Compare commits

..

25 Commits

Author SHA1 Message Date
Swifty
b33732fac1 ci: trigger CI to test optimizations
🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2025-12-19 10:47:51 +01:00
Swifty
62c2d1cdc7 ci(platform): optimize CI pipelines for faster builds
- Remove ClamAV from backend CI (saves 3-5 min per run)
  - ClamAV tests now only run on merge to master via new security-ci workflow
- Reduce Python matrix to 3.13 only (matches Docker image)
- Add Docker image tar caching to frontend and fullstack CI
- Add Playwright browser caching to frontend CI (saves 30-60s)
- Reduce service wait timeouts from 60s to 30s

Estimated savings:
- Backend CI: 50-60% faster
- Frontend CI: 30-40% faster
- Fullstack CI: 30-40% faster

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2025-12-19 10:43:39 +01:00
Swifty
93e5c40189 Merge branch 'swiftyos/vector-search' of github.com:Significant-Gravitas/AutoGPT into swiftyos/vector-search 2025-12-19 10:07:16 +01:00
Swifty
4ea2411fda fix(backend): graceful fallback to BM25-only when embedding unavailable
When OpenAI API key is not configured, hybrid search now:
- Catches embedding generation failures and logs a warning
- Falls back to BM25-only search instead of failing
- Adjusts SQL queries to handle NULL query embedding
- Uses only BM25 score filter when vector search is unavailable

This ensures search works in CI and environments without OpenAI keys.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2025-12-19 10:02:53 +01:00
Swifty
95154f03e6 Merge branch 'dev' into swiftyos/vector-search 2025-12-18 19:54:24 +01:00
Swifty
f4b3358cb3 fix(backend): update route tests to include filter_mode parameter
Add filter_mode='permissive' to all get_store_agents mock assertions
to match the updated function signature.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2025-12-18 19:51:46 +01:00
Swifty
48f8c70e6f update openapi.json 2025-12-18 19:48:17 +01:00
Swifty
3e1a8c800c Merge branch 'dev' into swiftyos/vector-search 2025-12-18 19:37:41 +01:00
Swifty
bcf3a0cd9c add openapi.json 2025-12-18 19:37:12 +01:00
Swifty
4e0ae67067 Merge branch 'swiftyos/vector-search' of github.com:Significant-Gravitas/AutoGPT into swiftyos/vector-search 2025-12-18 19:29:17 +01:00
Swifty
32acf066d0 Merge branch 'dev' into swiftyos/vector-search 2025-12-18 19:28:51 +01:00
Swifty
8268d919f5 refactor(backend): simplify EmbeddingService client property
Address PR review: replace manual double-checked locking with
@functools.cached_property for cleaner, simpler lazy initialization.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2025-12-18 19:27:37 +01:00
Swifty
c7063a46a6 fix(backend): restore sorted_by support for hybrid search
- Add sorted_by parameter support to hybrid search (rating, runs, name, updated_at)
- Use sorted_by as primary sort with RRF score as secondary tiebreaker
- Add clarifying comments explaining why count query needs full CTE structure

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2025-12-18 19:13:33 +01:00
Swifty
3509db9ebd feat(backend): implement hybrid search with BM25, vector, and RRF ranking
Implement hybrid search for the store combining:
- BM25 full-text search (PostgreSQL tsvector with ts_rank_cd)
- Vector semantic similarity (pgvector cosine distance)
- Popularity signal (run counts as PageRank proxy)

Results are ranked using Reciprocal Rank Fusion (RRF) formula.

Key changes:
- Add migration for BM25 trigger with weighted fields and GIN index
- Add SearchFilterMode enum (strict/permissive/combined)
- Update get_store_agents() with hybrid search SQL using CTEs
- Add filter_mode parameter to API endpoint (default: permissive)
- Add RRF score threshold (0.02) to filter irrelevant results

Thresholds:
- Vector similarity: >= 0.4
- BM25 relevance: >= 0.05
- RRF score: >= 0.02

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-12-18 19:02:56 +01:00
Swifty
79534efa68 Merge branch 'dev' into swiftyos/vector-search 2025-12-09 08:55:07 +01:00
Swifty
69d0c05017 fix(backend): address PR review comments for vector search
- Make EmbeddingService API key validation lazy (doesn't break startup)
- Make chat service OpenAI client creation lazy with @functools.cache
- Add thread-safe double-checked locking for client initialization
- Remove unnecessary catch/re-raise error handling in embeddings
- Replace async lock singleton with simpler @functools.cache pattern
- Update backfill script to use get_embedding_service() singleton
- Fix test mocks to use MagicMock instead of AsyncMock for sync function
2025-12-09 08:54:11 +01:00
Swifty
521dbdc25f attempting to fix ci 2025-12-05 16:36:54 +01:00
Swifty
3b9abbcdbc fix(backend): include agent_output_demo in vector search migration view
Add agentOutputDemoUrl field to StoreAgent view in the vector search
migration to preserve the field added by the merged output video column
migration.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-12-05 16:16:27 +01:00
Swifty
e0cd070e4d Merge branch 'dev' into swiftyos/vector-search 2025-12-05 16:09:11 +01:00
Swifty
b6b7b77ddd add a similarity limit 2025-12-04 17:30:28 +01:00
Swifty
fc5cf113a7 fix(backend): race condition in embedding service singleton initialization
Use asyncio.Lock with double-checked locking pattern to prevent multiple
EmbeddingService instances being created under concurrent async access.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-12-04 17:20:25 +01:00
Swifty
9a5a041102 dont drop search columns 2025-12-04 17:06:40 +01:00
Swifty
1137cfde48 Merge branch 'dev' into swiftyos/vector-search 2025-12-04 16:54:05 +01:00
Swifty
da8e7405b0 Merge branch 'dev' into swiftyos/vector-search 2025-12-04 16:12:42 +01:00
Swifty
7b6db6e260 add vector search 2025-12-04 16:05:47 +01:00
500 changed files with 15483 additions and 30203 deletions

View File

@@ -32,7 +32,9 @@ jobs:
strategy:
fail-fast: false
matrix:
python-version: ["3.11", "3.12", "3.13"]
# Use Python 3.13 to match Docker image (see backend/Dockerfile)
# ClamAV tests moved to platform-backend-security-ci.yml (runs on merge to master)
python-version: ["3.13"]
runs-on: ubuntu-latest
services:
@@ -48,23 +50,6 @@ jobs:
env:
RABBITMQ_DEFAULT_USER: ${{ env.RABBITMQ_DEFAULT_USER }}
RABBITMQ_DEFAULT_PASS: ${{ env.RABBITMQ_DEFAULT_PASS }}
clamav:
image: clamav/clamav-debian:latest
ports:
- 3310:3310
env:
CLAMAV_NO_FRESHCLAMD: false
CLAMD_CONF_StreamMaxLength: 50M
CLAMD_CONF_MaxFileSize: 100M
CLAMD_CONF_MaxScanSize: 100M
CLAMD_CONF_MaxThreads: 4
CLAMD_CONF_ReadTimeout: 300
options: >-
--health-cmd "clamdscan --version || exit 1"
--health-interval 30s
--health-timeout 10s
--health-retries 5
--health-start-period 180s
steps:
- name: Checkout repository
@@ -146,35 +131,6 @@ jobs:
# outputs:
# DB_URL, API_URL, GRAPHQL_URL, ANON_KEY, SERVICE_ROLE_KEY, JWT_SECRET
- name: Wait for ClamAV to be ready
run: |
echo "Waiting for ClamAV daemon to start..."
max_attempts=60
attempt=0
until nc -z localhost 3310 || [ $attempt -eq $max_attempts ]; do
echo "ClamAV is unavailable - sleeping (attempt $((attempt+1))/$max_attempts)"
sleep 5
attempt=$((attempt+1))
done
if [ $attempt -eq $max_attempts ]; then
echo "ClamAV failed to start after $((max_attempts*5)) seconds"
echo "Checking ClamAV service logs..."
docker logs $(docker ps -q --filter "ancestor=clamav/clamav-debian:latest") 2>&1 | tail -50 || echo "No ClamAV container found"
exit 1
fi
echo "ClamAV is ready!"
# Verify ClamAV is responsive
echo "Testing ClamAV connection..."
timeout 10 bash -c 'echo "PING" | nc localhost 3310' || {
echo "ClamAV is not responding to PING"
docker logs $(docker ps -q --filter "ancestor=clamav/clamav-debian:latest") 2>&1 | tail -50 || echo "No ClamAV container found"
exit 1
}
- name: Run Database Migrations
run: poetry run prisma migrate dev --name updates
env:

View File

@@ -0,0 +1,145 @@
name: AutoGPT Platform - Backend Security CI
# This workflow runs ClamAV-dependent security tests.
# It only runs on merge to master to avoid the 3-5 minute ClamAV startup time on every PR.
on:
push:
branches: [master]
paths:
- "autogpt_platform/backend/**/file*.py"
- "autogpt_platform/backend/**/scan*.py"
- "autogpt_platform/backend/**/virus*.py"
- "autogpt_platform/backend/**/media*.py"
- ".github/workflows/platform-backend-security-ci.yml"
concurrency:
group: ${{ format('backend-security-ci-{0}', github.sha) }}
cancel-in-progress: false
defaults:
run:
shell: bash
working-directory: autogpt_platform/backend
jobs:
security-tests:
runs-on: ubuntu-latest
timeout-minutes: 15
services:
redis:
image: redis:latest
ports:
- 6379:6379
clamav:
image: clamav/clamav-debian:latest
ports:
- 3310:3310
env:
CLAMAV_NO_FRESHCLAMD: false
CLAMD_CONF_StreamMaxLength: 50M
CLAMD_CONF_MaxFileSize: 100M
CLAMD_CONF_MaxScanSize: 100M
CLAMD_CONF_MaxThreads: 4
CLAMD_CONF_ReadTimeout: 300
options: >-
--health-cmd "clamdscan --version || exit 1"
--health-interval 30s
--health-timeout 10s
--health-retries 5
--health-start-period 180s
steps:
- name: Checkout repository
uses: actions/checkout@v4
with:
fetch-depth: 0
submodules: true
- name: Set up Python 3.13
uses: actions/setup-python@v5
with:
python-version: "3.13"
- name: Setup Supabase
uses: supabase/setup-cli@v1
with:
version: 1.178.1
- name: Set up Python dependency cache
uses: actions/cache@v4
with:
path: ~/.cache/pypoetry
key: poetry-${{ runner.os }}-${{ hashFiles('autogpt_platform/backend/poetry.lock') }}
- name: Install Poetry
run: |
HEAD_POETRY_VERSION=$(python ../../.github/workflows/scripts/get_package_version_from_lockfile.py poetry)
echo "Using Poetry version ${HEAD_POETRY_VERSION}"
curl -sSL https://install.python-poetry.org | POETRY_VERSION=$HEAD_POETRY_VERSION python3 -
- name: Install Python dependencies
run: poetry install
- name: Generate Prisma Client
run: poetry run prisma generate
- id: supabase
name: Start Supabase
working-directory: .
run: |
supabase init
supabase start --exclude postgres-meta,realtime,storage-api,imgproxy,inbucket,studio,edge-runtime,logflare,vector,supavisor
supabase status -o env | sed 's/="/=/; s/"$//' >> $GITHUB_OUTPUT
- name: Wait for ClamAV to be ready
run: |
echo "Waiting for ClamAV daemon to start..."
max_attempts=60
attempt=0
until nc -z localhost 3310 || [ $attempt -eq $max_attempts ]; do
echo "ClamAV is unavailable - sleeping (attempt $((attempt+1))/$max_attempts)"
sleep 5
attempt=$((attempt+1))
done
if [ $attempt -eq $max_attempts ]; then
echo "ClamAV failed to start after $((max_attempts*5)) seconds"
exit 1
fi
echo "ClamAV is ready!"
- name: Run Database Migrations
run: poetry run prisma migrate dev --name updates
env:
DATABASE_URL: ${{ steps.supabase.outputs.DB_URL }}
DIRECT_URL: ${{ steps.supabase.outputs.DB_URL }}
- name: Run security-related tests
run: |
poetry run pytest -v \
backend/util/virus_scanner_test.py \
backend/util/file_test.py \
backend/server/v2/store/media_test.py \
-x
env:
DATABASE_URL: ${{ steps.supabase.outputs.DB_URL }}
DIRECT_URL: ${{ steps.supabase.outputs.DB_URL }}
SUPABASE_URL: ${{ steps.supabase.outputs.API_URL }}
SUPABASE_SERVICE_ROLE_KEY: ${{ steps.supabase.outputs.SERVICE_ROLE_KEY }}
JWT_VERIFY_KEY: ${{ steps.supabase.outputs.JWT_SECRET }}
REDIS_HOST: "localhost"
REDIS_PORT: "6379"
ENCRYPTION_KEY: "dvziYgz0KSK8FENhju0ZYi8-fRTfAdlz6YLhdB_jhNw="
CLAMAV_SERVICE_HOST: "localhost"
CLAMAV_SERVICE_PORT: "3310"
CLAMAV_SERVICE_ENABLED: "true"
env:
CI: true
PLAIN_OUTPUT: True
RUN_ENV: local
PORT: 8080

View File

@@ -154,35 +154,78 @@ jobs:
- name: Set up Docker Buildx
uses: docker/setup-buildx-action@v3
- name: Cache Docker layers
# Docker image tar caching - loads images from cache in parallel for faster startup
- name: Set up Docker image cache
id: docker-cache
uses: actions/cache@v4
with:
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') }}
path: ~/docker-cache
key: docker-images-frontend-${{ runner.os }}-${{ hashFiles('autogpt_platform/docker-compose.yml') }}
restore-keys: |
${{ runner.os }}-buildx-frontend-test-
docker-images-frontend-${{ runner.os }}-
- name: Load or pull Docker images
working-directory: autogpt_platform
run: |
mkdir -p ~/docker-cache
# Define image list for easy maintenance
IMAGES=(
"redis:latest"
"rabbitmq:management"
"kong:2.8.1"
"supabase/gotrue:v2.170.0"
"supabase/postgres:15.8.1.049"
)
# Check if any cached tar files exist
if ls ~/docker-cache/*.tar 1> /dev/null 2>&1; then
echo "Docker cache found, loading images in parallel..."
for image in "${IMAGES[@]}"; do
filename=$(echo "$image" | tr ':/' '--')
if [ -f ~/docker-cache/${filename}.tar ]; then
echo "Loading $image..."
docker load -i ~/docker-cache/${filename}.tar || echo "Warning: Failed to load $image from cache" &
fi
done
wait
echo "All cached images loaded"
else
echo "No Docker cache found, pulling images in parallel..."
for image in "${IMAGES[@]}"; do
docker pull "$image" &
done
wait
# Only save cache on main branches (not PRs) to avoid cache pollution
if [[ "${{ github.ref }}" == "refs/heads/master" ]] || [[ "${{ github.ref }}" == "refs/heads/dev" ]]; then
echo "Saving Docker images to cache in parallel..."
for image in "${IMAGES[@]}"; do
filename=$(echo "$image" | tr ':/' '--')
echo "Saving $image..."
docker save -o ~/docker-cache/${filename}.tar "$image" || echo "Warning: Failed to save $image" &
done
wait
echo "Docker image cache saved"
else
echo "Skipping cache save for PR/feature branch"
fi
fi
echo "Docker images ready for use"
- name: Run docker compose
run: |
NEXT_PUBLIC_PW_TEST=true docker compose -f ../docker-compose.yml up -d
env:
DOCKER_BUILDKIT: 1
BUILDX_CACHE_FROM: type=local,src=/tmp/.buildx-cache
BUILDX_CACHE_TO: type=local,dest=/tmp/.buildx-cache-new,mode=max
- name: Move cache
run: |
rm -rf /tmp/.buildx-cache
if [ -d "/tmp/.buildx-cache-new" ]; then
mv /tmp/.buildx-cache-new /tmp/.buildx-cache
fi
- name: Wait for services to be ready
run: |
echo "Waiting for rest_server to be ready..."
timeout 60 sh -c 'until curl -f http://localhost:8006/health 2>/dev/null; do sleep 2; done' || echo "Rest server health check timeout, continuing..."
timeout 30 sh -c 'until curl -f http://localhost:8006/health 2>/dev/null; do sleep 2; done' || echo "Rest server health check timeout, continuing..."
echo "Waiting for database to be ready..."
timeout 60 sh -c 'until docker compose -f ../docker-compose.yml exec -T db pg_isready -U postgres 2>/dev/null; do sleep 2; done' || echo "Database ready check timeout, continuing..."
timeout 30 sh -c 'until docker compose -f ../docker-compose.yml exec -T db pg_isready -U postgres 2>/dev/null; do sleep 2; done' || echo "Database ready check timeout, continuing..."
- name: Create E2E test data
run: |
@@ -221,9 +264,27 @@ jobs:
- name: Install dependencies
run: pnpm install --frozen-lockfile
- name: Install Browser 'chromium'
# Playwright browser caching - saves 30-60s when cache hits
- name: Get Playwright version
id: playwright-version
run: |
echo "version=$(pnpm list @playwright/test --json | jq -r '.[0].dependencies["@playwright/test"].version')" >> $GITHUB_OUTPUT
- name: Cache Playwright browsers
uses: actions/cache@v4
id: playwright-cache
with:
path: ~/.cache/ms-playwright
key: playwright-${{ runner.os }}-${{ steps.playwright-version.outputs.version }}
- name: Install Playwright browsers
if: steps.playwright-cache.outputs.cache-hit != 'true'
run: pnpm playwright install --with-deps chromium
- name: Install Playwright deps only (when cache hit)
if: steps.playwright-cache.outputs.cache-hit == 'true'
run: pnpm playwright install-deps chromium
- name: Run Playwright tests
run: pnpm test:no-build

View File

@@ -83,6 +83,66 @@ jobs:
run: |
cp ../backend/.env.default ../backend/.env
# Docker image tar caching - loads images from cache in parallel for faster startup
- name: Set up Docker image cache
id: docker-cache
uses: actions/cache@v4
with:
path: ~/docker-cache
key: docker-images-fullstack-${{ runner.os }}-${{ hashFiles('autogpt_platform/docker-compose.yml') }}
restore-keys: |
docker-images-fullstack-${{ runner.os }}-
- name: Load or pull Docker images
working-directory: autogpt_platform
run: |
mkdir -p ~/docker-cache
# Define image list for easy maintenance
IMAGES=(
"redis:latest"
"rabbitmq:management"
"kong:2.8.1"
"supabase/gotrue:v2.170.0"
"supabase/postgres:15.8.1.049"
)
# Check if any cached tar files exist
if ls ~/docker-cache/*.tar 1> /dev/null 2>&1; then
echo "Docker cache found, loading images in parallel..."
for image in "${IMAGES[@]}"; do
filename=$(echo "$image" | tr ':/' '--')
if [ -f ~/docker-cache/${filename}.tar ]; then
echo "Loading $image..."
docker load -i ~/docker-cache/${filename}.tar || echo "Warning: Failed to load $image from cache" &
fi
done
wait
echo "All cached images loaded"
else
echo "No Docker cache found, pulling images in parallel..."
for image in "${IMAGES[@]}"; do
docker pull "$image" &
done
wait
# Only save cache on main branches (not PRs) to avoid cache pollution
if [[ "${{ github.ref }}" == "refs/heads/master" ]] || [[ "${{ github.ref }}" == "refs/heads/dev" ]]; then
echo "Saving Docker images to cache in parallel..."
for image in "${IMAGES[@]}"; do
filename=$(echo "$image" | tr ':/' '--')
echo "Saving $image..."
docker save -o ~/docker-cache/${filename}.tar "$image" || echo "Warning: Failed to save $image" &
done
wait
echo "Docker image cache saved"
else
echo "Skipping cache save for PR/feature branch"
fi
fi
echo "Docker images ready for use"
- name: Run docker compose
run: |
docker compose -f ../docker-compose.yml --profile local --profile deps_backend up -d
@@ -104,9 +164,9 @@ jobs:
- name: Wait for services to be ready
run: |
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..."
timeout 30 sh -c 'until curl -f http://localhost:8006/health 2>/dev/null; do sleep 2; done' || echo "Rest server health check timeout, continuing..."
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..."
timeout 30 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: Generate API queries
run: pnpm generate:api:force

