block with the given input data.
-        Args:
-            input_data: The input data with the structure of input_schema.
-
-        Kwargs: Currently 14/02/2025 these include
-            graph_id: The ID of the graph.
-            node_id: The ID of the node.
-            graph_exec_id: The ID of the graph execution.
-            node_exec_id: The ID of the node execution.
-            user_id: The ID of the user.
-
-        Returns:
-            A Generator that yields (output_name, output_data).
-            output_name: One of the output name defined in Block's output_schema.
-            output_data: The data for the output_name, matching the defined schema.
-        """
-        # --- satisfy the type checker, never executed -------------
-        if False:  # noqa: SIM115
-            yield "name", "value"  # pyright: ignore[reportMissingYield]
-        raise NotImplementedError(f"{self.name} does not implement the run method.")
-
-    async def run_once(
-        self, input_data: BlockSchemaInputType, output: str, **kwargs
-    ) -> Any:
-        async for item in self.run(input_data, **kwargs):
-            name, data = item
-            if name == output:
-                return data
-        raise ValueError(f"{self.name} did not produce any output for {output}")
-
-    def merge_stats(self, stats: "NodeExecutionStats") -> "NodeExecutionStats":
-        self.execution_stats += stats
-        return self.execution_stats
-
-    @property
-    def name(self):
-        return self.__class__.__name__
-
-    def to_dict(self):
-        return {
-            "id": self.id,
-            "name": self.name,
-            "inputSchema": self.input_schema.jsonschema(),
-            "outputSchema": self.output_schema.jsonschema(),
-            "description": self.description,
-            "categories": [category.dict() for category in self.categories],
-            "contributors": [
-                contributor.model_dump() for contributor in self.contributors
-            ],
-            "staticOutput": self.static_output,
-            "uiType": self.block_type.value,
-        }
-
-    def get_info(self) -> BlockInfo:
-        from backend.data.credit import get_block_cost
-
-        return BlockInfo(
-            id=self.id,
-            name=self.name,
-            inputSchema=self.input_schema.jsonschema(),
-            outputSchema=self.output_schema.jsonschema(),
-            costs=get_block_cost(self),
-            description=self.description,
-            categories=[category.dict() for category in self.categories],
-            contributors=[
-                contributor.model_dump() for contributor in self.contributors
-            ],
-            staticOutput=self.static_output,
-            uiType=self.block_type.value,
-        )
-
-    async def execute(self, input_data: BlockInput, **kwargs) -> BlockOutput:
-        try:
-            async for output_name, output_data in self._execute(input_data, **kwargs):
-                yield output_name, output_data
-        except Exception as ex:
-            if isinstance(ex, BlockError):
-                raise ex
-            else:
-                raise (
-                    BlockExecutionError
-                    if isinstance(ex, ValueError)
-                    else BlockUnknownError
-                )(
-                    message=str(ex),
-                    block_name=self.name,
-                    block_id=self.id,
-                ) from ex
-
-    async def is_block_exec_need_review(
-        self,
-        input_data: BlockInput,
-        *,
-        user_id: str,
-        node_id: str,
-        node_exec_id: str,
-        graph_exec_id: str,
-        graph_id: str,
-        graph_version: int,
-        execution_context: "ExecutionContext",
-        **kwargs,
-    ) -> tuple[bool, BlockInput]:
-        """
-        Check if this block execution needs human review and handle the review process.
-
-        Returns:
-            Tuple of (should_pause, input_data_to_use)
-            - should_pause: True if execution should be paused for review
-            - input_data_to_use: The input data to use (may be modified by reviewer)
-        """
-        if not (
-            self.is_sensitive_action and execution_context.sensitive_action_safe_mode
-        ):
-            return False, input_data
-
-        from backend.blocks.helpers.review import HITLReviewHelper
-
-        # Handle the review request and get decision
-        decision = await HITLReviewHelper.handle_review_decision(
-            input_data=input_data,
-            user_id=user_id,
-            node_id=node_id,
-            node_exec_id=node_exec_id,
-            graph_exec_id=graph_exec_id,
-            graph_id=graph_id,
-            graph_version=graph_version,
-            block_name=self.name,
-            editable=True,
-        )
-
-        if decision is None:
-            # We're awaiting review - pause execution
-            return True, input_data
-
-        if not decision.should_proceed:
-            # Review was rejected, raise an error to stop execution
-            raise BlockExecutionError(
-                message=f"Block execution rejected by reviewer: {decision.message}",
-                block_name=self.name,
-                block_id=self.id,
-            )
-
-        # Review was approved - use the potentially modified data
-        # ReviewResult.data must be a dict for block inputs
-        reviewed_data = decision.review_result.data
-        if not isinstance(reviewed_data, dict):
-            raise BlockExecutionError(
-                message=f"Review data must be a dict for block input, got {type(reviewed_data).__name__}",
-                block_name=self.name,
-                block_id=self.id,
-            )
-        return False, reviewed_data
-
-    async def _execute(self, input_data: BlockInput, **kwargs) -> BlockOutput:
-        # Check for review requirement only if running within a graph execution context
-        # Direct block execution (e.g., from chat) skips the review process
-        has_graph_context = all(
-            key in kwargs
-            for key in (
-                "node_exec_id",
-                "graph_exec_id",
-                "graph_id",
-                "execution_context",
-            )
-        )
-        if has_graph_context:
-            should_pause, input_data = await self.is_block_exec_need_review(
-                input_data, **kwargs
-            )
-            if should_pause:
-                return
-
-        # Validate the input data (original or reviewer-modified) once
-        if error := self.input_schema.validate_data(input_data):
-            raise BlockInputError(
-                message=f"Unable to execute block with invalid input data: {error}",
-                block_name=self.name,
-                block_id=self.id,
-            )
-
-        # Use the validated input data
-        async for output_name, output_data in self.run(
-            self.input_schema(**{k: v for k, v in input_data.items() if v is not None}),
-            **kwargs,
-        ):
-            if output_name == "error":
-                raise BlockExecutionError(
-                    message=output_data, block_name=self.name, block_id=self.id
-                )
-            if self.block_type == BlockType.STANDARD and (
-                error := self.output_schema.validate_field(output_name, output_data)
-            ):
-                raise BlockOutputError(
-                    message=f"Block produced an invalid output data: {error}",
-                    block_name=self.name,
-                    block_id=self.id,
-                )
-            yield output_name, output_data
-
-    def is_triggered_by_event_type(
-        self, trigger_config: dict[str, Any], event_type: str
-    ) -> bool:
-        if not self.webhook_config:
-            raise TypeError("This method can't be used on non-trigger blocks")
-        if not self.webhook_config.event_filter_input:
-            return True
-        event_filter = trigger_config.get(self.webhook_config.event_filter_input)
-        if not event_filter:
-            raise ValueError("Event filter is not configured on trigger")
-        return event_type in [
-            self.webhook_config.event_format.format(event=k)
-            for k in event_filter
-            if event_filter[k] is True
-        ]
-
-
-# Type alias for any block with standard input/output schemas
-AnyBlockSchema: TypeAlias = Block[BlockSchemaInput, BlockSchemaOutput]