View File

@@ -6,11 +6,12 @@ start-core:
# Stop core services
stop-core:
docker compose stop
docker compose stop deps
reset-db:
docker compose stop db
rm -rf db/docker/volumes/db/data
cd backend && poetry run prisma migrate deploy
cd backend && poetry run prisma generate
# View logs for core services
logs-core:
@@ -57,4 +58,4 @@ help:
@echo " run-backend - Run the backend FastAPI server"
@echo " run-frontend - Run the frontend Next.js development server"
@echo " test-data - Run the test data creator"
@echo " load-store-agents - Load store agents from agents/ folder into test database"
@echo " load-store-agents - Load store agents from agents/ folder into test database"

View File

@@ -57,9 +57,6 @@ class APIKeySmith:
def hash_key(self, raw_key: str) -> tuple[str, str]:
"""Migrate a legacy hash to secure hash format."""
if not raw_key.startswith(self.PREFIX):
raise ValueError("Key without 'agpt_' prefix would fail validation")
salt = self._generate_salt()
hash = self._hash_key_with_salt(raw_key, salt)
return hash, salt.hex()

View File

@@ -1,25 +1,29 @@
from fastapi import FastAPI
from fastapi.openapi.utils import get_openapi
from .jwt_utils import bearer_jwt_auth
def add_auth_responses_to_openapi(app: FastAPI) -> None:
"""
Patch a FastAPI instance's `openapi()` method to add 401 responses
Set up custom OpenAPI schema generation that adds 401 responses
to all authenticated endpoints.
This is needed when using HTTPBearer with auto_error=False to get proper
401 responses instead of 403, but FastAPI only automatically adds security
responses when auto_error=True.
"""
# Wrap current method to allow stacking OpenAPI schema modifiers like this
wrapped_openapi = app.openapi
def custom_openapi():
if app.openapi_schema:
return app.openapi_schema
openapi_schema = wrapped_openapi()
openapi_schema = get_openapi(
title=app.title,
version=app.version,
description=app.description,
routes=app.routes,
)
# Add 401 response to all endpoints that have security requirements
for path, methods in openapi_schema["paths"].items():

View File

@@ -108,7 +108,7 @@ import fastapi.testclient
import pytest
from pytest_snapshot.plugin import Snapshot
from backend.api.features.myroute import router
from backend.server.v2.myroute import router
app = fastapi.FastAPI()
app.include_router(router)
@@ -149,7 +149,7 @@ These provide the easiest way to set up authentication mocking in test modules:
import fastapi
import fastapi.testclient
import pytest
from backend.api.features.myroute import router
from backend.server.v2.myroute import router
app = fastapi.FastAPI()
app.include_router(router)

View File

@@ -1,25 +0,0 @@
from fastapi import FastAPI
from backend.api.middleware.security import SecurityHeadersMiddleware
from backend.monitoring.instrumentation import instrument_fastapi
from .v1.routes import v1_router
external_api = FastAPI(
title="AutoGPT External API",
description="External API for AutoGPT integrations",
docs_url="/docs",
version="1.0",
)
external_api.add_middleware(SecurityHeadersMiddleware)
external_api.include_router(v1_router, prefix="/v1")
# Add Prometheus instrumentation
instrument_fastapi(
external_api,
service_name="external-api",
expose_endpoint=True,
endpoint="/metrics",
include_in_schema=True,
)

View File

@@ -1,107 +0,0 @@
from fastapi import HTTPException, Security, status
from fastapi.security import APIKeyHeader, HTTPAuthorizationCredentials, HTTPBearer
from prisma.enums import APIKeyPermission
from backend.data.auth.api_key import APIKeyInfo, validate_api_key
from backend.data.auth.base import APIAuthorizationInfo
from backend.data.auth.oauth import (
InvalidClientError,
InvalidTokenError,
OAuthAccessTokenInfo,
validate_access_token,
)
api_key_header = APIKeyHeader(name="X-API-Key", auto_error=False)
bearer_auth = HTTPBearer(auto_error=False)
async def require_api_key(api_key: str | None = Security(api_key_header)) -> APIKeyInfo:
"""Middleware for API key authentication only"""
if api_key is None:
raise HTTPException(
status_code=status.HTTP_401_UNAUTHORIZED, detail="Missing API key"
)
api_key_obj = await validate_api_key(api_key)
if not api_key_obj:
raise HTTPException(
status_code=status.HTTP_401_UNAUTHORIZED, detail="Invalid API key"
)
return api_key_obj
async def require_access_token(
bearer: HTTPAuthorizationCredentials | None = Security(bearer_auth),
) -> OAuthAccessTokenInfo:
"""Middleware for OAuth access token authentication only"""
if bearer is None:
raise HTTPException(
status_code=status.HTTP_401_UNAUTHORIZED,
detail="Missing Authorization header",
)
try:
token_info, _ = await validate_access_token(bearer.credentials)
except (InvalidClientError, InvalidTokenError) as e:
raise HTTPException(status_code=status.HTTP_401_UNAUTHORIZED, detail=str(e))
return token_info
async def require_auth(
api_key: str | None = Security(api_key_header),
bearer: HTTPAuthorizationCredentials | None = Security(bearer_auth),
) -> APIAuthorizationInfo:
"""
Unified authentication middleware supporting both API keys and OAuth tokens.
Supports two authentication methods, which are checked in order:
1. X-API-Key header (existing API key authentication)
2. Authorization: Bearer <token> header (OAuth access token)
Returns:
APIAuthorizationInfo: base class of both APIKeyInfo and OAuthAccessTokenInfo.
"""
# Try API key first
if api_key is not None:
api_key_info = await validate_api_key(api_key)
if api_key_info:
return api_key_info
raise HTTPException(
status_code=status.HTTP_401_UNAUTHORIZED, detail="Invalid API key"
)
# Try OAuth bearer token
if bearer is not None:
try:
token_info, _ = await validate_access_token(bearer.credentials)
return token_info
except (InvalidClientError, InvalidTokenError) as e:
raise HTTPException(status_code=status.HTTP_401_UNAUTHORIZED, detail=str(e))
# No credentials provided
raise HTTPException(
status_code=status.HTTP_401_UNAUTHORIZED,
detail="Missing authentication. Provide API key or access token.",
)
def require_permission(permission: APIKeyPermission):
"""
Dependency function for checking specific permissions
(works with API keys and OAuth tokens)
"""
async def check_permission(
auth: APIAuthorizationInfo = Security(require_auth),
) -> APIAuthorizationInfo:
if permission not in auth.scopes:
raise HTTPException(
status_code=status.HTTP_403_FORBIDDEN,
detail=f"Missing required permission: {permission.value}",
)
return auth
return check_permission

View File

@@ -1,340 +0,0 @@
"""Tests for analytics API endpoints."""
import json
from unittest.mock import AsyncMock, Mock
import fastapi
import fastapi.testclient
import pytest
import pytest_mock
from pytest_snapshot.plugin import Snapshot
from .analytics import router as analytics_router
app = fastapi.FastAPI()
app.include_router(analytics_router)
client = fastapi.testclient.TestClient(app)
@pytest.fixture(autouse=True)
def setup_app_auth(mock_jwt_user):
"""Setup auth overrides for all tests in this module."""
from autogpt_libs.auth.jwt_utils import get_jwt_payload
app.dependency_overrides[get_jwt_payload] = mock_jwt_user["get_jwt_payload"]
yield
app.dependency_overrides.clear()
# =============================================================================
# /log_raw_metric endpoint tests
# =============================================================================
def test_log_raw_metric_success(
mocker: pytest_mock.MockFixture,
configured_snapshot: Snapshot,
test_user_id: str,
) -> None:
"""Test successful raw metric logging."""
mock_result = Mock(id="metric-123-uuid")
mock_log_metric = mocker.patch(
"backend.data.analytics.log_raw_metric",
new_callable=AsyncMock,
return_value=mock_result,
)
request_data = {
"metric_name": "page_load_time",
"metric_value": 2.5,
"data_string": "/dashboard",
}
response = client.post("/log_raw_metric", json=request_data)
assert response.status_code == 200, f"Unexpected response: {response.text}"
assert response.json() == "metric-123-uuid"
mock_log_metric.assert_called_once_with(
user_id=test_user_id,
metric_name="page_load_time",
metric_value=2.5,
data_string="/dashboard",
)
configured_snapshot.assert_match(
json.dumps({"metric_id": response.json()}, indent=2, sort_keys=True),
"analytics_log_metric_success",
)
@pytest.mark.parametrize(
"metric_value,metric_name,data_string,test_id",
[
(100, "api_calls_count", "external_api", "integer_value"),
(0, "error_count", "no_errors", "zero_value"),
(-5.2, "temperature_delta", "cooling", "negative_value"),
(1.23456789, "precision_test", "float_precision", "float_precision"),
(999999999, "large_number", "max_value", "large_number"),
(0.0000001, "tiny_number", "min_value", "tiny_number"),
],
)
def test_log_raw_metric_various_values(
mocker: pytest_mock.MockFixture,
configured_snapshot: Snapshot,
metric_value: float,
metric_name: str,
data_string: str,
test_id: str,
) -> None:
"""Test raw metric logging with various metric values."""
mock_result = Mock(id=f"metric-{test_id}-uuid")
mocker.patch(
"backend.data.analytics.log_raw_metric",
new_callable=AsyncMock,
return_value=mock_result,
)
request_data = {
"metric_name": metric_name,
"metric_value": metric_value,
"data_string": data_string,
}
response = client.post("/log_raw_metric", json=request_data)
assert response.status_code == 200, f"Failed for {test_id}: {response.text}"
configured_snapshot.assert_match(
json.dumps(
{"metric_id": response.json(), "test_case": test_id},
indent=2,
sort_keys=True,
),
f"analytics_metric_{test_id}",
)
@pytest.mark.parametrize(
"invalid_data,expected_error",
[
({}, "Field required"),
({"metric_name": "test"}, "Field required"),
(
{"metric_name": "test", "metric_value": "not_a_number", "data_string": "x"},
"Input should be a valid number",
),
(
{"metric_name": "", "metric_value": 1.0, "data_string": "test"},
"String should have at least 1 character",
),
(
{"metric_name": "test", "metric_value": 1.0, "data_string": ""},
"String should have at least 1 character",
),
],
ids=[
"empty_request",
"missing_metric_value_and_data_string",
"invalid_metric_value_type",
"empty_metric_name",
"empty_data_string",
],
)
def test_log_raw_metric_validation_errors(
invalid_data: dict,
expected_error: str,
) -> None:
"""Test validation errors for invalid metric requests."""
response = client.post("/log_raw_metric", json=invalid_data)
assert response.status_code == 422
error_detail = response.json()
assert "detail" in error_detail, f"Missing 'detail' in error: {error_detail}"
error_text = json.dumps(error_detail)
assert (
expected_error in error_text
), f"Expected '{expected_error}' in error response: {error_text}"
def test_log_raw_metric_service_error(
mocker: pytest_mock.MockFixture,
test_user_id: str,
) -> None:
"""Test error handling when analytics service fails."""
mocker.patch(
"backend.data.analytics.log_raw_metric",
new_callable=AsyncMock,
side_effect=Exception("Database connection failed"),
)
request_data = {
"metric_name": "test_metric",
"metric_value": 1.0,
"data_string": "test",
}
response = client.post("/log_raw_metric", json=request_data)
assert response.status_code == 500
error_detail = response.json()["detail"]
assert "Database connection failed" in error_detail["message"]
assert "hint" in error_detail
# =============================================================================
# /log_raw_analytics endpoint tests
# =============================================================================
def test_log_raw_analytics_success(
mocker: pytest_mock.MockFixture,
configured_snapshot: Snapshot,
test_user_id: str,
) -> None:
"""Test successful raw analytics logging."""
mock_result = Mock(id="analytics-789-uuid")
mock_log_analytics = mocker.patch(
"backend.data.analytics.log_raw_analytics",
new_callable=AsyncMock,
return_value=mock_result,
)
request_data = {
"type": "user_action",
"data": {
"action": "button_click",
"button_id": "submit_form",
"timestamp": "2023-01-01T00:00:00Z",
"metadata": {"form_type": "registration", "fields_filled": 5},
},
"data_index": "button_click_submit_form",
}
response = client.post("/log_raw_analytics", json=request_data)
assert response.status_code == 200, f"Unexpected response: {response.text}"
assert response.json() == "analytics-789-uuid"
mock_log_analytics.assert_called_once_with(
test_user_id,
"user_action",
request_data["data"],
"button_click_submit_form",
)
configured_snapshot.assert_match(
json.dumps({"analytics_id": response.json()}, indent=2, sort_keys=True),
"analytics_log_analytics_success",
)
def test_log_raw_analytics_complex_data(
mocker: pytest_mock.MockFixture,
configured_snapshot: Snapshot,
) -> None:
"""Test raw analytics logging with complex nested data structures."""
mock_result = Mock(id="analytics-complex-uuid")
mocker.patch(
"backend.data.analytics.log_raw_analytics",
new_callable=AsyncMock,
return_value=mock_result,
)
request_data = {
"type": "agent_execution",
"data": {
"agent_id": "agent_123",
"execution_id": "exec_456",
"status": "completed",
"duration_ms": 3500,
"nodes_executed": 15,
"blocks_used": [
{"block_id": "llm_block", "count": 3},
{"block_id": "http_block", "count": 5},
{"block_id": "code_block", "count": 2},
],
"errors": [],
"metadata": {
"trigger": "manual",
"user_tier": "premium",
"environment": "production",
},
},
"data_index": "agent_123_exec_456",
}
response = client.post("/log_raw_analytics", json=request_data)
assert response.status_code == 200
configured_snapshot.assert_match(
json.dumps(
{"analytics_id": response.json(), "logged_data": request_data["data"]},
indent=2,
sort_keys=True,
),
"analytics_log_analytics_complex_data",
)
@pytest.mark.parametrize(
"invalid_data,expected_error",
[
({}, "Field required"),
({"type": "test"}, "Field required"),
(
{"type": "test", "data": "not_a_dict", "data_index": "test"},
"Input should be a valid dictionary",
),
({"type": "test", "data": {"key": "value"}}, "Field required"),
],
ids=[
"empty_request",
"missing_data_and_data_index",
"invalid_data_type",
"missing_data_index",
],
)
def test_log_raw_analytics_validation_errors(
invalid_data: dict,
expected_error: str,
) -> None:
"""Test validation errors for invalid analytics requests."""
response = client.post("/log_raw_analytics", json=invalid_data)
assert response.status_code == 422
error_detail = response.json()
assert "detail" in error_detail, f"Missing 'detail' in error: {error_detail}"
error_text = json.dumps(error_detail)
assert (
expected_error in error_text
), f"Expected '{expected_error}' in error response: {error_text}"
def test_log_raw_analytics_service_error(
mocker: pytest_mock.MockFixture,
test_user_id: str,
) -> None:
"""Test error handling when analytics service fails."""
mocker.patch(
"backend.data.analytics.log_raw_analytics",
new_callable=AsyncMock,
side_effect=Exception("Analytics DB unreachable"),
)
request_data = {
"type": "test_event",
"data": {"key": "value"},
"data_index": "test_index",
}
response = client.post("/log_raw_analytics", json=request_data)
assert response.status_code == 500
error_detail = response.json()["detail"]
assert "Analytics DB unreachable" in error_detail["message"]
assert "hint" in error_detail

View File

@@ -1,215 +0,0 @@
"""Database operations for chat sessions."""
import logging
from datetime import UTC, datetime
from typing import Any, cast
from prisma.models import ChatMessage as PrismaChatMessage
from prisma.models import ChatSession as PrismaChatSession
from prisma.types import (
ChatMessageCreateInput,
ChatSessionCreateInput,
ChatSessionUpdateInput,
)
from backend.util.json import SafeJson
logger = logging.getLogger(__name__)
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": True},
)
if session and session.Messages:
# Sort messages by sequence in Python since Prisma doesn't support order_by in include
session.Messages.sort(key=lambda m: m.sequence)
return session
async def create_chat_session(
session_id: str,
user_id: str | None,
) -> PrismaChatSession:
"""Create a new chat session in the database."""
data = ChatSessionCreateInput(
id=session_id,
userId=user_id,
credentials=SafeJson({}),
successfulAgentRuns=SafeJson({}),
successfulAgentSchedules=SafeJson({}),
)
return await PrismaChatSession.prisma().create(
data=data,
include={"Messages": True},
)
async def update_chat_session(
session_id: str,
credentials: dict[str, Any] | None = None,
successful_agent_runs: dict[str, Any] | None = None,
successful_agent_schedules: dict[str, Any] | None = None,
total_prompt_tokens: int | None = None,
total_completion_tokens: int | None = None,
title: str | None = None,
) -> PrismaChatSession | None:
"""Update a chat session's metadata."""
data: ChatSessionUpdateInput = {"updatedAt": datetime.now(UTC)}
if credentials is not None:
data["credentials"] = SafeJson(credentials)
if successful_agent_runs is not None:
data["successfulAgentRuns"] = SafeJson(successful_agent_runs)
if successful_agent_schedules is not None:
data["successfulAgentSchedules"] = SafeJson(successful_agent_schedules)
if total_prompt_tokens is not None:
data["totalPromptTokens"] = total_prompt_tokens
if total_completion_tokens is not None:
data["totalCompletionTokens"] = total_completion_tokens
if title is not None:
data["title"] = title
session = await PrismaChatSession.prisma().update(
where={"id": session_id},
data=data,
include={"Messages": True},
)
if session and session.Messages:
session.Messages.sort(key=lambda m: m.sequence)
return session
async def add_chat_message(
session_id: str,
role: str,
sequence: int,
content: str | None = None,
name: str | None = None,
tool_call_id: str | None = None,
refusal: str | None = None,
tool_calls: list[dict[str, Any]] | None = None,
function_call: dict[str, Any] | None = None,
) -> PrismaChatMessage:
"""Add a message to a chat session."""
# Build the input dict dynamically - only include optional fields when they
# have values, as Prisma TypedDict validation fails when optional fields
# are explicitly set to None
data: dict[str, Any] = {
"Session": {"connect": {"id": session_id}},
"role": role,
"sequence": sequence,
}
# Add optional string fields
if content is not None:
data["content"] = content
if name is not None:
data["name"] = name
if tool_call_id is not None:
data["toolCallId"] = tool_call_id
if refusal is not None:
data["refusal"] = refusal
# Add optional JSON fields only when they have values
if tool_calls is not None:
data["toolCalls"] = SafeJson(tool_calls)
if function_call is not None:
data["functionCall"] = SafeJson(function_call)
# Update session's updatedAt timestamp
await PrismaChatSession.prisma().update(
where={"id": session_id},
data={"updatedAt": datetime.now(UTC)},
)
return await PrismaChatMessage.prisma().create(
data=cast(ChatMessageCreateInput, data)
)
async def add_chat_messages_batch(
session_id: str,
messages: list[dict[str, Any]],
start_sequence: int,
) -> list[PrismaChatMessage]:
"""Add multiple messages to a chat session in a batch."""
if not messages:
return []
created_messages = []
for i, msg in enumerate(messages):
# Build the input dict dynamically - only include optional JSON fields
# when they have values, as Prisma TypedDict validation fails when
# optional fields are explicitly set to None
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 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"])
created = await PrismaChatMessage.prisma().create(
data=cast(ChatMessageCreateInput, data)
)
created_messages.append(created)
# Update session's updatedAt timestamp
await PrismaChatSession.prisma().update(
where={"id": session_id},
data={"updatedAt": datetime.now(UTC)},
)
return created_messages
async def get_user_chat_sessions(
user_id: str,
limit: int = 50,
offset: int = 0,
) -> list[PrismaChatSession]:
"""Get chat sessions for a user, ordered by most recent."""
return await PrismaChatSession.prisma().find_many(
where={"userId": user_id},
order={"updatedAt": "desc"},
take=limit,
skip=offset,
)
async def get_user_session_count(user_id: str) -> int:
"""Get the total number of chat sessions for a user."""
return await PrismaChatSession.prisma().count(where={"userId": user_id})
async def delete_chat_session(session_id: str) -> bool:
"""Delete a chat session and all its messages."""
try:
await PrismaChatSession.prisma().delete(where={"id": session_id})
return True
except Exception as e:
logger.error(f"Failed to delete chat session {session_id}: {e}")
return False
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

View File

@@ -1,473 +0,0 @@
import logging
import uuid
from datetime import UTC, datetime
from openai.types.chat import (
ChatCompletionAssistantMessageParam,
ChatCompletionDeveloperMessageParam,
ChatCompletionFunctionMessageParam,
ChatCompletionMessageParam,
ChatCompletionSystemMessageParam,
ChatCompletionToolMessageParam,
ChatCompletionUserMessageParam,
)
from openai.types.chat.chat_completion_assistant_message_param import FunctionCall
from openai.types.chat.chat_completion_message_tool_call_param import (
ChatCompletionMessageToolCallParam,
Function,
)
from prisma.models import ChatMessage as PrismaChatMessage
from prisma.models import ChatSession as PrismaChatSession
from pydantic import BaseModel
from backend.data.redis_client import get_redis_async
from backend.util import json
from backend.util.exceptions import RedisError
from . import db as chat_db
from .config import ChatConfig
logger = logging.getLogger(__name__)
config = ChatConfig()
class ChatMessage(BaseModel):
role: str
content: str | None = None
name: str | None = None
tool_call_id: str | None = None
refusal: str | None = None
tool_calls: list[dict] | None = None
function_call: dict | None = None
class Usage(BaseModel):
prompt_tokens: int
completion_tokens: int
total_tokens: int
class ChatSession(BaseModel):
session_id: str
user_id: str | None
title: str | None = None
messages: list[ChatMessage]
usage: list[Usage]
credentials: dict[str, dict] = {} # Map of provider -> credential metadata
started_at: datetime
updated_at: datetime
successful_agent_runs: dict[str, int] = {}
successful_agent_schedules: dict[str, int] = {}
@staticmethod
def new(user_id: str | None) -> "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_prisma(
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:
tool_calls = None
if msg.toolCalls:
tool_calls = (
json.loads(msg.toolCalls)
if isinstance(msg.toolCalls, str)
else msg.toolCalls
)
function_call = None
if msg.functionCall:
function_call = (
json.loads(msg.functionCall)
if isinstance(msg.functionCall, str)
else msg.functionCall
)
messages.append(
ChatMessage(
role=msg.role,
content=msg.content,
name=msg.name,
tool_call_id=msg.toolCallId,
refusal=msg.refusal,
tool_calls=tool_calls,
function_call=function_call,
)
)
# Parse JSON fields from Prisma
credentials = (
json.loads(prisma_session.credentials)
if isinstance(prisma_session.credentials, str)
else prisma_session.credentials or {}
)
successful_agent_runs = (
json.loads(prisma_session.successfulAgentRuns)
if isinstance(prisma_session.successfulAgentRuns, str)
else prisma_session.successfulAgentRuns or {}
)
successful_agent_schedules = (
json.loads(prisma_session.successfulAgentSchedules)
if isinstance(prisma_session.successfulAgentSchedules, str)
else prisma_session.successfulAgentSchedules or {}
)
# Calculate usage from token counts
usage = []
if prisma_session.totalPromptTokens or prisma_session.totalCompletionTokens:
usage.append(
Usage(
prompt_tokens=prisma_session.totalPromptTokens or 0,
completion_tokens=prisma_session.totalCompletionTokens or 0,
total_tokens=(prisma_session.totalPromptTokens or 0)
+ (prisma_session.totalCompletionTokens or 0),
)
)
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,
updated_at=prisma_session.updatedAt,
successful_agent_runs=successful_agent_runs,
successful_agent_schedules=successful_agent_schedules,
)
def to_openai_messages(self) -> list[ChatCompletionMessageParam]:
messages = []
for message in self.messages:
if message.role == "developer":
m = ChatCompletionDeveloperMessageParam(
role="developer",
content=message.content or "",
)
if message.name:
m["name"] = message.name
messages.append(m)
elif message.role == "system":
m = ChatCompletionSystemMessageParam(
role="system",
content=message.content or "",
)
if message.name:
m["name"] = message.name
messages.append(m)
elif message.role == "user":
m = ChatCompletionUserMessageParam(
role="user",
content=message.content or "",
)
if message.name:
m["name"] = message.name
messages.append(m)
elif message.role == "assistant":
m = ChatCompletionAssistantMessageParam(
role="assistant",
content=message.content or "",
)
if message.function_call:
m["function_call"] = FunctionCall(
arguments=message.function_call["arguments"],
name=message.function_call["name"],
)
if message.refusal:
m["refusal"] = message.refusal
if message.tool_calls:
t: list[ChatCompletionMessageToolCallParam] = []
for tool_call in message.tool_calls:
# Tool calls are stored with nested structure: {id, type, function: {name, arguments}}
function_data = tool_call.get("function", {})
# Skip tool calls that are missing required fields
if "id" not in tool_call or "name" not in function_data:
logger.warning(
f"Skipping invalid tool call: missing required fields. "
f"Got: {tool_call.keys()}, function keys: {function_data.keys()}"
)
continue
# Arguments are stored as a JSON string
arguments_str = function_data.get("arguments", "{}")
t.append(
ChatCompletionMessageToolCallParam(
id=tool_call["id"],
type="function",
function=Function(
arguments=arguments_str,
name=function_data["name"],
),
)
)
m["tool_calls"] = t
if message.name:
m["name"] = message.name
messages.append(m)
elif message.role == "tool":
messages.append(
ChatCompletionToolMessageParam(
role="tool",
content=message.content or "",
tool_call_id=message.tool_call_id or "",
)
)
elif message.role == "function":
messages.append(
ChatCompletionFunctionMessageParam(
role="function",
content=message.content,
name=message.name or "",
)
)
return messages
async def _get_session_from_cache(session_id: str) -> ChatSession | None:
"""Get a chat session from Redis cache."""
redis_key = f"chat:session:{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 _cache_session(session: ChatSession) -> None:
"""Cache a chat session in Redis."""
redis_key = f"chat:session:{session.session_id}"
async_redis = await get_redis_async()
await async_redis.setex(redis_key, config.session_ttl, session.model_dump_json())
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_prisma(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,
) -> ChatSession | None:
"""Get a chat session by ID.
Checks Redis cache first, falls back to database if not found.
Caches database results back to Redis.
"""
# Try cache first
try:
session = await _get_session_from_cache(session_id)
if session:
# Verify user ownership
if session.user_id is not None and session.user_id != user_id:
logger.warning(
f"Session {session_id} user id mismatch: {session.user_id} != {user_id}"
)
return None
return session
except RedisError:
logger.warning(f"Cache error for session {session_id}, trying database")
except Exception as e:
logger.warning(f"Unexpected cache error for session {session_id}: {e}")
# Fall back to database
logger.info(f"Session {session_id} not in cache, checking database")
session = await _get_session_from_db(session_id)
if session is None:
logger.warning(f"Session {session_id} not found in cache or database")
return None
# Verify user ownership
if session.user_id is not None and session.user_id != user_id:
logger.warning(
f"Session {session_id} user id mismatch: {session.user_id} != {user_id}"
)
return None
# Cache the session from DB
try:
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}")
return session
async def upsert_chat_session(
session: ChatSession,
) -> ChatSession:
"""Update a chat session in both cache and database."""
# Get existing message count from DB for incremental saves
existing_message_count = await chat_db.get_chat_session_message_count(
session.session_id
)
# Save to database
try:
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}")
# Continue to cache even if DB fails
# Save to cache
try:
await _cache_session(session)
except Exception as e:
raise RedisError(
f"Failed to persist chat session {session.session_id} to Redis: {e}"
) from e
return session
async def create_chat_session(user_id: str | None) -> ChatSession:
"""Create a new chat session and persist it."""
session = ChatSession.new(user_id)
# Create in database first
try:
await chat_db.create_chat_session(
session_id=session.session_id,
user_id=user_id,
)
except Exception as e:
logger.error(f"Failed to create session in database: {e}")
# Continue even if DB fails - cache will still work
# Cache the session
try:
await _cache_session(session)
except Exception as e:
logger.warning(f"Failed to cache new session: {e}")
return session
async def get_user_sessions(
user_id: str,
limit: int = 50,
offset: int = 0,
) -> list[ChatSession]:
"""Get all chat sessions for a user from the database."""
prisma_sessions = await chat_db.get_user_chat_sessions(user_id, limit, offset)
sessions = []
for prisma_session in prisma_sessions:
# Convert without messages for listing (lighter weight)
sessions.append(ChatSession.from_prisma(prisma_session, None))
return sessions
async def delete_chat_session(session_id: str) -> bool:
"""Delete a chat session from both cache and database."""
# Delete from cache
try:
redis_key = f"chat:session:{session_id}"
async_redis = await get_redis_async()
await async_redis.delete(redis_key)
except Exception as e:
logger.warning(f"Failed to delete session {session_id} from cache: {e}")
# Delete from database
return await chat_db.delete_chat_session(session_id)

View File

@@ -1,117 +0,0 @@
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():
s = ChatSession.new(user_id=None)
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():
s = ChatSession.new(user_id="abc123")
s.messages = messages
s = await upsert_chat_session(s)
s2 = await get_chat_session(s.session_id, None)
assert s2 is None
@pytest.mark.asyncio(loop_scope="session")
async def test_chatsession_db_storage():
"""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=None)
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)

View File

@@ -1,192 +0,0 @@
You are Otto, an AI Co-Pilot and Forward Deployed Engineer for AutoGPT, an AI Business Automation tool. Your mission is to help users quickly find, create, and set up AutoGPT agents to solve their business problems.
Here are the functions available to you:
<functions>
**Understanding & Discovery:**
1. **add_understanding** - Save information about the user's business context (use this as you learn about them)
2. **find_agent** - Search the marketplace for pre-built agents that solve the user's problem
3. **find_library_agent** - Search the user's personal library of saved agents
4. **find_block** - Search for individual blocks (building components for agents)
5. **search_platform_docs** - Search AutoGPT documentation for help
**Agent Creation & Editing:**
6. **create_agent** - Create a new custom agent from scratch based on user requirements
7. **edit_agent** - Modify an existing agent (add/remove blocks, change configuration)
**Execution & Output:**
8. **run_agent** - Run or schedule an agent (automatically handles setup)
9. **run_block** - Run a single block directly without creating an agent
10. **agent_output** - Get the output/results from a running or completed agent execution
</functions>
## ALWAYS GET THE USER'S NAME
**This is critical:** If you don't know the user's name, ask for it in your first response. Use a friendly, natural approach:
- "Hi! I'm Otto. What's your name?"
- "Hey there! Before we dive in, what should I call you?"
Once you have their name, immediately save it with `add_understanding(user_name="...")` and use it throughout the conversation.
## BUILDING USER UNDERSTANDING
**If no User Business Context is provided below**, gather information naturally during conversation - don't interrogate them.
**Key information to gather (in priority order):**
1. Their name (ALWAYS first if unknown)
2. Their job title and role
3. Their business/company and industry
4. Pain points and what they want to automate
5. Tools they currently use
**How to gather this information:**
- Ask naturally as part of helping them (e.g., "What's your role?" or "What industry are you in?")
- When they share information, immediately save it using `add_understanding`
- Don't ask all questions at once - spread them across the conversation
- Prioritize understanding their immediate problem first
**Example:**
```
User: "I need help automating my social media"
Otto: I can help with that! I'm Otto - what's your name?
User: "I'm Sarah"
Otto: [calls add_understanding with user_name="Sarah"]
Nice to meet you, Sarah! What's your role - are you a social media manager or business owner?
User: "I'm the marketing director at a fintech startup"
Otto: [calls add_understanding with job_title="Marketing Director", industry="fintech", business_size="startup"]
Great! Let me find social media automation agents for you.
[calls find_agent with query="social media automation marketing"]
```
## WHEN TO USE WHICH TOOL
**Finding existing agents:**
- `find_agent` - Search the marketplace for pre-built agents others have created
- `find_library_agent` - Search agents the user has already saved to their library
**Creating/editing agents:**
- `create_agent` - When user wants a custom agent that doesn't exist, or has specific requirements
- `edit_agent` - When user wants to modify an existing agent (change inputs, add blocks, etc.)
**Running agents:**
- `run_agent` - To execute an agent (handles credentials and inputs automatically)
- `agent_output` - To check the results of a running or completed agent execution
**Direct execution:**
- `run_block` - Run a single block directly without needing a full agent
## HOW run_agent WORKS
The `run_agent` tool automatically handles the entire setup flow:
1. **First call** (no inputs) → Returns available inputs so user can decide what values to use
2. **Credentials check** → If missing, UI automatically prompts user to add them (you don't need to mention this)
3. **Execution** → Runs when you provide `inputs` OR set `use_defaults=true`
Parameters:
- `username_agent_slug` (required): Agent identifier like "creator/agent-name"
- `inputs`: Object with input values for the agent
- `use_defaults`: Set to `true` to run with default values (only after user confirms)
- `schedule_name` + `cron`: For scheduled execution
## HOW create_agent WORKS
Use `create_agent` when the user wants to build a custom automation:
- Describe what the agent should do
- The tool will create the agent structure with appropriate blocks
- Returns the agent ID for further editing or running
## HOW agent_output WORKS
Use `agent_output` to get results from agent executions:
- Pass the execution_id from a run_agent response
- Returns the current status and any outputs produced
- Useful for checking if an agent has completed and what it produced
## WORKFLOW
1. **Get their name** - If unknown, ask for it first
2. **Understand context** - Ask 1-2 questions about their problem while helping
3. **Find or create** - Use find_agent for existing solutions, create_agent for custom needs
4. **Set up and run** - Use run_agent to execute, agent_output to get results
## YOUR APPROACH
**Step 1: Greet and Identify**
- If you don't know their name, ask for it
- Be friendly and conversational
**Step 2: Understand the Problem**
- Ask maximum 1-2 targeted questions
- Focus on: What business problem are they solving?
- If they want to create/edit an agent, understand what it should do
**Step 3: Find or Create**
- For existing solutions: Use `find_agent` with relevant keywords
- For custom needs: Use `create_agent` with their requirements
- For modifications: Use `edit_agent` on an existing agent
**Step 4: Execute**
- Call `run_agent` without inputs first to see what's available
- Ask user what values they want or if defaults are okay
- Call `run_agent` again with inputs or `use_defaults=true`
- Use `agent_output` to check results when needed
## USING add_understanding
Call `add_understanding` whenever you learn something about the user:
**User info:** `user_name`, `job_title`
**Business:** `business_name`, `industry`, `business_size` (1-10, 11-50, 51-200, 201-1000, 1000+), `user_role` (decision maker, implementer, end user)
**Processes:** `key_workflows` (array), `daily_activities` (array)
**Pain points:** `pain_points` (array), `bottlenecks` (array), `manual_tasks` (array), `automation_goals` (array)
**Tools:** `current_software` (array), `existing_automation` (array)
**Other:** `additional_notes`
Example: `add_understanding(user_name="Sarah", job_title="Marketing Director", industry="fintech")`
## KEY RULES
**What You DON'T Do:**
- Don't help with login (frontend handles this)
- Don't mention or explain credentials to the user (frontend handles this automatically)
- Don't run agents without first showing available inputs to the user
- Don't use `use_defaults=true` without user explicitly confirming
- Don't write responses longer than 3 sentences
- Don't interrogate users with many questions - gather info naturally
**What You DO:**
- ALWAYS ask for user's name if you don't have it
- Save user information with `add_understanding` as you learn it
- Use their name when addressing them
- Always call run_agent first without inputs to see what's available
- Ask user what values they want OR if they want to use defaults
- Keep all responses to maximum 3 sentences
- Include the agent link in your response after successful execution
**Error Handling:**
- Authentication needed → "Please sign in via the interface"
- Credentials missing → The UI handles this automatically. Focus on asking the user about input values instead.
## RESPONSE STRUCTURE
Before responding, wrap your analysis in <thinking> tags to systematically plan your approach:
- Check if you know the user's name - if not, ask for it
- Check if you have user context - if not, plan to gather some naturally
- Extract the key business problem or request from the user's message
- Determine what function call (if any) you need to make next
- Plan your response to stay under the 3-sentence maximum
Example interaction:
```
User: "Hi, I want to build an agent that monitors my competitors"
Otto: <thinking>I don't know this user's name. I should ask for it while acknowledging their request.</thinking>
Hi! I'm Otto and I'd love to help you build a competitor monitoring agent. What's your name?
User: "I'm Mike"
Otto: [calls add_understanding with user_name="Mike"]
<thinking>Now I know Mike wants competitor monitoring. I should search for existing agents first.</thinking>
Great to meet you, Mike! Let me search for competitor monitoring agents.
[calls find_agent with query="competitor monitoring analysis"]
```
KEEP ANSWERS TO 3 SENTENCES

View File

@@ -1,155 +0,0 @@
You are Otto, an AI Co-Pilot helping new users get started with AutoGPT, an AI Business Automation platform. Your mission is to welcome them, learn about their needs, and help them run their first successful agent.
Here are the functions available to you:
<functions>
**Understanding & Discovery:**
1. **add_understanding** - Save information about the user's business context (use this as you learn about them)
2. **find_agent** - Search the marketplace for pre-built agents that solve the user's problem
3. **find_library_agent** - Search the user's personal library of saved agents
4. **find_block** - Search for individual blocks (building components for agents)
5. **search_platform_docs** - Search AutoGPT documentation for help
**Agent Creation & Editing:**
6. **create_agent** - Create a new custom agent from scratch based on user requirements
7. **edit_agent** - Modify an existing agent (add/remove blocks, change configuration)
**Execution & Output:**
8. **run_agent** - Run or schedule an agent (automatically handles setup)
9. **run_block** - Run a single block directly without creating an agent
10. **agent_output** - Get the output/results from a running or completed agent execution
</functions>
## YOUR ONBOARDING MISSION
You are guiding a new user through their first experience with AutoGPT. Your goal is to:
1. Welcome them warmly and get their name
2. Learn about them and their business
3. Find or create an agent that solves a real problem for them
4. Get that agent running successfully
5. Celebrate their success and point them to next steps
## PHASE 1: WELCOME & INTRODUCTION
**Start every conversation by:**
- Giving a warm, friendly greeting
- Introducing yourself as Otto, their AI assistant
- Asking for their name immediately
**Example opening:**
```
Hi! I'm Otto, your AI assistant. Welcome to AutoGPT! I'm here to help you set up your first automation. What's your name?
```
Once you have their name, save it immediately with `add_understanding(user_name="...")` and use it throughout.
## PHASE 2: DISCOVERY
**After getting their name, learn about them:**
- What's their role/job title?
- What industry/business are they in?
- What's one thing they'd love to automate?
**Keep it conversational - don't interrogate. Example:**
```
Nice to meet you, Sarah! What do you do for work, and what's one task you wish you could automate?
```
Save everything you learn with `add_understanding`.
## PHASE 3: FIND OR CREATE AN AGENT
**Once you understand their need:**
- Search for existing agents with `find_agent`
- Present the best match and explain how it helps them
- If nothing fits, offer to create a custom agent with `create_agent`
**Be enthusiastic about the solution:**
```
I found a great agent for you! The "Social Media Scheduler" can automatically post to your accounts on a schedule. Want to try it?
```
## PHASE 4: SETUP & RUN
**Guide them through running the agent:**
1. Call `run_agent` without inputs first to see what's needed
2. Explain each input in simple terms
3. Ask what values they want to use
4. Run the agent with their inputs or defaults
**Don't mention credentials** - the UI handles that automatically.
## PHASE 5: CELEBRATE & HANDOFF
**After successful execution:**
- Congratulate them on their first automation!
- Tell them where to find this agent (their Library)
- Mention they can explore more agents in the Marketplace
- Offer to help with anything else
**Example:**
```
You did it! Your first agent is running. You can find it anytime in your Library. Ready to explore more automations?
```
## KEY RULES
**What You DON'T Do:**
- Don't help with login (frontend handles this)
- Don't mention credentials (UI handles automatically)
- Don't run agents without showing inputs first
- Don't use `use_defaults=true` without explicit confirmation
- Don't write responses longer than 3 sentences
- Don't overwhelm with too many questions at once
**What You DO:**
- ALWAYS get the user's name first
- Be warm, encouraging, and celebratory
- Save info with `add_understanding` as you learn it
- Use their name when addressing them
- Keep responses to maximum 3 sentences
- Make them feel successful at each step
## USING add_understanding
Save information as you learn it:
**User info:** `user_name`, `job_title`
**Business:** `business_name`, `industry`, `business_size`, `user_role`
**Pain points:** `pain_points`, `manual_tasks`, `automation_goals`
**Tools:** `current_software`
Example: `add_understanding(user_name="Sarah", job_title="Marketing Manager", automation_goals=["social media scheduling"])`
## HOW run_agent WORKS
1. **First call** (no inputs) → Shows available inputs
2. **Credentials** → UI handles automatically (don't mention)
3. **Execution** → Run with `inputs={...}` or `use_defaults=true`
## RESPONSE STRUCTURE
Before responding, plan your approach in <thinking> tags:
- What phase am I in? (Welcome/Discovery/Find/Setup/Celebrate)
- Do I know their name? If not, ask for it
- What's the next step to move them forward?
- Keep response under 3 sentences
**Example flow:**
```
User: "Hi"
Otto: <thinking>Phase 1 - I need to welcome them and get their name.</thinking>
Hi! I'm Otto, welcome to AutoGPT! I'm here to help you set up your first automation - what's your name?
User: "I'm Alex"
Otto: [calls add_understanding with user_name="Alex"]
<thinking>Got their name. Phase 2 - learn about them.</thinking>
Great to meet you, Alex! What do you do for work, and what's one task you'd love to automate?
User: "I run an e-commerce store and spend hours on customer support emails"
Otto: [calls add_understanding with industry="e-commerce", pain_points=["customer support emails"]]
<thinking>Phase 3 - search for agents.</thinking>
[calls find_agent with query="customer support email automation"]
```
KEEP ANSWERS TO 3 SENTENCES - Be warm, helpful, and focused on their success!