autogpt_platform/backend/backend/blocks/_utils.py

@@ -1,122 +0,0 @@
-import logging
-import os
-
-from backend.integrations.providers import ProviderName
-
-from ._base import AnyBlockSchema
-
-logger = logging.getLogger(__name__)
-
-
-def is_block_auth_configured(
-    block_cls: type[AnyBlockSchema],
-) -> bool:
-    """
-    Check if a block has a valid authentication method configured at runtime.
-
-    For example if a block is an OAuth-only block and there env vars are not set,
-    do not show it in the UI.
-
-    """
-    from backend.sdk.registry import AutoRegistry
-
-    # Create an instance to access input_schema
-    try:
-        block = block_cls()
-    except Exception as e:
-        # If we can't create a block instance, assume it's not OAuth-only
-        logger.error(f"Error creating block instance for {block_cls.__name__}: {e}")
-        return True
-    logger.debug(
-        f"Checking if block {block_cls.__name__} has a valid provider configured"
-    )
-
-    # Get all credential inputs from input schema
-    credential_inputs = block.input_schema.get_credentials_fields_info()
-    required_inputs = block.input_schema.get_required_fields()
-    if not credential_inputs:
-        logger.debug(
-            f"Block {block_cls.__name__} has no credential inputs - Treating as valid"
-        )
-        return True
-
-    # Check credential inputs
-    if len(required_inputs.intersection(credential_inputs.keys())) == 0:
-        logger.debug(
-            f"Block {block_cls.__name__} has only optional credential inputs"
-            " - will work without credentials configured"
-        )
-
-    # Check if the credential inputs for this block are correctly configured
-    for field_name, field_info in credential_inputs.items():
-        provider_names = field_info.provider
-        if not provider_names:
-            logger.warning(
-                f"Block {block_cls.__name__} "
-                f"has credential input '{field_name}' with no provider options"
-                " - Disabling"
-            )
-            return False
-
-        # If a field has multiple possible providers, each one needs to be usable to
-        # prevent breaking the UX
-        for _provider_name in provider_names:
-            provider_name = _provider_name.value
-            if provider_name in ProviderName.__members__.values():
-                logger.debug(
-                    f"Block {block_cls.__name__} credential input '{field_name}' "
-                    f"provider '{provider_name}' is part of the legacy provider system"
-                    " - Treating as valid"
-                )
-                break
-
-            provider = AutoRegistry.get_provider(provider_name)
-            if not provider:
-                logger.warning(
-                    f"Block {block_cls.__name__} credential input '{field_name}' "
-                    f"refers to unknown provider '{provider_name}' - Disabling"
-                )
-                return False
-
-            # Check the provider's supported auth types
-            if field_info.supported_types != provider.supported_auth_types:
-                logger.warning(
-                    f"Block {block_cls.__name__} credential input '{field_name}' "
-                    f"has mismatched supported auth types (field <> Provider): "
-                    f"{field_info.supported_types} != {provider.supported_auth_types}"
-                )
-
-            if not (supported_auth_types := provider.supported_auth_types):
-                # No auth methods are been configured for this provider
-                logger.warning(
-                    f"Block {block_cls.__name__} credential input '{field_name}' "
-                    f"provider '{provider_name}' "
-                    "has no authentication methods configured - Disabling"
-                )
-                return False
-
-            # Check if provider supports OAuth
-            if "oauth2" in supported_auth_types:
-                # Check if OAuth environment variables are set
-                if (oauth_config := provider.oauth_config) and bool(
-                    os.getenv(oauth_config.client_id_env_var)
-                    and os.getenv(oauth_config.client_secret_env_var)
-                ):
-                    logger.debug(
-                        f"Block {block_cls.__name__} credential input '{field_name}' "
-                        f"provider '{provider_name}' is configured for OAuth"
-                    )
-                else:
-                    logger.error(
-                        f"Block {block_cls.__name__} credential input '{field_name}' "
-                        f"provider '{provider_name}' "
-                        "is missing OAuth client ID or secret - Disabling"
-                    )
-                    return False
-
-        logger.debug(
-            f"Block {block_cls.__name__} credential input '{field_name}' is valid; "
-            f"supported credential types: {', '.join(field_info.supported_types)}"
-        )
-
-    return True