View File

@@ -1,472 +0,0 @@
"""Chat API routes for chat session management and streaming via SSE."""
import logging
from collections.abc import AsyncGenerator
from typing import Annotated
from autogpt_libs import auth
from fastapi import APIRouter, Depends, Query, Security
from fastapi.responses import StreamingResponse
from pydantic import BaseModel
from backend.util.exceptions import NotFoundError
from . import service as chat_service
from .config import ChatConfig
config = ChatConfig()
logger = logging.getLogger(__name__)
router = APIRouter(
tags=["chat"],
)
# ========== Request/Response Models ==========
class StreamChatRequest(BaseModel):
"""Request model for streaming chat with optional context."""
message: str
is_user_message: bool = True
context: dict[str, str] | None = None # {url: str, content: str}
class CreateSessionResponse(BaseModel):
"""Response model containing information on a newly created chat session."""
id: str
created_at: str
user_id: str | None
class SessionDetailResponse(BaseModel):
"""Response model providing complete details for a chat session, including messages."""
id: str
created_at: str
updated_at: str
user_id: str | None
messages: list[dict]
class SessionSummaryResponse(BaseModel):
"""Response model for a session summary (without messages)."""
id: str
created_at: str
updated_at: str
title: str | None = None
class ListSessionsResponse(BaseModel):
"""Response model for listing chat sessions."""
sessions: list[SessionSummaryResponse]
total: int
# ========== Routes ==========
@router.get(
"/sessions",
dependencies=[Security(auth.requires_user)],
)
async def list_sessions(
user_id: Annotated[str, Security(auth.get_user_id)],
limit: int = Query(default=50, ge=1, le=100),
offset: int = Query(default=0, ge=0),
) -> ListSessionsResponse:
"""
List chat sessions for the authenticated user.
Returns a paginated list of chat sessions belonging to the current user,
ordered by most recently updated.
Args:
user_id: The authenticated user's ID.
limit: Maximum number of sessions to return (1-100).
offset: Number of sessions to skip for pagination.
Returns:
ListSessionsResponse: List of session summaries and total count.
"""
sessions = await chat_service.get_user_sessions(user_id, limit, offset)
return ListSessionsResponse(
sessions=[
SessionSummaryResponse(
id=session.session_id,
created_at=session.started_at.isoformat(),
updated_at=session.updated_at.isoformat(),
title=None, # TODO: Add title support
)
for session in sessions
],
total=len(sessions),
)
@router.post(
"/sessions",
)
async def create_session(
user_id: Annotated[str | None, Depends(auth.get_user_id)],
) -> CreateSessionResponse:
"""
Create a new chat session.
Initiates a new chat session for either an authenticated or anonymous user.
Args:
user_id: The optional authenticated user ID parsed from the JWT. If missing, creates an anonymous session.
Returns:
CreateSessionResponse: Details of the created session.
"""
logger.info(
f"Creating session with user_id: "
f"...{user_id[-8:] if user_id and len(user_id) > 8 else '<redacted>'}"
)
session = await chat_service.create_chat_session(user_id)
return CreateSessionResponse(
id=session.session_id,
created_at=session.started_at.isoformat(),
user_id=session.user_id or None,
)
@router.get(
"/sessions/{session_id}",
)
async def get_session(
session_id: str,
user_id: Annotated[str | None, Depends(auth.get_user_id)],
) -> SessionDetailResponse:
"""
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.
Args:
session_id: The unique identifier for the desired chat session.
user_id: The optional authenticated user ID, or None for anonymous access.
Returns:
SessionDetailResponse: Details for the requested session; raises NotFoundError if not found.
"""
session = await chat_service.get_session(session_id, user_id)
if not session:
raise NotFoundError(f"Session {session_id} not found")
messages = [message.model_dump() for message in session.messages]
logger.info(
f"Returning session {session_id}: "
f"message_count={len(messages)}, "
f"roles={[m.get('role') for m in messages]}"
)
return SessionDetailResponse(
id=session.session_id,
created_at=session.started_at.isoformat(),
updated_at=session.updated_at.isoformat(),
user_id=session.user_id or None,
messages=messages,
)
@router.post(
"/sessions/{session_id}/stream",
)
async def stream_chat_post(
session_id: str,
request: StreamChatRequest,
user_id: str | None = Depends(auth.get_user_id),
):
"""
Stream chat responses for a session (POST with context support).
Streams the AI/completion responses in real time over Server-Sent Events (SSE), including:
- Text fragments as they are generated
- Tool call UI elements (if invoked)
- Tool execution results
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.
"""
# Validate session exists before starting the stream
# This prevents errors after the response has already started
session = await chat_service.get_session(session_id, user_id)
if not session:
raise NotFoundError(f"Session {session_id} not found. ")
if session.user_id is None and user_id is not None:
session = await chat_service.assign_user_to_session(session_id, user_id)
async def event_generator() -> AsyncGenerator[str, None]:
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,
):
yield chunk.to_sse()
return StreamingResponse(
event_generator(),
media_type="text/event-stream",
headers={
"Cache-Control": "no-cache",
"Connection": "keep-alive",
"X-Accel-Buffering": "no", # Disable nginx buffering
},
)
@router.get(
"/sessions/{session_id}/stream",
)
async def stream_chat_get(
session_id: str,
message: Annotated[str, Query(min_length=1, max_length=10000)],
user_id: str | None = Depends(auth.get_user_id),
is_user_message: bool = Query(default=True),
):
"""
Stream chat responses for a session (GET - legacy endpoint).
Streams the AI/completion responses in real time over Server-Sent Events (SSE), including:
- Text fragments as they are generated
- Tool call UI elements (if invoked)
- Tool execution results
Args:
session_id: The chat session identifier to associate with the streamed messages.
message: The user's new message to process.
user_id: Optional authenticated user ID.
is_user_message: Whether the message is a user message.
Returns:
StreamingResponse: SSE-formatted response chunks.
"""
# Validate session exists before starting the stream
# This prevents errors after the response has already started
session = await chat_service.get_session(session_id, user_id)
if not session:
raise NotFoundError(f"Session {session_id} not found. ")
if session.user_id is None and user_id is not None:
session = await chat_service.assign_user_to_session(session_id, user_id)
async def event_generator() -> AsyncGenerator[str, None]:
async for chunk in chat_service.stream_chat_completion(
session_id,
message,
is_user_message=is_user_message,
user_id=user_id,
session=session, # Pass pre-fetched session to avoid double-fetch
):
yield chunk.to_sse()
return StreamingResponse(
event_generator(),
media_type="text/event-stream",
headers={
"Cache-Control": "no-cache",
"Connection": "keep-alive",
"X-Accel-Buffering": "no", # Disable nginx buffering
},
)
@router.patch(
"/sessions/{session_id}/assign-user",
dependencies=[Security(auth.requires_user)],
status_code=200,
)
async def session_assign_user(
session_id: str,
user_id: Annotated[str, Security(auth.get_user_id)],
) -> dict:
"""
Assign an authenticated user to a chat session.
Used (typically post-login) to claim an existing anonymous session as the current authenticated user.
Args:
session_id: The identifier for the (previously anonymous) session.
user_id: The authenticated user's ID to associate with the session.
Returns:
dict: Status of the assignment.
"""
await chat_service.assign_user_to_session(session_id, user_id)
return {"status": "ok"}
# ========== Onboarding Routes ==========
# These routes use a specialized onboarding system prompt
@router.post(
"/onboarding/sessions",
)
async def create_onboarding_session(
user_id: Annotated[str | None, Depends(auth.get_user_id)],
) -> CreateSessionResponse:
"""
Create a new onboarding chat session.
Initiates a new chat session specifically for user onboarding,
using a specialized prompt that guides users through their first
experience with AutoGPT.
Args:
user_id: The optional authenticated user ID parsed from the JWT.
Returns:
CreateSessionResponse: Details of the created onboarding session.
"""
logger.info(
f"Creating onboarding session with user_id: "
f"...{user_id[-8:] if user_id and len(user_id) > 8 else '<redacted>'}"
)
session = await chat_service.create_chat_session(user_id)
return CreateSessionResponse(
id=session.session_id,
created_at=session.started_at.isoformat(),
user_id=session.user_id or None,
)
@router.get(
"/onboarding/sessions/{session_id}",
)
async def get_onboarding_session(
session_id: str,
user_id: Annotated[str | None, Depends(auth.get_user_id)],
) -> SessionDetailResponse:
"""
Retrieve the details of an onboarding chat session.
Args:
session_id: The unique identifier for the onboarding session.
user_id: The optional authenticated user ID.
Returns:
SessionDetailResponse: Details for the requested session.
"""
session = await chat_service.get_session(session_id, user_id)
if not session:
raise NotFoundError(f"Session {session_id} not found")
messages = [message.model_dump() for message in session.messages]
logger.info(
f"Returning onboarding session {session_id}: "
f"message_count={len(messages)}, "
f"roles={[m.get('role') for m in messages]}"
)
return SessionDetailResponse(
id=session.session_id,
created_at=session.started_at.isoformat(),
updated_at=session.updated_at.isoformat(),
user_id=session.user_id or None,
messages=messages,
)
@router.post(
"/onboarding/sessions/{session_id}/stream",
)
async def stream_onboarding_chat(
session_id: str,
request: StreamChatRequest,
user_id: str | None = Depends(auth.get_user_id),
):
"""
Stream onboarding chat responses for a session.
Uses the specialized onboarding system prompt to guide new users
through their first experience with AutoGPT. Streams AI responses
in real time over Server-Sent Events (SSE).
Args:
session_id: The onboarding session identifier.
request: Request body containing message and optional context.
user_id: Optional authenticated user ID.
Returns:
StreamingResponse: SSE-formatted response chunks.
"""
session = await chat_service.get_session(session_id, user_id)
if not session:
raise NotFoundError(f"Session {session_id} not found.")
if session.user_id is None and user_id is not None:
session = await chat_service.assign_user_to_session(session_id, user_id)
async def event_generator() -> AsyncGenerator[str, None]:
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,
context=request.context,
prompt_type="onboarding", # Use onboarding system prompt
):
yield chunk.to_sse()
return StreamingResponse(
event_generator(),
media_type="text/event-stream",
headers={
"Cache-Control": "no-cache",
"Connection": "keep-alive",
"X-Accel-Buffering": "no",
},
)
# ========== Health Check ==========
@router.get("/health", status_code=200)
async def health_check() -> dict:
"""
Health check endpoint for the chat service.
Performs a full cycle test of session creation, assignment, and retrieval. Should always return healthy
if the service and data layer are operational.
Returns:
dict: A status dictionary indicating health, service name, and API version.
"""
session = await chat_service.create_chat_session(None)
await chat_service.assign_user_to_session(session.session_id, "test_user")
await chat_service.get_session(session.session_id, "test_user")
return {
"status": "healthy",
"service": "chat",
"version": "0.1.0",
}

View File

@@ -1,202 +0,0 @@
"""Tool for capturing user business understanding incrementally."""
import logging
from typing import Any
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
logger = logging.getLogger(__name__)
class AddUnderstandingTool(BaseTool):
"""Tool for capturing user's business understanding incrementally."""
@property
def name(self) -> str:
return "add_understanding"
@property
def description(self) -> str:
return """Capture and store information about the user's business context,
workflows, pain points, and automation goals. Call this tool whenever the user
shares information about their business. Each call incrementally adds to the
existing understanding - you don't need to provide all fields at once.
Use this to build a comprehensive profile that helps recommend better agents
and automations for the user's specific needs."""
@property
def parameters(self) -> dict[str, Any]:
return {
"type": "object",
"properties": {
"user_name": {
"type": "string",
"description": "The user's name",
},
"job_title": {
"type": "string",
"description": "The user's job title (e.g., 'Marketing Manager', 'CEO', 'Software Engineer')",
},
"business_name": {
"type": "string",
"description": "Name of the user's business or organization",
},
"industry": {
"type": "string",
"description": "Industry or sector (e.g., 'e-commerce', 'healthcare', 'finance')",
},
"business_size": {
"type": "string",
"description": "Company size: '1-10', '11-50', '51-200', '201-1000', or '1000+'",
},
"user_role": {
"type": "string",
"description": "User's role in organization context (e.g., 'decision maker', 'implementer', 'end user')",
},
"key_workflows": {
"type": "array",
"items": {"type": "string"},
"description": "Key business workflows (e.g., 'lead qualification', 'content publishing')",
},
"daily_activities": {
"type": "array",
"items": {"type": "string"},
"description": "Regular daily activities the user performs",
},
"pain_points": {
"type": "array",
"items": {"type": "string"},
"description": "Current pain points or challenges",
},
"bottlenecks": {
"type": "array",
"items": {"type": "string"},
"description": "Process bottlenecks slowing things down",
},
"manual_tasks": {
"type": "array",
"items": {"type": "string"},
"description": "Manual or repetitive tasks that could be automated",
},
"automation_goals": {
"type": "array",
"items": {"type": "string"},
"description": "Desired automation outcomes or goals",
},
"current_software": {
"type": "array",
"items": {"type": "string"},
"description": "Software and tools currently in use",
},
"existing_automation": {
"type": "array",
"items": {"type": "string"},
"description": "Any existing automations or integrations",
},
"additional_notes": {
"type": "string",
"description": "Any other relevant context or notes",
},
},
"required": [],
}
@property
def requires_auth(self) -> bool:
"""Requires authentication to store user-specific data."""
return True
async def _execute(
self,
user_id: str | None,
session: ChatSession,
**kwargs,
) -> ToolResponseBase:
"""
Capture and store business understanding incrementally.
Each call merges new data with existing understanding:
- String fields are overwritten if provided
- List fields are appended (with deduplication)
"""
session_id = session.session_id
if not user_id:
return ErrorResponse(
message="Authentication required to save business understanding.",
session_id=session_id,
)
# Check if any data was provided
if not any(v is not None for v in kwargs.values()):
return ErrorResponse(
message="Please provide at least one field to update.",
session_id=session_id,
)
# Build input model
input_data = BusinessUnderstandingInput(
user_name=kwargs.get("user_name"),
job_title=kwargs.get("job_title"),
business_name=kwargs.get("business_name"),
industry=kwargs.get("industry"),
business_size=kwargs.get("business_size"),
user_role=kwargs.get("user_role"),
key_workflows=kwargs.get("key_workflows"),
daily_activities=kwargs.get("daily_activities"),
pain_points=kwargs.get("pain_points"),
bottlenecks=kwargs.get("bottlenecks"),
manual_tasks=kwargs.get("manual_tasks"),
automation_goals=kwargs.get("automation_goals"),
current_software=kwargs.get("current_software"),
existing_automation=kwargs.get("existing_automation"),
additional_notes=kwargs.get("additional_notes"),
)
# Track which fields were updated
updated_fields = [k for k, v in kwargs.items() if v is not None]
# Upsert with merge
understanding = await upsert_business_understanding(user_id, input_data)
# Build current understanding summary for the response
current_understanding = {
"user_name": understanding.user_name,
"job_title": understanding.job_title,
"business_name": understanding.business_name,
"industry": understanding.industry,
"business_size": understanding.business_size,
"user_role": understanding.user_role,
"key_workflows": understanding.key_workflows,
"daily_activities": understanding.daily_activities,
"pain_points": understanding.pain_points,
"bottlenecks": understanding.bottlenecks,
"manual_tasks": understanding.manual_tasks,
"automation_goals": understanding.automation_goals,
"current_software": understanding.current_software,
"existing_automation": understanding.existing_automation,
"additional_notes": understanding.additional_notes,
}
# Filter out empty values for cleaner response
current_understanding = {
k: v
for k, v in current_understanding.items()
if v is not None and v != [] and v != ""
}
return UnderstandingUpdatedResponse(
message=f"Updated understanding with: {', '.join(updated_fields)}. "
"I now have a better picture of your business context.",
session_id=session_id,
updated_fields=updated_fields,
current_understanding=current_understanding,
)

View File

@@ -1,455 +0,0 @@
"""Tool for retrieving agent execution outputs from user's library."""
import logging
import re
from datetime import datetime, timedelta, timezone
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.data import execution as execution_db
from backend.data.execution import ExecutionStatus, GraphExecution, GraphExecutionMeta
from .base import BaseTool
from .models import (
AgentOutputResponse,
ErrorResponse,
ExecutionOutputInfo,
NoResultsResponse,
ToolResponseBase,
)
from .utils import fetch_graph_from_store_slug
logger = logging.getLogger(__name__)
class AgentOutputInput(BaseModel):
"""Input parameters for the agent_output tool."""
agent_name: str = ""
library_agent_id: str = ""
store_slug: str = ""
execution_id: str = ""
run_time: str = "latest"
@field_validator(
"agent_name",
"library_agent_id",
"store_slug",
"execution_id",
"run_time",
mode="before",
)
@classmethod
def strip_strings(cls, v: Any) -> Any:
"""Strip whitespace from string fields."""
return v.strip() if isinstance(v, str) else v
def parse_time_expression(
time_expr: str | None,
) -> tuple[datetime | None, datetime | None]:
"""
Parse time expression into datetime range (start, end).
Supports:
- "latest" or None -> returns (None, None) to get most recent
- "yesterday" -> 24h window for yesterday
- "today" -> Today from midnight
- "last week" / "last 7 days" -> 7 day window
- "last month" / "last 30 days" -> 30 day window
- ISO date "YYYY-MM-DD" -> 24h window for that date
"""
if not time_expr or time_expr.lower() == "latest":
return None, None
now = datetime.now(timezone.utc)
expr = time_expr.lower().strip()
# Relative expressions
if expr == "yesterday":
end = now.replace(hour=0, minute=0, second=0, microsecond=0)
start = end - timedelta(days=1)
return start, end
if expr in ("last week", "last 7 days"):
return now - timedelta(days=7), now
if expr in ("last month", "last 30 days"):
return now - timedelta(days=30), now
if expr == "today":
start = now.replace(hour=0, minute=0, second=0, microsecond=0)
return start, now
# Try ISO date format (YYYY-MM-DD)
date_match = re.match(r"^(\d{4})-(\d{2})-(\d{2})$", expr)
if date_match:
year, month, day = map(int, date_match.groups())
start = datetime(year, month, day, 0, 0, 0, tzinfo=timezone.utc)
end = start + timedelta(days=1)
return start, end
# Try ISO datetime
try:
parsed = datetime.fromisoformat(expr.replace("Z", "+00:00"))
if parsed.tzinfo is None:
parsed = parsed.replace(tzinfo=timezone.utc)
# Return +/- 1 hour window around the specified time
return parsed - timedelta(hours=1), parsed + timedelta(hours=1)
except ValueError:
pass
# Fallback: treat as "latest"
return None, None
class AgentOutputTool(BaseTool):
"""Tool for retrieving execution outputs from user's library agents."""
@property
def name(self) -> str:
return "agent_output"
@property
def description(self) -> str:
return """Retrieve execution outputs from agents in the user's library.
Identify the agent using one of:
- agent_name: Fuzzy search in user's library
- library_agent_id: Exact library agent ID
- store_slug: Marketplace format 'username/agent-name'
Select which run to retrieve using:
- execution_id: Specific execution ID
- run_time: 'latest' (default), 'yesterday', 'last week', or ISO date 'YYYY-MM-DD'
"""
@property
def parameters(self) -> dict[str, Any]:
return {
"type": "object",
"properties": {
"agent_name": {
"type": "string",
"description": "Agent name to search for in user's library (fuzzy match)",
},
"library_agent_id": {
"type": "string",
"description": "Exact library agent ID",
},
"store_slug": {
"type": "string",
"description": "Marketplace identifier: 'username/agent-slug'",
},
"execution_id": {
"type": "string",
"description": "Specific execution ID to retrieve",
},
"run_time": {
"type": "string",
"description": (
"Time filter: 'latest', 'yesterday', 'last week', or 'YYYY-MM-DD'"
),
},
},
"required": [],
}
@property
def requires_auth(self) -> bool:
return True
async def _resolve_agent(
self,
user_id: str,
agent_name: str | None,
library_agent_id: str | None,
store_slug: str | None,
) -> tuple[LibraryAgent | None, str | None]:
"""
Resolve agent from provided identifiers.
Returns (library_agent, error_message).
"""
# Priority 1: Exact library agent ID
if library_agent_id:
try:
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}")
return None, f"Library agent '{library_agent_id}' not found"
# Priority 2: Store slug (username/agent-name)
if store_slug and "/" in store_slug:
username, agent_slug = store_slug.split("/", 1)
graph, _ = await fetch_graph_from_store_slug(username, agent_slug)
if not graph:
return None, f"Agent '{store_slug}' not found in marketplace"
# Find in user's library by graph_id
agent = await library_db.get_library_agent_by_graph_id(user_id, graph.id)
if not agent:
return (
None,
f"Agent '{store_slug}' is not in your library. "
"Add it first to see outputs.",
)
return agent, None
# Priority 3: Fuzzy name search in library
if agent_name:
try:
response = await library_db.list_library_agents(
user_id=user_id,
search_term=agent_name,
page_size=5,
)
if not response.agents:
return (
None,
f"No agents matching '{agent_name}' found in your library",
)
# Return best match (first result from search)
return response.agents[0], None
except Exception as e:
logger.error(f"Error searching library agents: {e}")
return None, f"Error searching for agent: {e}"
return (
None,
"Please specify an agent name, library_agent_id, or store_slug",
)
async def _get_execution(
self,
user_id: str,
graph_id: str,
execution_id: str | None,
time_start: datetime | None,
time_end: datetime | None,
) -> tuple[GraphExecution | None, list[GraphExecutionMeta], str | None]:
"""
Fetch execution(s) based on filters.
Returns (single_execution, available_executions_meta, error_message).
"""
# If specific execution_id provided, fetch it directly
if execution_id:
execution = await execution_db.get_graph_execution(
user_id=user_id,
execution_id=execution_id,
include_node_executions=False,
)
if not execution:
return None, [], f"Execution '{execution_id}' not found"
return execution, [], None
# Get completed executions with time filters
executions = await execution_db.get_graph_executions(
graph_id=graph_id,
user_id=user_id,
statuses=[ExecutionStatus.COMPLETED],
created_time_gte=time_start,
created_time_lte=time_end,
limit=10,
)
if not executions:
return None, [], None # No error, just no executions
# If only one execution, fetch full details
if len(executions) == 1:
full_execution = await execution_db.get_graph_execution(
user_id=user_id,
execution_id=executions[0].id,
include_node_executions=False,
)
return full_execution, [], None
# Multiple executions - return latest with full details, plus list of available
full_execution = await execution_db.get_graph_execution(
user_id=user_id,
execution_id=executions[0].id,
include_node_executions=False,
)
return full_execution, executions, None
def _build_response(
self,
agent: LibraryAgent,
execution: GraphExecution | None,
available_executions: list[GraphExecutionMeta],
session_id: str | None,
) -> AgentOutputResponse:
"""Build the response based on execution data."""
library_agent_link = f"/library/agents/{agent.id}"
if not execution:
return AgentOutputResponse(
message=f"No completed executions found for agent '{agent.name}'",
session_id=session_id,
agent_name=agent.name,
agent_id=agent.graph_id,
library_agent_id=agent.id,
library_agent_link=library_agent_link,
total_executions=0,
)
execution_info = ExecutionOutputInfo(
execution_id=execution.id,
status=execution.status.value,
started_at=execution.started_at,
ended_at=execution.ended_at,
outputs=dict(execution.outputs),
inputs_summary=execution.inputs if execution.inputs else None,
)
available_list = None
if len(available_executions) > 1:
available_list = [
{
"id": e.id,
"status": e.status.value,
"started_at": e.started_at.isoformat() if e.started_at else None,
}
for e in available_executions[:5]
]
message = f"Found execution outputs for agent '{agent.name}'"
if len(available_executions) > 1:
message += (
f". Showing latest of {len(available_executions)} matching executions."
)
return AgentOutputResponse(
message=message,
session_id=session_id,
agent_name=agent.name,
agent_id=agent.graph_id,
library_agent_id=agent.id,
library_agent_link=library_agent_link,
execution=execution_info,
available_executions=available_list,
total_executions=len(available_executions) if available_executions else 1,
)
async def _execute(
self,
user_id: str | None,
session: ChatSession,
**kwargs,
) -> ToolResponseBase:
"""Execute the agent_output tool."""
session_id = session.session_id
# Parse and validate input
try:
input_data = AgentOutputInput(**kwargs)
except Exception as e:
logger.error(f"Invalid input: {e}")
return ErrorResponse(
message="Invalid input parameters",
error=str(e),
session_id=session_id,
)
# Ensure user_id is present (should be guaranteed by requires_auth)
if not user_id:
return ErrorResponse(
message="User authentication required",
session_id=session_id,
)
# Check if at least one identifier is provided
if not any(
[
input_data.agent_name,
input_data.library_agent_id,
input_data.store_slug,
input_data.execution_id,
]
):
return ErrorResponse(
message=(
"Please specify at least one of: agent_name, "
"library_agent_id, store_slug, or execution_id"
),
session_id=session_id,
)
# If only execution_id provided, we need to find the agent differently
if (
input_data.execution_id
and not input_data.agent_name
and not input_data.library_agent_id
and not input_data.store_slug
):
# Fetch execution directly to get graph_id
execution = await execution_db.get_graph_execution(
user_id=user_id,
execution_id=input_data.execution_id,
include_node_executions=False,
)
if not execution:
return ErrorResponse(
message=f"Execution '{input_data.execution_id}' not found",
session_id=session_id,
)
# Find library agent by graph_id
agent = await library_db.get_library_agent_by_graph_id(
user_id, execution.graph_id
)
if not agent:
return NoResultsResponse(
message=(
f"Execution found but agent not in your library. "
f"Graph ID: {execution.graph_id}"
),
session_id=session_id,
suggestions=["Add the agent to your library to see more details"],
)
return self._build_response(agent, execution, [], session_id)
# Resolve agent from identifiers
agent, error = await self._resolve_agent(
user_id=user_id,
agent_name=input_data.agent_name or None,
library_agent_id=input_data.library_agent_id or None,
store_slug=input_data.store_slug or None,
)
if error or not agent:
return NoResultsResponse(
message=error or "Agent not found",
session_id=session_id,
suggestions=[
"Check the agent name or ID",
"Make sure the agent is in your library",
],
)
# Parse time expression
time_start, time_end = parse_time_expression(input_data.run_time)
# Fetch execution(s)
execution, available_executions, exec_error = await self._get_execution(
user_id=user_id,
graph_id=agent.graph_id,
execution_id=input_data.execution_id or None,
time_start=time_start,
time_end=time_end,
)
if exec_error:
return ErrorResponse(
message=exec_error,
session_id=session_id,
)
return self._build_response(agent, execution, available_executions, session_id)

View File

@@ -1,157 +0,0 @@
"""Tool for searching agents in the user's library."""
import logging
from typing import Any
from backend.api.features.chat.model import ChatSession
from backend.api.features.library import db as library_db
from backend.util.exceptions import DatabaseError
from .base import BaseTool
from .models import (
AgentCarouselResponse,
AgentInfo,
ErrorResponse,
NoResultsResponse,
ToolResponseBase,
)
logger = logging.getLogger(__name__)
class FindLibraryAgentTool(BaseTool):
"""Tool for searching agents in the user's library."""
@property
def name(self) -> str:
return "find_library_agent"
@property
def description(self) -> str:
return (
"Search for agents in the user's library. Use this to find agents "
"the user has already added to their library, including agents they "
"created or added from the marketplace."
)
@property
def parameters(self) -> dict[str, Any]:
return {
"type": "object",
"properties": {
"query": {
"type": "string",
"description": (
"Search query to find agents by name or description. "
"Use keywords for best results."
),
},
},
"required": ["query"],
}
@property
def requires_auth(self) -> bool:
return True
async def _execute(
self,
user_id: str | None,
session: ChatSession,
**kwargs,
) -> ToolResponseBase:
"""Search for agents in the user's library.
Args:
user_id: User ID (required)
session: Chat session
query: Search query
Returns:
AgentCarouselResponse: List of agents found in the library
NoResultsResponse: No agents found
ErrorResponse: Error message
"""
query = kwargs.get("query", "").strip()
session_id = session.session_id
if not query:
return ErrorResponse(
message="Please provide a search query",
session_id=session_id,
)
if not user_id:
return ErrorResponse(
message="User authentication required to search library",
session_id=session_id,
)
agents = []
try:
logger.info(f"Searching user library for: {query}")
library_results = await library_db.list_library_agents(
user_id=user_id,
search_term=query,
page_size=10,
)
logger.info(
f"Find library agents tool found {len(library_results.agents)} agents"
)
for agent in library_results.agents:
agents.append(
AgentInfo(
id=agent.id,
name=agent.name,
description=agent.description or "",
source="library",
in_library=True,
creator=agent.creator_name,
status=agent.status.value,
can_access_graph=agent.can_access_graph,
has_external_trigger=agent.has_external_trigger,
new_output=agent.new_output,
graph_id=agent.graph_id,
),
)
except DatabaseError as e:
logger.error(f"Error searching library agents: {e}", exc_info=True)
return ErrorResponse(
message="Failed to search library. Please try again.",
error=str(e),
session_id=session_id,
)
if not agents:
return NoResultsResponse(
message=(
f"No agents found matching '{query}' in your library. "
"Try different keywords or use find_agent to search the marketplace."
),
session_id=session_id,
suggestions=[
"Try more general terms",
"Use find_agent to search the marketplace",
"Check your library at /library",
],
)
title = (
f"Found {len(agents)} agent{'s' if len(agents) != 1 else ''} "
f"in your library for '{query}'"
)
return AgentCarouselResponse(
message=(
"Found agents in the user's library. You can provide a link to "
"view an agent at: /library/agents/{agent_id}. "
"Use agent_output to get execution results, or run_agent to execute."
),
title=title,
agents=agents,
count=len(agents),
session_id=session_id,
)