autogpt_platform/backend/backend/blocks/agent.py

@@ -1,7 +1,7 @@
 import logging
-from typing import TYPE_CHECKING, Any, Optional
+from typing import Any, Optional
 
-from backend.blocks._base import (
+from backend.data.block import (
     Block,
     BlockCategory,
     BlockInput,
@@ -9,15 +9,13 @@ from backend.blocks._base import (
     BlockSchema,
     BlockSchemaInput,
     BlockType,
+    get_block,
 )
 from backend.data.execution import ExecutionContext, ExecutionStatus, NodesInputMasks
 from backend.data.model import NodeExecutionStats, SchemaField
 from backend.util.json import validate_with_jsonschema
 from backend.util.retry import func_retry
 
-if TYPE_CHECKING:
-    from backend.executor.utils import LogMetadata
-
 _logger = logging.getLogger(__name__)
 
 
@@ -126,10 +124,9 @@ class AgentExecutorBlock(Block):
         graph_version: int,
         graph_exec_id: str,
         user_id: str,
-        logger: "LogMetadata",
+        logger,
     ) -> BlockOutput:
 
-        from backend.blocks import get_block
         from backend.data.execution import ExecutionEventType
         from backend.executor import utils as execution_utils
 
@@ -201,7 +198,7 @@ class AgentExecutorBlock(Block):
         self,
         graph_exec_id: str,
         user_id: str,
-        logger: "LogMetadata",
+        logger,
     ) -> None:
         from backend.executor import utils as execution_utils
 

autogpt_platform/backend/backend/blocks/ai_condition.py

@@ -1,11 +1,5 @@
 from typing import Any
 
-from backend.blocks._base import (
-    BlockCategory,
-    BlockOutput,
-    BlockSchemaInput,
-    BlockSchemaOutput,
-)
 from backend.blocks.llm import (
     DEFAULT_LLM_MODEL,
     TEST_CREDENTIALS,
@@ -17,6 +11,12 @@ from backend.blocks.llm import (
     LLMResponse,
     llm_call,
 )
+from backend.data.block import (
+    BlockCategory,
+    BlockOutput,
+    BlockSchemaInput,
+    BlockSchemaOutput,
+)
 from backend.data.model import APIKeyCredentials, NodeExecutionStats, SchemaField
 
 

autogpt_platform/backend/backend/blocks/ai_image_customizer.py

@@ -6,7 +6,7 @@ from pydantic import SecretStr
 from replicate.client import Client as ReplicateClient
 from replicate.helpers import FileOutput
 
-from backend.blocks._base import (
+from backend.data.block import (
     Block,
     BlockCategory,
     BlockOutput,

autogpt_platform/backend/backend/blocks/ai_image_generator_block.py

@@ -5,12 +5,7 @@ from pydantic import SecretStr
 from replicate.client import Client as ReplicateClient
 from replicate.helpers import FileOutput
 
-from backend.blocks._base import (
-    Block,
-    BlockCategory,
-    BlockSchemaInput,
-    BlockSchemaOutput,
-)
+from backend.data.block import Block, BlockCategory, BlockSchemaInput, BlockSchemaOutput
 from backend.data.execution import ExecutionContext
 from backend.data.model import (
     APIKeyCredentials,

autogpt_platform/backend/backend/blocks/ai_music_generator.py

@@ -6,7 +6,7 @@ from typing import Literal
 from pydantic import SecretStr
 from replicate.client import Client as ReplicateClient
 
-from backend.blocks._base import (
+from backend.data.block import (
     Block,
     BlockCategory,
     BlockOutput,

autogpt_platform/backend/backend/blocks/ai_shortform_video_block.py

@@ -6,7 +6,7 @@ from typing import Literal
 
 from pydantic import SecretStr
 
-from backend.blocks._base import (
+from backend.data.block import (
     Block,
     BlockCategory,
     BlockOutput,