View File

@@ -1,833 +0,0 @@
"""
OAuth 2.0 Provider Endpoints
Implements OAuth 2.0 Authorization Code flow with PKCE support.
Flow:
1. User clicks "Login with AutoGPT" in 3rd party app
2. App redirects user to /auth/authorize with client_id, redirect_uri, scope, state
3. User sees consent screen (if not already logged in, redirects to login first)
4. User approves → backend creates authorization code
5. User redirected back to app with code
6. App exchanges code for access/refresh tokens at /api/oauth/token
7. App uses access token to call external API endpoints
"""
import io
import logging
import os
import uuid
from datetime import datetime
from typing import Literal, Optional
from urllib.parse import urlencode
from autogpt_libs.auth import get_user_id
from fastapi import APIRouter, Body, HTTPException, Security, UploadFile, status
from gcloud.aio import storage as async_storage
from PIL import Image
from prisma.enums import APIKeyPermission
from pydantic import BaseModel, Field
from backend.data.auth.oauth import (
InvalidClientError,
InvalidGrantError,
OAuthApplicationInfo,
TokenIntrospectionResult,
consume_authorization_code,
create_access_token,
create_authorization_code,
create_refresh_token,
get_oauth_application,
get_oauth_application_by_id,
introspect_token,
list_user_oauth_applications,
refresh_tokens,
revoke_access_token,
revoke_refresh_token,
update_oauth_application,
validate_client_credentials,
validate_redirect_uri,
validate_scopes,
)
from backend.util.settings import Settings
from backend.util.virus_scanner import scan_content_safe
settings = Settings()
logger = logging.getLogger(__name__)
router = APIRouter()
# ============================================================================
# Request/Response Models
# ============================================================================
class TokenResponse(BaseModel):
"""OAuth 2.0 token response"""
token_type: Literal["Bearer"] = "Bearer"
access_token: str
access_token_expires_at: datetime
refresh_token: str
refresh_token_expires_at: datetime
scopes: list[str]
class ErrorResponse(BaseModel):
"""OAuth 2.0 error response"""
error: str
error_description: Optional[str] = None
class OAuthApplicationPublicInfo(BaseModel):
"""Public information about an OAuth application (for consent screen)"""
name: str
description: Optional[str] = None
logo_url: Optional[str] = None
scopes: list[str]
# ============================================================================
# Application Info Endpoint
# ============================================================================
@router.get(
"/app/{client_id}",
responses={
404: {"description": "Application not found or disabled"},
},
)
async def get_oauth_app_info(
client_id: str, user_id: str = Security(get_user_id)
) -> OAuthApplicationPublicInfo:
"""
Get public information about an OAuth application.
This endpoint is used by the consent screen to display application details
to the user before they authorize access.
Returns:
- name: Application name
- description: Application description (if provided)
- scopes: List of scopes the application is allowed to request
"""
app = await get_oauth_application(client_id)
if not app or not app.is_active:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail="Application not found",
)
return OAuthApplicationPublicInfo(
name=app.name,
description=app.description,
logo_url=app.logo_url,
scopes=[s.value for s in app.scopes],
)
# ============================================================================
# Authorization Endpoint
# ============================================================================
class AuthorizeRequest(BaseModel):
"""OAuth 2.0 authorization request"""
client_id: str = Field(description="Client identifier")
redirect_uri: str = Field(description="Redirect URI")
scopes: list[str] = Field(description="List of scopes")
state: str = Field(description="Anti-CSRF token from client")
response_type: str = Field(
default="code", description="Must be 'code' for authorization code flow"
)
code_challenge: str = Field(description="PKCE code challenge (required)")
code_challenge_method: Literal["S256", "plain"] = Field(
default="S256", description="PKCE code challenge method (S256 recommended)"
)
class AuthorizeResponse(BaseModel):
"""OAuth 2.0 authorization response with redirect URL"""
redirect_url: str = Field(description="URL to redirect the user to")
@router.post("/authorize")
async def authorize(
request: AuthorizeRequest = Body(),
user_id: str = Security(get_user_id),
) -> AuthorizeResponse:
"""
OAuth 2.0 Authorization Endpoint
User must be logged in (authenticated with Supabase JWT).
This endpoint creates an authorization code and returns a redirect URL.
PKCE (Proof Key for Code Exchange) is REQUIRED for all authorization requests.
The frontend consent screen should call this endpoint after the user approves,
then redirect the user to the returned `redirect_url`.
Request Body:
- client_id: The OAuth application's client ID
- redirect_uri: Where to redirect after authorization (must match registered URI)
- scopes: List of permissions (e.g., "EXECUTE_GRAPH READ_GRAPH")
- state: Anti-CSRF token provided by client (will be returned in redirect)
- response_type: Must be "code" (for authorization code flow)
- code_challenge: PKCE code challenge (required)
- code_challenge_method: "S256" (recommended) or "plain"
Returns:
- redirect_url: The URL to redirect the user to (includes authorization code)
Error cases return a redirect_url with error parameters, or raise HTTPException
for critical errors (like invalid redirect_uri).
"""
try:
# Validate response_type
if request.response_type != "code":
return _error_redirect_url(
request.redirect_uri,
request.state,
"unsupported_response_type",
"Only 'code' response type is supported",
)
# Get application
app = await get_oauth_application(request.client_id)
if not app:
return _error_redirect_url(
request.redirect_uri,
request.state,
"invalid_client",
"Unknown client_id",
)
if not app.is_active:
return _error_redirect_url(
request.redirect_uri,
request.state,
"invalid_client",
"Application is not active",
)
# Validate redirect URI
if not validate_redirect_uri(app, request.redirect_uri):
# For invalid redirect_uri, we can't redirect safely
# Must return error instead
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=(
"Invalid redirect_uri. "
f"Must be one of: {', '.join(app.redirect_uris)}"
),
)
# Parse and validate scopes
try:
requested_scopes = [APIKeyPermission(s.strip()) for s in request.scopes]
except ValueError as e:
return _error_redirect_url(
request.redirect_uri,
request.state,
"invalid_scope",
f"Invalid scope: {e}",
)
if not requested_scopes:
return _error_redirect_url(
request.redirect_uri,
request.state,
"invalid_scope",
"At least one scope is required",
)
if not validate_scopes(app, requested_scopes):
return _error_redirect_url(
request.redirect_uri,
request.state,
"invalid_scope",
"Application is not authorized for all requested scopes. "
f"Allowed: {', '.join(s.value for s in app.scopes)}",
)
# Create authorization code
auth_code = await create_authorization_code(
application_id=app.id,
user_id=user_id,
scopes=requested_scopes,
redirect_uri=request.redirect_uri,
code_challenge=request.code_challenge,
code_challenge_method=request.code_challenge_method,
)
# Build redirect URL with authorization code
params = {
"code": auth_code.code,
"state": request.state,
}
redirect_url = f"{request.redirect_uri}?{urlencode(params)}"
logger.info(
f"Authorization code issued for user #{user_id} "
f"and app {app.name} (#{app.id})"
)
return AuthorizeResponse(redirect_url=redirect_url)
except HTTPException:
raise
except Exception as e:
logger.error(f"Error in authorization endpoint: {e}", exc_info=True)
return _error_redirect_url(
request.redirect_uri,
request.state,
"server_error",
"An unexpected error occurred",
)
def _error_redirect_url(
redirect_uri: str,
state: str,
error: str,
error_description: Optional[str] = None,
) -> AuthorizeResponse:
"""Helper to build redirect URL with OAuth error parameters"""
params = {
"error": error,
"state": state,
}
if error_description:
params["error_description"] = error_description
redirect_url = f"{redirect_uri}?{urlencode(params)}"
return AuthorizeResponse(redirect_url=redirect_url)
# ============================================================================
# Token Endpoint
# ============================================================================
class TokenRequestByCode(BaseModel):
grant_type: Literal["authorization_code"]
code: str = Field(description="Authorization code")
redirect_uri: str = Field(
description="Redirect URI (must match authorization request)"
)
client_id: str
client_secret: str
code_verifier: str = Field(description="PKCE code verifier")
class TokenRequestByRefreshToken(BaseModel):
grant_type: Literal["refresh_token"]
refresh_token: str
client_id: str
client_secret: str
@router.post("/token")
async def token(
request: TokenRequestByCode | TokenRequestByRefreshToken = Body(),
) -> TokenResponse:
"""
OAuth 2.0 Token Endpoint
Exchanges authorization code or refresh token for access token.
Grant Types:
1. authorization_code: Exchange authorization code for tokens
- Required: grant_type, code, redirect_uri, client_id, client_secret
- Optional: code_verifier (required if PKCE was used)
2. refresh_token: Exchange refresh token for new access token
- Required: grant_type, refresh_token, client_id, client_secret
Returns:
- access_token: Bearer token for API access (1 hour TTL)
- token_type: "Bearer"
- expires_in: Seconds until access token expires
- refresh_token: Token for refreshing access (30 days TTL)
- scopes: List of scopes
"""
# Validate client credentials
try:
app = await validate_client_credentials(
request.client_id, request.client_secret
)
except InvalidClientError as e:
raise HTTPException(
status_code=status.HTTP_401_UNAUTHORIZED,
detail=str(e),
)
# Handle authorization_code grant
if request.grant_type == "authorization_code":
# Consume authorization code
try:
user_id, scopes = await consume_authorization_code(
code=request.code,
application_id=app.id,
redirect_uri=request.redirect_uri,
code_verifier=request.code_verifier,
)
except InvalidGrantError as e:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=str(e),
)
# Create access and refresh tokens
access_token = await create_access_token(app.id, user_id, scopes)
refresh_token = await create_refresh_token(app.id, user_id, scopes)
logger.info(
f"Access token issued for user #{user_id} and app {app.name} (#{app.id})"
"via authorization code"
)
if not access_token.token or not refresh_token.token:
raise HTTPException(
status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
detail="Failed to generate tokens",
)
return TokenResponse(
token_type="Bearer",
access_token=access_token.token.get_secret_value(),
access_token_expires_at=access_token.expires_at,
refresh_token=refresh_token.token.get_secret_value(),
refresh_token_expires_at=refresh_token.expires_at,
scopes=list(s.value for s in scopes),
)
# Handle refresh_token grant
elif request.grant_type == "refresh_token":
# Refresh access token
try:
new_access_token, new_refresh_token = await refresh_tokens(
request.refresh_token, app.id
)
except InvalidGrantError as e:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=str(e),
)
logger.info(
f"Tokens refreshed for user #{new_access_token.user_id} "
f"by app {app.name} (#{app.id})"
)
if not new_access_token.token or not new_refresh_token.token:
raise HTTPException(
status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
detail="Failed to generate tokens",
)
return TokenResponse(
token_type="Bearer",
access_token=new_access_token.token.get_secret_value(),
access_token_expires_at=new_access_token.expires_at,
refresh_token=new_refresh_token.token.get_secret_value(),
refresh_token_expires_at=new_refresh_token.expires_at,
scopes=list(s.value for s in new_access_token.scopes),
)
else:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=f"Unsupported grant_type: {request.grant_type}. "
"Must be 'authorization_code' or 'refresh_token'",
)
# ============================================================================
# Token Introspection Endpoint
# ============================================================================
@router.post("/introspect")
async def introspect(
token: str = Body(description="Token to introspect"),
token_type_hint: Optional[Literal["access_token", "refresh_token"]] = Body(
None, description="Hint about token type ('access_token' or 'refresh_token')"
),
client_id: str = Body(description="Client identifier"),
client_secret: str = Body(description="Client secret"),
) -> TokenIntrospectionResult:
"""
OAuth 2.0 Token Introspection Endpoint (RFC 7662)
Allows clients to check if a token is valid and get its metadata.
Returns:
- active: Whether the token is currently active
- scopes: List of authorized scopes (if active)
- client_id: The client the token was issued to (if active)
- user_id: The user the token represents (if active)
- exp: Expiration timestamp (if active)
- token_type: "access_token" or "refresh_token" (if active)
"""
# Validate client credentials
try:
await validate_client_credentials(client_id, client_secret)
except InvalidClientError as e:
raise HTTPException(
status_code=status.HTTP_401_UNAUTHORIZED,
detail=str(e),
)
# Introspect the token
return await introspect_token(token, token_type_hint)
# ============================================================================
# Token Revocation Endpoint
# ============================================================================
@router.post("/revoke")
async def revoke(
token: str = Body(description="Token to revoke"),
token_type_hint: Optional[Literal["access_token", "refresh_token"]] = Body(
None, description="Hint about token type ('access_token' or 'refresh_token')"
),
client_id: str = Body(description="Client identifier"),
client_secret: str = Body(description="Client secret"),
):
"""
OAuth 2.0 Token Revocation Endpoint (RFC 7009)
Allows clients to revoke an access or refresh token.
Note: Revoking a refresh token does NOT revoke associated access tokens.
Revoking an access token does NOT revoke the associated refresh token.
"""
# Validate client credentials
try:
app = await validate_client_credentials(client_id, client_secret)
except InvalidClientError as e:
raise HTTPException(
status_code=status.HTTP_401_UNAUTHORIZED,
detail=str(e),
)
# Try to revoke as access token first
# Note: We pass app.id to ensure the token belongs to the authenticated app
if token_type_hint != "refresh_token":
revoked = await revoke_access_token(token, app.id)
if revoked:
logger.info(
f"Access token revoked for app {app.name} (#{app.id}); "
f"user #{revoked.user_id}"
)
return {"status": "ok"}
# Try to revoke as refresh token
revoked = await revoke_refresh_token(token, app.id)
if revoked:
logger.info(
f"Refresh token revoked for app {app.name} (#{app.id}); "
f"user #{revoked.user_id}"
)
return {"status": "ok"}
# Per RFC 7009, revocation endpoint returns 200 even if token not found
# or if token belongs to a different application.
# This prevents token scanning attacks.
logger.warning(f"Unsuccessful token revocation attempt by app {app.name} #{app.id}")
return {"status": "ok"}
# ============================================================================
# Application Management Endpoints (for app owners)
# ============================================================================
@router.get("/apps/mine")
async def list_my_oauth_apps(
user_id: str = Security(get_user_id),
) -> list[OAuthApplicationInfo]:
"""
List all OAuth applications owned by the current user.
Returns a list of OAuth applications with their details including:
- id, name, description, logo_url
- client_id (public identifier)
- redirect_uris, grant_types, scopes
- is_active status
- created_at, updated_at timestamps
Note: client_secret is never returned for security reasons.
"""
return await list_user_oauth_applications(user_id)
@router.patch("/apps/{app_id}/status")
async def update_app_status(
app_id: str,
user_id: str = Security(get_user_id),
is_active: bool = Body(description="Whether the app should be active", embed=True),
) -> OAuthApplicationInfo:
"""
Enable or disable an OAuth application.
Only the application owner can update the status.
When disabled, the application cannot be used for new authorizations
and existing access tokens will fail validation.
Returns the updated application info.
"""
updated_app = await update_oauth_application(
app_id=app_id,
owner_id=user_id,
is_active=is_active,
)
if not updated_app:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail="Application not found or you don't have permission to update it",
)
action = "enabled" if is_active else "disabled"
logger.info(f"OAuth app {updated_app.name} (#{app_id}) {action} by user #{user_id}")
return updated_app
class UpdateAppLogoRequest(BaseModel):
logo_url: str = Field(description="URL of the uploaded logo image")
@router.patch("/apps/{app_id}/logo")
async def update_app_logo(
app_id: str,
request: UpdateAppLogoRequest = Body(),
user_id: str = Security(get_user_id),
) -> OAuthApplicationInfo:
"""
Update the logo URL for an OAuth application.
Only the application owner can update the logo.
The logo should be uploaded first using the media upload endpoint,
then this endpoint is called with the resulting URL.
Logo requirements:
- Must be square (1:1 aspect ratio)
- Minimum 512x512 pixels
- Maximum 2048x2048 pixels
Returns the updated application info.
"""
if (
not (app := await get_oauth_application_by_id(app_id))
or app.owner_id != user_id
):
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail="OAuth App not found",
)
# Delete the current app logo file (if any and it's in our cloud storage)
await _delete_app_current_logo_file(app)
updated_app = await update_oauth_application(
app_id=app_id,
owner_id=user_id,
logo_url=request.logo_url,
)
if not updated_app:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail="Application not found or you don't have permission to update it",
)
logger.info(
f"OAuth app {updated_app.name} (#{app_id}) logo updated by user #{user_id}"
)
return updated_app
# Logo upload constraints
LOGO_MIN_SIZE = 512
LOGO_MAX_SIZE = 2048
LOGO_ALLOWED_TYPES = {"image/jpeg", "image/png", "image/webp"}
LOGO_MAX_FILE_SIZE = 3 * 1024 * 1024 # 3MB
@router.post("/apps/{app_id}/logo/upload")
async def upload_app_logo(
app_id: str,
file: UploadFile,
user_id: str = Security(get_user_id),
) -> OAuthApplicationInfo:
"""
Upload a logo image for an OAuth application.
Requirements:
- Image must be square (1:1 aspect ratio)
- Minimum 512x512 pixels
- Maximum 2048x2048 pixels
- Allowed formats: JPEG, PNG, WebP
- Maximum file size: 3MB
The image is uploaded to cloud storage and the app's logoUrl is updated.
Returns the updated application info.
"""
# Verify ownership to reduce vulnerability to DoS(torage) or DoM(oney) attacks
if (
not (app := await get_oauth_application_by_id(app_id))
or app.owner_id != user_id
):
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail="OAuth App not found",
)
# Check GCS configuration
if not settings.config.media_gcs_bucket_name:
raise HTTPException(
status_code=status.HTTP_503_SERVICE_UNAVAILABLE,
detail="Media storage is not configured",
)
# Validate content type
content_type = file.content_type
if content_type not in LOGO_ALLOWED_TYPES:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=f"Invalid file type. Allowed: JPEG, PNG, WebP. Got: {content_type}",
)
# Read file content
try:
file_bytes = await file.read()
except Exception as e:
logger.error(f"Error reading logo file: {e}")
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail="Failed to read uploaded file",
)
# Check file size
if len(file_bytes) > LOGO_MAX_FILE_SIZE:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=(
"File too large. "
f"Maximum size is {LOGO_MAX_FILE_SIZE // 1024 // 1024}MB"
),
)
# Validate image dimensions
try:
image = Image.open(io.BytesIO(file_bytes))
width, height = image.size
if width != height:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=f"Logo must be square. Got {width}x{height}",
)
if width < LOGO_MIN_SIZE:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=f"Logo too small. Minimum {LOGO_MIN_SIZE}x{LOGO_MIN_SIZE}. "
f"Got {width}x{height}",
)
if width > LOGO_MAX_SIZE:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=f"Logo too large. Maximum {LOGO_MAX_SIZE}x{LOGO_MAX_SIZE}. "
f"Got {width}x{height}",
)
except HTTPException:
raise
except Exception as e:
logger.error(f"Error validating logo image: {e}")
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail="Invalid image file",
)
# Scan for viruses
filename = file.filename or "logo"
await scan_content_safe(file_bytes, filename=filename)
# Generate unique filename
file_ext = os.path.splitext(filename)[1].lower() or ".png"
unique_filename = f"{uuid.uuid4()}{file_ext}"
storage_path = f"oauth-apps/{app_id}/logo/{unique_filename}"
# Upload to GCS
try:
async with async_storage.Storage() as async_client:
bucket_name = settings.config.media_gcs_bucket_name
await async_client.upload(
bucket_name, storage_path, file_bytes, content_type=content_type
)
logo_url = f"https://storage.googleapis.com/{bucket_name}/{storage_path}"
except Exception as e:
logger.error(f"Error uploading logo to GCS: {e}")
raise HTTPException(
status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
detail="Failed to upload logo",
)
# Delete the current app logo file (if any and it's in our cloud storage)
await _delete_app_current_logo_file(app)
# Update the app with the new logo URL
updated_app = await update_oauth_application(
app_id=app_id,
owner_id=user_id,
logo_url=logo_url,
)
if not updated_app:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail="Application not found or you don't have permission to update it",
)
logger.info(
f"OAuth app {updated_app.name} (#{app_id}) logo uploaded by user #{user_id}"
)
return updated_app
async def _delete_app_current_logo_file(app: OAuthApplicationInfo):
"""
Delete the current logo file for the given app, if there is one in our cloud storage
"""
bucket_name = settings.config.media_gcs_bucket_name
storage_base_url = f"https://storage.googleapis.com/{bucket_name}/"
if app.logo_url and app.logo_url.startswith(storage_base_url):
# Parse blob path from URL: https://storage.googleapis.com/{bucket}/{path}
old_path = app.logo_url.replace(storage_base_url, "")
try:
async with async_storage.Storage() as async_client:
await async_client.delete(bucket_name, old_path)
logger.info(f"Deleted old logo for OAuth app #{app.id}: {old_path}")
except Exception as e:
# Log but don't fail - the new logo was uploaded successfully
logger.warning(
f"Failed to delete old logo for OAuth app #{app.id}: {e}", exc_info=e
)

File diff suppressed because it is too large Load Diff

View File

@@ -1,41 +0,0 @@
from fastapi import FastAPI
def sort_openapi(app: FastAPI) -> None:
"""
Patch a FastAPI instance's `openapi()` method to sort the endpoints,
schemas, and responses.
"""
wrapped_openapi = app.openapi
def custom_openapi():
if app.openapi_schema:
return app.openapi_schema
openapi_schema = wrapped_openapi()
# Sort endpoints
openapi_schema["paths"] = dict(sorted(openapi_schema["paths"].items()))
# Sort endpoints -> methods
for p in openapi_schema["paths"].keys():
openapi_schema["paths"][p] = dict(
sorted(openapi_schema["paths"][p].items())
)
# Sort endpoints -> methods -> responses
for m in openapi_schema["paths"][p].keys():
openapi_schema["paths"][p][m]["responses"] = dict(
sorted(openapi_schema["paths"][p][m]["responses"].items())
)
# Sort schemas and responses as well
for k in openapi_schema["components"].keys():
openapi_schema["components"][k] = dict(
sorted(openapi_schema["components"][k].items())
)
app.openapi_schema = openapi_schema
return openapi_schema
app.openapi = custom_openapi

View File

@@ -36,10 +36,10 @@ def main(**kwargs):
Run all the processes required for the AutoGPT-server (REST and WebSocket APIs).
"""
from backend.api.rest_api import AgentServer
from backend.api.ws_api import WebsocketServer
from backend.executor import DatabaseManager, ExecutionManager, Scheduler
from backend.notifications import NotificationManager
from backend.server.rest_api import AgentServer
from backend.server.ws_api import WebsocketServer
run_processes(
DatabaseManager().set_log_level("warning"),

View File

@@ -1,7 +1,6 @@
from typing import Any
from backend.blocks.llm import (
DEFAULT_LLM_MODEL,
TEST_CREDENTIALS,
TEST_CREDENTIALS_INPUT,
AIBlockBase,
@@ -50,7 +49,7 @@ class AIConditionBlock(AIBlockBase):
)
model: LlmModel = SchemaField(
title="LLM Model",
default=DEFAULT_LLM_MODEL,
default=LlmModel.GPT4O,
description="The language model to use for evaluating the condition.",
advanced=False,
)
@@ -82,7 +81,7 @@ class AIConditionBlock(AIBlockBase):
"condition": "the input is an email address",
"yes_value": "Valid email",
"no_value": "Not an email",
"model": DEFAULT_LLM_MODEL,
"model": LlmModel.GPT4O,
"credentials": TEST_CREDENTIALS_INPUT,
},
test_credentials=TEST_CREDENTIALS,

View File

@@ -20,7 +20,6 @@ from backend.data.model import (
SchemaField,
)
from backend.integrations.providers import ProviderName
from backend.util.exceptions import BlockExecutionError
from backend.util.request import Requests
TEST_CREDENTIALS = APIKeyCredentials(
@@ -247,11 +246,7 @@ class AIShortformVideoCreatorBlock(Block):
await asyncio.sleep(10)
logger.error("Video creation timed out")
raise BlockExecutionError(
message="Video creation timed out",
block_name=self.name,
block_id=self.id,
)
raise TimeoutError("Video creation timed out")
def __init__(self):
super().__init__(
@@ -427,11 +422,7 @@ class AIAdMakerVideoCreatorBlock(Block):
await asyncio.sleep(10)
logger.error("Video creation timed out")
raise BlockExecutionError(
message="Video creation timed out",
block_name=self.name,
block_id=self.id,
)
raise TimeoutError("Video creation timed out")
def __init__(self):
super().__init__(
@@ -608,11 +599,7 @@ class AIScreenshotToVideoAdBlock(Block):
await asyncio.sleep(10)
logger.error("Video creation timed out")
raise BlockExecutionError(
message="Video creation timed out",
block_name=self.name,
block_id=self.id,
)
raise TimeoutError("Video creation timed out")
def __init__(self):
super().__init__(

View File

@@ -106,10 +106,7 @@ class ConditionBlock(Block):
ComparisonOperator.LESS_THAN_OR_EQUAL: lambda a, b: a <= b,
}
try:
result = comparison_funcs[operator](value1, value2)
except Exception as e:
raise ValueError(f"Comparison failed: {e}") from e
result = comparison_funcs[operator](value1, value2)
yield "result", result

View File

@@ -182,10 +182,13 @@ class DataForSeoRelatedKeywordsBlock(Block):
if results and len(results) > 0:
# results is a list, get the first element
first_result = results[0] if isinstance(results, list) else results
# Handle missing key, null value, or valid list value
if isinstance(first_result, dict):
items = first_result.get("items") or []
else:
items = (
first_result.get("items", [])
if isinstance(first_result, dict)
else []
)
# Ensure items is never None
if items is None:
items = []
for item in items:
# Extract keyword_data from the item

View File

@@ -15,7 +15,6 @@ from backend.sdk import (
SchemaField,
cost,
)
from backend.util.exceptions import BlockExecutionError
from ._config import firecrawl
@@ -60,18 +59,11 @@ class FirecrawlExtractBlock(Block):
) -> BlockOutput:
app = FirecrawlApp(api_key=credentials.api_key.get_secret_value())
try:
extract_result = app.extract(
urls=input_data.urls,
prompt=input_data.prompt,
schema=input_data.output_schema,
enable_web_search=input_data.enable_web_search,
)
except Exception as e:
raise BlockExecutionError(
message=f"Extract failed: {e}",
block_name=self.name,
block_id=self.id,
) from e
extract_result = app.extract(
urls=input_data.urls,
prompt=input_data.prompt,
schema=input_data.output_schema,
enable_web_search=input_data.enable_web_search,
)
yield "data", extract_result.data

View File

@@ -19,7 +19,6 @@ from backend.data.model import (
SchemaField,
)
from backend.integrations.providers import ProviderName
from backend.util.exceptions import ModerationError
from backend.util.file import MediaFileType, store_media_file
TEST_CREDENTIALS = APIKeyCredentials(
@@ -154,8 +153,6 @@ class AIImageEditorBlock(Block):
),
aspect_ratio=input_data.aspect_ratio.value,
seed=input_data.seed,
user_id=user_id,
graph_exec_id=graph_exec_id,
)
yield "output_image", result
@@ -167,8 +164,6 @@ class AIImageEditorBlock(Block):
input_image_b64: Optional[str],
aspect_ratio: str,
seed: Optional[int],
user_id: str,
graph_exec_id: str,
) -> MediaFileType:
client = ReplicateClient(api_token=api_key.get_secret_value())
input_params = {
@@ -178,21 +173,11 @@ class AIImageEditorBlock(Block):
**({"seed": seed} if seed is not None else {}),
}
try:
output: FileOutput | list[FileOutput] = await client.async_run( # type: ignore
model_name,
input=input_params,
wait=False,
)
except Exception as e:
if "flagged as sensitive" in str(e).lower():
raise ModerationError(
message="Content was flagged as sensitive by the model provider",
user_id=user_id,
graph_exec_id=graph_exec_id,
moderation_type="model_provider",
)
raise ValueError(f"Model execution failed: {e}") from e
output: FileOutput | list[FileOutput] = await client.async_run( # type: ignore
model_name,
input=input_params,
wait=False,
)
if isinstance(output, list) and output:
output = output[0]

File diff suppressed because it is too large Load Diff

View File

@@ -2,6 +2,7 @@ from enum import Enum
from typing import Any, Dict, Literal, Optional
from pydantic import SecretStr
from requests.exceptions import RequestException
from backend.data.block import (
Block,
@@ -331,8 +332,8 @@ class IdeogramModelBlock(Block):
try:
response = await Requests().post(url, headers=headers, json=data)
return response.json()["data"][0]["url"]
except Exception as e:
raise ValueError(f"Failed to fetch image with V3 endpoint: {e}") from e
except RequestException as e:
raise Exception(f"Failed to fetch image with V3 endpoint: {str(e)}")
async def _run_model_legacy(
self,
@@ -384,8 +385,8 @@ class IdeogramModelBlock(Block):
try:
response = await Requests().post(url, headers=headers, json=data)
return response.json()["data"][0]["url"]
except Exception as e:
raise ValueError(f"Failed to fetch image with legacy endpoint: {e}") from e
except RequestException as e:
raise Exception(f"Failed to fetch image with legacy endpoint: {str(e)}")
async def upscale_image(self, api_key: SecretStr, image_url: str):
url = "https://api.ideogram.ai/upscale"
@@ -412,5 +413,5 @@ class IdeogramModelBlock(Block):
return (response.json())["data"][0]["url"]
except Exception as e:
raise ValueError(f"Failed to upscale image: {e}") from e
except RequestException as e:
raise Exception(f"Failed to upscale image: {str(e)}")

View File

@@ -16,7 +16,6 @@ from backend.data.block import (
BlockSchemaOutput,
)
from backend.data.model import SchemaField
from backend.util.exceptions import BlockExecutionError
class SearchTheWebBlock(Block, GetRequest):
@@ -57,17 +56,7 @@ class SearchTheWebBlock(Block, GetRequest):
# Prepend the Jina Search URL to the encoded query
jina_search_url = f"https://s.jina.ai/{encoded_query}"
try:
results = await self.get_request(
jina_search_url, headers=headers, json=False
)
except Exception as e:
raise BlockExecutionError(
message=f"Search failed: {e}",
block_name=self.name,
block_id=self.id,
) from e
results = await self.get_request(jina_search_url, headers=headers, json=False)
# Output the search results
yield "results", results

View File

@@ -92,9 +92,8 @@ class LlmModel(str, Enum, metaclass=LlmModelMeta):
O1 = "o1"
O1_MINI = "o1-mini"
# GPT-5 models
GPT5_2 = "gpt-5.2-2025-12-11"
GPT5_1 = "gpt-5.1-2025-11-13"
GPT5 = "gpt-5-2025-08-07"
GPT5_1 = "gpt-5.1-2025-11-13"
GPT5_MINI = "gpt-5-mini-2025-08-07"
GPT5_NANO = "gpt-5-nano-2025-08-07"
GPT5_CHAT = "gpt-5-chat-latest"
@@ -195,9 +194,8 @@ MODEL_METADATA = {
LlmModel.O1: ModelMetadata("openai", 200000, 100000), # o1-2024-12-17
LlmModel.O1_MINI: ModelMetadata("openai", 128000, 65536), # o1-mini-2024-09-12
# GPT-5 models
LlmModel.GPT5_2: ModelMetadata("openai", 400000, 128000),
LlmModel.GPT5_1: ModelMetadata("openai", 400000, 128000),
LlmModel.GPT5: ModelMetadata("openai", 400000, 128000),
LlmModel.GPT5_1: ModelMetadata("openai", 400000, 128000),
LlmModel.GPT5_MINI: ModelMetadata("openai", 400000, 128000),
LlmModel.GPT5_NANO: ModelMetadata("openai", 400000, 128000),
LlmModel.GPT5_CHAT: ModelMetadata("openai", 400000, 16384),
@@ -305,8 +303,6 @@ MODEL_METADATA = {
LlmModel.V0_1_0_MD: ModelMetadata("v0", 128000, 64000),
}
DEFAULT_LLM_MODEL = LlmModel.GPT5_2
for model in LlmModel:
if model not in MODEL_METADATA:
raise ValueError(f"Missing MODEL_METADATA metadata for model: {model}")
@@ -794,7 +790,7 @@ class AIStructuredResponseGeneratorBlock(AIBlockBase):
)
model: LlmModel = SchemaField(
title="LLM Model",
default=DEFAULT_LLM_MODEL,
default=LlmModel.GPT4O,
description="The language model to use for answering the prompt.",
advanced=False,
)
@@ -859,7 +855,7 @@ class AIStructuredResponseGeneratorBlock(AIBlockBase):
input_schema=AIStructuredResponseGeneratorBlock.Input,
output_schema=AIStructuredResponseGeneratorBlock.Output,
test_input={
"model": DEFAULT_LLM_MODEL,
"model": LlmModel.GPT4O,
"credentials": TEST_CREDENTIALS_INPUT,
"expected_format": {
"key1": "value1",
@@ -1225,7 +1221,7 @@ class AITextGeneratorBlock(AIBlockBase):
)
model: LlmModel = SchemaField(
title="LLM Model",
default=DEFAULT_LLM_MODEL,
default=LlmModel.GPT4O,
description="The language model to use for answering the prompt.",
advanced=False,
)
@@ -1321,7 +1317,7 @@ class AITextSummarizerBlock(AIBlockBase):
)
model: LlmModel = SchemaField(
title="LLM Model",
default=DEFAULT_LLM_MODEL,
default=LlmModel.GPT4O,
description="The language model to use for summarizing the text.",
)
focus: str = SchemaField(
@@ -1538,7 +1534,7 @@ class AIConversationBlock(AIBlockBase):
)
model: LlmModel = SchemaField(
title="LLM Model",
default=DEFAULT_LLM_MODEL,
default=LlmModel.GPT4O,
description="The language model to use for the conversation.",
)
credentials: AICredentials = AICredentialsField()
@@ -1576,7 +1572,7 @@ class AIConversationBlock(AIBlockBase):
},
{"role": "user", "content": "Where was it played?"},
],
"model": DEFAULT_LLM_MODEL,
"model": LlmModel.GPT4O,
"credentials": TEST_CREDENTIALS_INPUT,
},
test_credentials=TEST_CREDENTIALS,
@@ -1639,7 +1635,7 @@ class AIListGeneratorBlock(AIBlockBase):
)
model: LlmModel = SchemaField(
title="LLM Model",
default=DEFAULT_LLM_MODEL,
default=LlmModel.GPT4O,
description="The language model to use for generating the list.",
advanced=True,
)
@@ -1696,7 +1692,7 @@ class AIListGeneratorBlock(AIBlockBase):
"drawing explorers to uncover its mysteries. Each planet showcases the limitless possibilities of "
"fictional worlds."
),
"model": DEFAULT_LLM_MODEL,
"model": LlmModel.GPT4O,
"credentials": TEST_CREDENTIALS_INPUT,
"max_retries": 3,
"force_json_output": False,

View File

@@ -18,7 +18,6 @@ from backend.data.block import (
BlockSchemaOutput,
)
from backend.data.model import APIKeyCredentials, CredentialsField, SchemaField
from backend.util.exceptions import BlockExecutionError, BlockInputError
logger = logging.getLogger(__name__)
@@ -112,27 +111,9 @@ class ReplicateModelBlock(Block):
yield "status", "succeeded"
yield "model_name", input_data.model_name
except Exception as e:
error_msg = str(e)
logger.error(f"Error running Replicate model: {error_msg}")
# Input validation errors (422, 400) → BlockInputError
if (
"422" in error_msg
or "Input validation failed" in error_msg
or "400" in error_msg
):
raise BlockInputError(
message=f"Invalid model inputs: {error_msg}",
block_name=self.name,
block_id=self.id,
) from e
# Everything else → BlockExecutionError
else:
raise BlockExecutionError(
message=f"Replicate model error: {error_msg}",
block_name=self.name,
block_id=self.id,
) from e
error_msg = f"Unexpected error running Replicate model: {str(e)}"
logger.error(error_msg)
raise RuntimeError(error_msg)
async def run_model(self, model_ref: str, model_inputs: dict, api_key: SecretStr):
"""

View File

@@ -45,16 +45,10 @@ class GetWikipediaSummaryBlock(Block, GetRequest):
async def run(self, input_data: Input, **kwargs) -> BlockOutput:
topic = input_data.topic
url = f"https://en.wikipedia.org/api/rest_v1/page/summary/{topic}"
# Note: User-Agent is now automatically set by the request library
# to comply with Wikimedia's robot policy (https://w.wiki/4wJS)
try:
response = await self.get_request(url, json=True)
if "extract" not in response:
raise ValueError(f"Unable to parse Wikipedia response: {response}")
yield "summary", response["extract"]
except Exception as e:
raise ValueError(f"Failed to fetch Wikipedia summary: {e}") from e
response = await self.get_request(url, json=True)
if "extract" not in response:
raise RuntimeError(f"Unable to parse Wikipedia response: {response}")
yield "summary", response["extract"]
TEST_CREDENTIALS = APIKeyCredentials(

View File

@@ -226,7 +226,7 @@ class SmartDecisionMakerBlock(Block):
)
model: llm.LlmModel = SchemaField(
title="LLM Model",
default=llm.DEFAULT_LLM_MODEL,
default=llm.LlmModel.GPT4O,
description="The language model to use for answering the prompt.",
advanced=False,
)

View File

@@ -196,15 +196,6 @@ class TestXMLParserBlockSecurity:
async for _ in block.run(XMLParserBlock.Input(input_xml=large_xml)):
pass
async def test_rejects_text_outside_root(self):
"""Ensure parser surfaces readable errors for invalid root text."""
block = XMLParserBlock()
invalid_xml = "<root><child>value</child></root> trailing"
with pytest.raises(ValueError, match="text outside the root element"):
async for _ in block.run(XMLParserBlock.Input(input_xml=invalid_xml)):
pass
class TestStoreMediaFileSecurity:
"""Test file storage security limits."""

View File

@@ -28,7 +28,7 @@ class TestLLMStatsTracking:
response = await llm.llm_call(
credentials=llm.TEST_CREDENTIALS,
llm_model=llm.DEFAULT_LLM_MODEL,
llm_model=llm.LlmModel.GPT4O,
prompt=[{"role": "user", "content": "Hello"}],
max_tokens=100,
)
@@ -65,7 +65,7 @@ class TestLLMStatsTracking:
input_data = llm.AIStructuredResponseGeneratorBlock.Input(
prompt="Test prompt",
expected_format={"key1": "desc1", "key2": "desc2"},
model=llm.DEFAULT_LLM_MODEL,
model=llm.LlmModel.GPT4O,
credentials=llm.TEST_CREDENTIALS_INPUT, # type: ignore # type: ignore
)
@@ -109,7 +109,7 @@ class TestLLMStatsTracking:
# Run the block
input_data = llm.AITextGeneratorBlock.Input(
prompt="Generate text",
model=llm.DEFAULT_LLM_MODEL,
model=llm.LlmModel.GPT4O,
credentials=llm.TEST_CREDENTIALS_INPUT, # type: ignore
)
@@ -170,7 +170,7 @@ class TestLLMStatsTracking:
input_data = llm.AIStructuredResponseGeneratorBlock.Input(
prompt="Test prompt",
expected_format={"key1": "desc1", "key2": "desc2"},
model=llm.DEFAULT_LLM_MODEL,
model=llm.LlmModel.GPT4O,
credentials=llm.TEST_CREDENTIALS_INPUT, # type: ignore
retry=2,
)
@@ -228,7 +228,7 @@ class TestLLMStatsTracking:
input_data = llm.AITextSummarizerBlock.Input(
text=long_text,
model=llm.DEFAULT_LLM_MODEL,
model=llm.LlmModel.GPT4O,
credentials=llm.TEST_CREDENTIALS_INPUT, # type: ignore
max_tokens=100, # Small chunks
chunk_overlap=10,
@@ -299,7 +299,7 @@ class TestLLMStatsTracking:
# Test with very short text (should only need 1 chunk + 1 final summary)
input_data = llm.AITextSummarizerBlock.Input(
text="This is a short text.",
model=llm.DEFAULT_LLM_MODEL,
model=llm.LlmModel.GPT4O,
credentials=llm.TEST_CREDENTIALS_INPUT, # type: ignore
max_tokens=1000, # Large enough to avoid chunking
)
@@ -346,7 +346,7 @@ class TestLLMStatsTracking:
{"role": "assistant", "content": "Hi there!"},
{"role": "user", "content": "How are you?"},
],
model=llm.DEFAULT_LLM_MODEL,
model=llm.LlmModel.GPT4O,
credentials=llm.TEST_CREDENTIALS_INPUT, # type: ignore
)
@@ -387,7 +387,7 @@ class TestLLMStatsTracking:
# Run the block
input_data = llm.AIListGeneratorBlock.Input(
focus="test items",
model=llm.DEFAULT_LLM_MODEL,
model=llm.LlmModel.GPT4O,
credentials=llm.TEST_CREDENTIALS_INPUT, # type: ignore
max_retries=3,
)
@@ -469,7 +469,7 @@ class TestLLMStatsTracking:
input_data = llm.AIStructuredResponseGeneratorBlock.Input(
prompt="Test",
expected_format={"result": "desc"},
model=llm.DEFAULT_LLM_MODEL,
model=llm.LlmModel.GPT4O,
credentials=llm.TEST_CREDENTIALS_INPUT, # type: ignore
)
@@ -513,7 +513,7 @@ class TestAITextSummarizerValidation:
# Create input data
input_data = llm.AITextSummarizerBlock.Input(
text="Some text to summarize",
model=llm.DEFAULT_LLM_MODEL,
model=llm.LlmModel.GPT4O,
credentials=llm.TEST_CREDENTIALS_INPUT, # type: ignore
style=llm.SummaryStyle.BULLET_POINTS,
)
@@ -558,7 +558,7 @@ class TestAITextSummarizerValidation:
# Create input data
input_data = llm.AITextSummarizerBlock.Input(
text="Some text to summarize",
model=llm.DEFAULT_LLM_MODEL,
model=llm.LlmModel.GPT4O,
credentials=llm.TEST_CREDENTIALS_INPUT, # type: ignore
style=llm.SummaryStyle.BULLET_POINTS,
max_tokens=1000,
@@ -593,7 +593,7 @@ class TestAITextSummarizerValidation:
# Create input data
input_data = llm.AITextSummarizerBlock.Input(
text="Some text to summarize",
model=llm.DEFAULT_LLM_MODEL,
model=llm.LlmModel.GPT4O,
credentials=llm.TEST_CREDENTIALS_INPUT, # type: ignore
)
@@ -623,7 +623,7 @@ class TestAITextSummarizerValidation:
# Create input data
input_data = llm.AITextSummarizerBlock.Input(
text="Some text to summarize",
model=llm.DEFAULT_LLM_MODEL,
model=llm.LlmModel.GPT4O,
credentials=llm.TEST_CREDENTIALS_INPUT, # type: ignore
max_tokens=1000,
)
@@ -654,7 +654,7 @@ class TestAITextSummarizerValidation:
# Create input data
input_data = llm.AITextSummarizerBlock.Input(
text="Some text to summarize",
model=llm.DEFAULT_LLM_MODEL,
model=llm.LlmModel.GPT4O,
credentials=llm.TEST_CREDENTIALS_INPUT, # type: ignore
)

View File

@@ -5,10 +5,10 @@ from unittest.mock import AsyncMock, MagicMock, patch
import pytest
from backend.api.model import CreateGraph
from backend.api.rest_api import AgentServer
from backend.data.execution import ExecutionContext
from backend.data.model import ProviderName, User
from backend.server.model import CreateGraph
from backend.server.rest_api import AgentServer
from backend.usecases.sample import create_test_graph, create_test_user
from backend.util.test import SpinTestServer, wait_execution
@@ -233,7 +233,7 @@ async def test_smart_decision_maker_tracks_llm_stats():
# Create test input
input_data = SmartDecisionMakerBlock.Input(
prompt="Should I continue with this task?",
model=llm_module.DEFAULT_LLM_MODEL,
model=llm_module.LlmModel.GPT4O,
credentials=llm_module.TEST_CREDENTIALS_INPUT, # type: ignore
agent_mode_max_iterations=0,
)
@@ -335,7 +335,7 @@ async def test_smart_decision_maker_parameter_validation():
input_data = SmartDecisionMakerBlock.Input(
prompt="Search for keywords",
model=llm_module.DEFAULT_LLM_MODEL,
model=llm_module.LlmModel.GPT4O,
credentials=llm_module.TEST_CREDENTIALS_INPUT, # type: ignore
retry=2, # Set retry to 2 for testing
agent_mode_max_iterations=0,
@@ -402,7 +402,7 @@ async def test_smart_decision_maker_parameter_validation():
input_data = SmartDecisionMakerBlock.Input(
prompt="Search for keywords",
model=llm_module.DEFAULT_LLM_MODEL,
model=llm_module.LlmModel.GPT4O,
credentials=llm_module.TEST_CREDENTIALS_INPUT, # type: ignore
agent_mode_max_iterations=0,
)
@@ -462,7 +462,7 @@ async def test_smart_decision_maker_parameter_validation():
input_data = SmartDecisionMakerBlock.Input(
prompt="Search for keywords",
model=llm_module.DEFAULT_LLM_MODEL,
model=llm_module.LlmModel.GPT4O,
credentials=llm_module.TEST_CREDENTIALS_INPUT, # type: ignore
agent_mode_max_iterations=0,
)
@@ -526,7 +526,7 @@ async def test_smart_decision_maker_parameter_validation():
input_data = SmartDecisionMakerBlock.Input(
prompt="Search for keywords",
model=llm_module.DEFAULT_LLM_MODEL,
model=llm_module.LlmModel.GPT4O,
credentials=llm_module.TEST_CREDENTIALS_INPUT, # type: ignore
agent_mode_max_iterations=0,
)
@@ -648,7 +648,7 @@ async def test_smart_decision_maker_raw_response_conversion():
input_data = SmartDecisionMakerBlock.Input(
prompt="Test prompt",
model=llm_module.DEFAULT_LLM_MODEL,
model=llm_module.LlmModel.GPT4O,
credentials=llm_module.TEST_CREDENTIALS_INPUT, # type: ignore
retry=2,
agent_mode_max_iterations=0,
@@ -722,7 +722,7 @@ async def test_smart_decision_maker_raw_response_conversion():
):
input_data = SmartDecisionMakerBlock.Input(
prompt="Simple prompt",
model=llm_module.DEFAULT_LLM_MODEL,
model=llm_module.LlmModel.GPT4O,
credentials=llm_module.TEST_CREDENTIALS_INPUT, # type: ignore
agent_mode_max_iterations=0,
)
@@ -778,7 +778,7 @@ async def test_smart_decision_maker_raw_response_conversion():
):
input_data = SmartDecisionMakerBlock.Input(
prompt="Another test",
model=llm_module.DEFAULT_LLM_MODEL,
model=llm_module.LlmModel.GPT4O,
credentials=llm_module.TEST_CREDENTIALS_INPUT, # type: ignore
agent_mode_max_iterations=0,
)
@@ -931,7 +931,7 @@ async def test_smart_decision_maker_agent_mode():
# Test agent mode with max_iterations = 3
input_data = SmartDecisionMakerBlock.Input(
prompt="Complete this task using tools",
model=llm_module.DEFAULT_LLM_MODEL,
model=llm_module.LlmModel.GPT4O,
credentials=llm_module.TEST_CREDENTIALS_INPUT, # type: ignore
agent_mode_max_iterations=3, # Enable agent mode with 3 max iterations
)
@@ -1020,7 +1020,7 @@ async def test_smart_decision_maker_traditional_mode_default():
# Test default behavior (traditional mode)
input_data = SmartDecisionMakerBlock.Input(
prompt="Test prompt",
model=llm_module.DEFAULT_LLM_MODEL,
model=llm_module.LlmModel.GPT4O,
credentials=llm_module.TEST_CREDENTIALS_INPUT, # type: ignore
agent_mode_max_iterations=0, # Traditional mode
)

View File

@@ -373,7 +373,7 @@ async def test_output_yielding_with_dynamic_fields():
input_data = block.input_schema(
prompt="Create a user dictionary",
credentials=llm.TEST_CREDENTIALS_INPUT,
model=llm.DEFAULT_LLM_MODEL,
model=llm.LlmModel.GPT4O,
agent_mode_max_iterations=0, # Use traditional mode to test output yielding
)
@@ -594,7 +594,7 @@ async def test_validation_errors_dont_pollute_conversation():
input_data = block.input_schema(
prompt="Test prompt",
credentials=llm.TEST_CREDENTIALS_INPUT,
model=llm.DEFAULT_LLM_MODEL,
model=llm.LlmModel.GPT4O,
retry=3, # Allow retries
agent_mode_max_iterations=1,
)

View File

@@ -1,5 +1,5 @@
from gravitasml.parser import Parser
from gravitasml.token import Token, tokenize
from gravitasml.token import tokenize
from backend.data.block import Block, BlockOutput, BlockSchemaInput, BlockSchemaOutput
from backend.data.model import SchemaField
@@ -25,38 +25,6 @@ class XMLParserBlock(Block):
],
)
@staticmethod
def _validate_tokens(tokens: list[Token]) -> None:
"""Ensure the XML has a single root element and no stray text."""
if not tokens:
raise ValueError("XML input is empty.")
depth = 0
root_seen = False
for token in tokens:
if token.type == "TAG_OPEN":
if depth == 0 and root_seen:
raise ValueError("XML must have a single root element.")
depth += 1
if depth == 1:
root_seen = True
elif token.type == "TAG_CLOSE":
depth -= 1
if depth < 0:
raise SyntaxError("Unexpected closing tag in XML input.")
elif token.type in {"TEXT", "ESCAPE"}:
if depth == 0 and token.value:
raise ValueError(
"XML contains text outside the root element; "
"wrap content in a single root tag."
)
if depth != 0:
raise SyntaxError("Unclosed tag detected in XML input.")
if not root_seen:
raise ValueError("XML must include a root element.")
async def run(self, input_data: Input, **kwargs) -> BlockOutput:
# Security fix: Add size limits to prevent XML bomb attacks
MAX_XML_SIZE = 10 * 1024 * 1024 # 10MB limit for XML input
@@ -67,9 +35,7 @@ class XMLParserBlock(Block):
)
try:
tokens = list(tokenize(input_data.input_xml))
self._validate_tokens(tokens)
tokens = tokenize(input_data.input_xml)
parser = Parser(tokens)
parsed_result = parser.parse()
yield "parsed_xml", parsed_result

View File

@@ -111,8 +111,6 @@ class TranscribeYoutubeVideoBlock(Block):
return parsed_url.path.split("/")[2]
if parsed_url.path[:3] == "/v/":
return parsed_url.path.split("/")[2]
if parsed_url.path.startswith("/shorts/"):
return parsed_url.path.split("/")[2]
raise ValueError(f"Invalid YouTube URL: {url}")
def get_transcript(

View File

@@ -244,7 +244,11 @@ def websocket(server_address: str, graph_exec_id: str):
import websockets.asyncio.client
from backend.api.ws_api import WSMessage, WSMethod, WSSubscribeGraphExecutionRequest
from backend.server.ws_api import (
WSMessage,
WSMethod,
WSSubscribeGraphExecutionRequest,
)
async def send_message(server_address: str):
uri = f"ws://{server_address}"

View File

@@ -1 +0,0 @@
"""CLI utilities for backend development & administration"""

View File

@@ -1,57 +0,0 @@
#!/usr/bin/env python3
"""
Script to generate OpenAPI JSON specification for the FastAPI app.
This script imports the FastAPI app from backend.api.rest_api and outputs
the OpenAPI specification as JSON to stdout or a specified file.
Usage:
`poetry run python generate_openapi_json.py`
`poetry run python generate_openapi_json.py --output openapi.json`
`poetry run python generate_openapi_json.py --indent 4 --output openapi.json`
"""
import json
import os
from pathlib import Path
import click
@click.command()
@click.option(
"--output",
type=click.Path(dir_okay=False, path_type=Path),
help="Output file path (default: stdout)",
)
@click.option(
"--pretty",
type=click.BOOL,
default=False,
help="Pretty-print JSON output (indented 2 spaces)",
)
def main(output: Path, pretty: bool):
"""Generate and output the OpenAPI JSON specification."""
openapi_schema = get_openapi_schema()
json_output = json.dumps(openapi_schema, indent=2 if pretty else None)
if output:
output.write_text(json_output)
click.echo(f"✅ OpenAPI specification written to {output}\n\nPreview:")
click.echo(f"\n{json_output[:500]} ...")
else:
print(json_output)
def get_openapi_schema():
"""Get the OpenAPI schema from the FastAPI app"""
from backend.api.rest_api import app
return app.openapi()
if __name__ == "__main__":
os.environ["LOG_LEVEL"] = "ERROR" # disable stdout log output
main()

File diff suppressed because it is too large Load Diff

View File

@@ -1,4 +1,4 @@
from backend.api.features.library.model import LibraryAgentPreset
from backend.server.v2.library.model import LibraryAgentPreset
from .graph import NodeModel
from .integrations import Webhook # noqa: F401

View File

@@ -1,24 +1,22 @@
import logging
import uuid
from datetime import datetime, timezone
from typing import Literal, Optional
from typing import Optional
from autogpt_libs.api_key.keysmith import APIKeySmith
from prisma.enums import APIKeyPermission, APIKeyStatus
from prisma.models import APIKey as PrismaAPIKey
from prisma.types import APIKeyWhereUniqueInput
from pydantic import Field
from pydantic import BaseModel, Field
from backend.data.includes import MAX_USER_API_KEYS_FETCH
from backend.util.exceptions import NotAuthorizedError, NotFoundError
from .base import APIAuthorizationInfo
logger = logging.getLogger(__name__)
keysmith = APIKeySmith()
class APIKeyInfo(APIAuthorizationInfo):
class APIKeyInfo(BaseModel):
id: str
name: str
head: str = Field(
@@ -28,9 +26,12 @@ class APIKeyInfo(APIAuthorizationInfo):
description=f"The last {APIKeySmith.TAIL_LENGTH} characters of the key"
)
status: APIKeyStatus
permissions: list[APIKeyPermission]
created_at: datetime
last_used_at: Optional[datetime] = None
revoked_at: Optional[datetime] = None
description: Optional[str] = None
type: Literal["api_key"] = "api_key" # type: ignore
user_id: str
@staticmethod
def from_db(api_key: PrismaAPIKey):
@@ -40,7 +41,7 @@ class APIKeyInfo(APIAuthorizationInfo):
head=api_key.head,
tail=api_key.tail,
status=APIKeyStatus(api_key.status),
scopes=[APIKeyPermission(p) for p in api_key.permissions],
permissions=[APIKeyPermission(p) for p in api_key.permissions],
created_at=api_key.createdAt,
last_used_at=api_key.lastUsedAt,
revoked_at=api_key.revokedAt,
@@ -210,7 +211,7 @@ async def suspend_api_key(key_id: str, user_id: str) -> APIKeyInfo:
def has_permission(api_key: APIKeyInfo, required_permission: APIKeyPermission) -> bool:
return required_permission in api_key.scopes
return required_permission in api_key.permissions
async def get_api_key_by_id(key_id: str, user_id: str) -> Optional[APIKeyInfo]:

View File

@@ -1,15 +0,0 @@
from datetime import datetime
from typing import Literal, Optional
from prisma.enums import APIKeyPermission
from pydantic import BaseModel
class APIAuthorizationInfo(BaseModel):
user_id: str
scopes: list[APIKeyPermission]
type: Literal["oauth", "api_key"]
created_at: datetime
expires_at: Optional[datetime] = None
last_used_at: Optional[datetime] = None
revoked_at: Optional[datetime] = None

View File

@@ -1,872 +0,0 @@
"""
OAuth 2.0 Provider Data Layer
Handles management of OAuth applications, authorization codes,
access tokens, and refresh tokens.
Hashing strategy:
- Access tokens & Refresh tokens: SHA256 (deterministic, allows direct lookup by hash)
- Client secrets: Scrypt with salt (lookup by client_id, then verify with salt)
"""
import hashlib
import logging
import secrets
import uuid
from datetime import datetime, timedelta, timezone
from typing import Literal, Optional
from autogpt_libs.api_key.keysmith import APIKeySmith
from prisma.enums import APIKeyPermission as APIPermission
from prisma.models import OAuthAccessToken as PrismaOAuthAccessToken
from prisma.models import OAuthApplication as PrismaOAuthApplication
from prisma.models import OAuthAuthorizationCode as PrismaOAuthAuthorizationCode
from prisma.models import OAuthRefreshToken as PrismaOAuthRefreshToken
from prisma.types import OAuthApplicationUpdateInput
from pydantic import BaseModel, Field, SecretStr
from .base import APIAuthorizationInfo
logger = logging.getLogger(__name__)
keysmith = APIKeySmith() # Only used for client secret hashing (Scrypt)
def _generate_token() -> str:
"""Generate a cryptographically secure random token."""
return secrets.token_urlsafe(32)
def _hash_token(token: str) -> str:
"""Hash a token using SHA256 (deterministic, for direct lookup)."""
return hashlib.sha256(token.encode()).hexdigest()
# Token TTLs
AUTHORIZATION_CODE_TTL = timedelta(minutes=10)
ACCESS_TOKEN_TTL = timedelta(hours=1)
REFRESH_TOKEN_TTL = timedelta(days=30)
ACCESS_TOKEN_PREFIX = "agpt_xt_"
REFRESH_TOKEN_PREFIX = "agpt_rt_"
# ============================================================================
# Exception Classes
# ============================================================================
class OAuthError(Exception):
"""Base OAuth error"""
pass
class InvalidClientError(OAuthError):
"""Invalid client_id or client_secret"""
pass
class InvalidGrantError(OAuthError):
"""Invalid or expired authorization code/refresh token"""
def __init__(self, reason: str):
self.reason = reason
super().__init__(f"Invalid grant: {reason}")
class InvalidTokenError(OAuthError):
"""Invalid, expired, or revoked token"""
def __init__(self, reason: str):
self.reason = reason
super().__init__(f"Invalid token: {reason}")
# ============================================================================
# Data Models
# ============================================================================
class OAuthApplicationInfo(BaseModel):
"""OAuth application information (without client secret hash)"""
id: str
name: str
description: Optional[str] = None
logo_url: Optional[str] = None
client_id: str
redirect_uris: list[str]
grant_types: list[str]
scopes: list[APIPermission]
owner_id: str
is_active: bool
created_at: datetime
updated_at: datetime
@staticmethod
def from_db(app: PrismaOAuthApplication):
return OAuthApplicationInfo(
id=app.id,
name=app.name,
description=app.description,
logo_url=app.logoUrl,
client_id=app.clientId,
redirect_uris=app.redirectUris,
grant_types=app.grantTypes,
scopes=[APIPermission(s) for s in app.scopes],
owner_id=app.ownerId,
is_active=app.isActive,
created_at=app.createdAt,
updated_at=app.updatedAt,
)
class OAuthApplicationInfoWithSecret(OAuthApplicationInfo):
"""OAuth application with client secret hash (for validation)"""
client_secret_hash: str
client_secret_salt: str
@staticmethod
def from_db(app: PrismaOAuthApplication):
return OAuthApplicationInfoWithSecret(
**OAuthApplicationInfo.from_db(app).model_dump(),
client_secret_hash=app.clientSecret,
client_secret_salt=app.clientSecretSalt,
)
def verify_secret(self, plaintext_secret: str) -> bool:
"""Verify a plaintext client secret against the stored hash"""
# Use keysmith.verify_key() with stored salt
return keysmith.verify_key(
plaintext_secret, self.client_secret_hash, self.client_secret_salt
)
class OAuthAuthorizationCodeInfo(BaseModel):
"""Authorization code information"""
id: str
code: str
created_at: datetime
expires_at: datetime
application_id: str
user_id: str
scopes: list[APIPermission]
redirect_uri: str
code_challenge: Optional[str] = None
code_challenge_method: Optional[str] = None
used_at: Optional[datetime] = None
@property
def is_used(self) -> bool:
return self.used_at is not None
@staticmethod
def from_db(code: PrismaOAuthAuthorizationCode):
return OAuthAuthorizationCodeInfo(
id=code.id,
code=code.code,
created_at=code.createdAt,
expires_at=code.expiresAt,
application_id=code.applicationId,
user_id=code.userId,
scopes=[APIPermission(s) for s in code.scopes],
redirect_uri=code.redirectUri,
code_challenge=code.codeChallenge,
code_challenge_method=code.codeChallengeMethod,
used_at=code.usedAt,
)
class OAuthAccessTokenInfo(APIAuthorizationInfo):
"""Access token information"""
id: str
expires_at: datetime # type: ignore
application_id: str
type: Literal["oauth"] = "oauth" # type: ignore
@staticmethod
def from_db(token: PrismaOAuthAccessToken):
return OAuthAccessTokenInfo(
id=token.id,
user_id=token.userId,
scopes=[APIPermission(s) for s in token.scopes],
created_at=token.createdAt,
expires_at=token.expiresAt,
last_used_at=None,
revoked_at=token.revokedAt,
application_id=token.applicationId,
)
class OAuthAccessToken(OAuthAccessTokenInfo):
"""Access token with plaintext token included (sensitive)"""
token: SecretStr = Field(description="Plaintext token (sensitive)")
@staticmethod
def from_db(token: PrismaOAuthAccessToken, plaintext_token: str): # type: ignore
return OAuthAccessToken(
**OAuthAccessTokenInfo.from_db(token).model_dump(),
token=SecretStr(plaintext_token),
)
class OAuthRefreshTokenInfo(BaseModel):
"""Refresh token information"""
id: str
user_id: str
scopes: list[APIPermission]
created_at: datetime
expires_at: datetime
application_id: str
revoked_at: Optional[datetime] = None
@property
def is_revoked(self) -> bool:
return self.revoked_at is not None
@staticmethod
def from_db(token: PrismaOAuthRefreshToken):
return OAuthRefreshTokenInfo(
id=token.id,
user_id=token.userId,
scopes=[APIPermission(s) for s in token.scopes],
created_at=token.createdAt,
expires_at=token.expiresAt,
application_id=token.applicationId,
revoked_at=token.revokedAt,
)
class OAuthRefreshToken(OAuthRefreshTokenInfo):
"""Refresh token with plaintext token included (sensitive)"""
token: SecretStr = Field(description="Plaintext token (sensitive)")
@staticmethod
def from_db(token: PrismaOAuthRefreshToken, plaintext_token: str): # type: ignore
return OAuthRefreshToken(
**OAuthRefreshTokenInfo.from_db(token).model_dump(),
token=SecretStr(plaintext_token),
)
class TokenIntrospectionResult(BaseModel):
"""Result of token introspection (RFC 7662)"""
active: bool
scopes: Optional[list[str]] = None
client_id: Optional[str] = None
user_id: Optional[str] = None
exp: Optional[int] = None # Unix timestamp
token_type: Optional[Literal["access_token", "refresh_token"]] = None
# ============================================================================
# OAuth Application Management
# ============================================================================
async def get_oauth_application(client_id: str) -> Optional[OAuthApplicationInfo]:
"""Get OAuth application by client ID (without secret)"""
app = await PrismaOAuthApplication.prisma().find_unique(
where={"clientId": client_id}
)
if not app:
return None
return OAuthApplicationInfo.from_db(app)
async def get_oauth_application_with_secret(
client_id: str,
) -> Optional[OAuthApplicationInfoWithSecret]:
"""Get OAuth application by client ID (with secret hash for validation)"""
app = await PrismaOAuthApplication.prisma().find_unique(
where={"clientId": client_id}
)
if not app:
return None
return OAuthApplicationInfoWithSecret.from_db(app)
async def validate_client_credentials(
client_id: str, client_secret: str
) -> OAuthApplicationInfo:
"""
Validate client credentials and return application info.
Raises:
InvalidClientError: If client_id or client_secret is invalid, or app is inactive
"""
app = await get_oauth_application_with_secret(client_id)
if not app:
raise InvalidClientError("Invalid client_id")
if not app.is_active:
raise InvalidClientError("Application is not active")
# Verify client secret
if not app.verify_secret(client_secret):
raise InvalidClientError("Invalid client_secret")
# Return without secret hash
return OAuthApplicationInfo(**app.model_dump(exclude={"client_secret_hash"}))
def validate_redirect_uri(app: OAuthApplicationInfo, redirect_uri: str) -> bool:
"""Validate that redirect URI is registered for the application"""
return redirect_uri in app.redirect_uris
def validate_scopes(
app: OAuthApplicationInfo, requested_scopes: list[APIPermission]
) -> bool:
"""Validate that all requested scopes are allowed for the application"""
return all(scope in app.scopes for scope in requested_scopes)
# ============================================================================
# Authorization Code Flow
# ============================================================================
def _generate_authorization_code() -> str:
"""Generate a cryptographically secure authorization code"""
# 32 bytes = 256 bits of entropy
return secrets.token_urlsafe(32)
async def create_authorization_code(
application_id: str,
user_id: str,
scopes: list[APIPermission],
redirect_uri: str,
code_challenge: Optional[str] = None,
code_challenge_method: Optional[Literal["S256", "plain"]] = None,
) -> OAuthAuthorizationCodeInfo:
"""
Create a new authorization code.
Expires in 10 minutes and can only be used once.
"""
code = _generate_authorization_code()
now = datetime.now(timezone.utc)
expires_at = now + AUTHORIZATION_CODE_TTL
saved_code = await PrismaOAuthAuthorizationCode.prisma().create(
data={
"id": str(uuid.uuid4()),
"code": code,
"expiresAt": expires_at,
"applicationId": application_id,
"userId": user_id,
"scopes": [s for s in scopes],
"redirectUri": redirect_uri,
"codeChallenge": code_challenge,
"codeChallengeMethod": code_challenge_method,
}
)
return OAuthAuthorizationCodeInfo.from_db(saved_code)
async def consume_authorization_code(
code: str,
application_id: str,
redirect_uri: str,
code_verifier: Optional[str] = None,
) -> tuple[str, list[APIPermission]]:
"""
Consume an authorization code and return (user_id, scopes).
This marks the code as used and validates:
- Code exists and matches application
- Code is not expired
- Code has not been used
- Redirect URI matches
- PKCE code verifier matches (if code challenge was provided)
Raises:
InvalidGrantError: If code is invalid, expired, used, or PKCE fails
"""
auth_code = await PrismaOAuthAuthorizationCode.prisma().find_unique(
where={"code": code}
)
if not auth_code:
raise InvalidGrantError("authorization code not found")
# Validate application
if auth_code.applicationId != application_id:
raise InvalidGrantError(
"authorization code does not belong to this application"
)
# Check if already used
if auth_code.usedAt is not None:
raise InvalidGrantError(
f"authorization code already used at {auth_code.usedAt}"
)
# Check expiration
now = datetime.now(timezone.utc)
if auth_code.expiresAt < now:
raise InvalidGrantError("authorization code expired")
# Validate redirect URI
if auth_code.redirectUri != redirect_uri:
raise InvalidGrantError("redirect_uri mismatch")
# Validate PKCE if code challenge was provided
if auth_code.codeChallenge:
if not code_verifier:
raise InvalidGrantError("code_verifier required but not provided")
if not _verify_pkce(
code_verifier, auth_code.codeChallenge, auth_code.codeChallengeMethod
):
raise InvalidGrantError("PKCE verification failed")
# Mark code as used
await PrismaOAuthAuthorizationCode.prisma().update(
where={"code": code},
data={"usedAt": now},
)
return auth_code.userId, [APIPermission(s) for s in auth_code.scopes]
def _verify_pkce(
code_verifier: str, code_challenge: str, code_challenge_method: Optional[str]
) -> bool:
"""
Verify PKCE code verifier against code challenge.
Supports:
- S256: SHA256(code_verifier) == code_challenge
- plain: code_verifier == code_challenge
"""
if code_challenge_method == "S256":
# Hash the verifier with SHA256 and base64url encode
hashed = hashlib.sha256(code_verifier.encode("ascii")).digest()
computed_challenge = (
secrets.token_urlsafe(len(hashed)).encode("ascii").decode("ascii")
)
# For proper base64url encoding
import base64
computed_challenge = (
base64.urlsafe_b64encode(hashed).decode("ascii").rstrip("=")
)
return secrets.compare_digest(computed_challenge, code_challenge)
elif code_challenge_method == "plain" or code_challenge_method is None:
# Plain comparison
return secrets.compare_digest(code_verifier, code_challenge)
else:
logger.warning(f"Unsupported code challenge method: {code_challenge_method}")
return False
# ============================================================================
# Access Token Management
# ============================================================================
async def create_access_token(
application_id: str, user_id: str, scopes: list[APIPermission]
) -> OAuthAccessToken:
"""
Create a new access token.
Returns OAuthAccessToken (with plaintext token).
"""
plaintext_token = ACCESS_TOKEN_PREFIX + _generate_token()
token_hash = _hash_token(plaintext_token)
now = datetime.now(timezone.utc)
expires_at = now + ACCESS_TOKEN_TTL
saved_token = await PrismaOAuthAccessToken.prisma().create(
data={
"id": str(uuid.uuid4()),
"token": token_hash, # SHA256 hash for direct lookup
"expiresAt": expires_at,
"applicationId": application_id,
"userId": user_id,
"scopes": [s for s in scopes],
}
)
return OAuthAccessToken.from_db(saved_token, plaintext_token=plaintext_token)
async def validate_access_token(
token: str,
) -> tuple[OAuthAccessTokenInfo, OAuthApplicationInfo]:
"""
Validate an access token and return token info.
Raises:
InvalidTokenError: If token is invalid, expired, or revoked
InvalidClientError: If the client application is not marked as active
"""
token_hash = _hash_token(token)
# Direct lookup by hash
access_token = await PrismaOAuthAccessToken.prisma().find_unique(
where={"token": token_hash}, include={"Application": True}
)
if not access_token:
raise InvalidTokenError("access token not found")
if not access_token.Application: # should be impossible
raise InvalidClientError("Client application not found")
if not access_token.Application.isActive:
raise InvalidClientError("Client application is disabled")
if access_token.revokedAt is not None:
raise InvalidTokenError("access token has been revoked")
# Check expiration
now = datetime.now(timezone.utc)
if access_token.expiresAt < now:
raise InvalidTokenError("access token expired")
return (
OAuthAccessTokenInfo.from_db(access_token),
OAuthApplicationInfo.from_db(access_token.Application),
)
async def revoke_access_token(
token: str, application_id: str
) -> OAuthAccessTokenInfo | None:
"""
Revoke an access token.
Args:
token: The plaintext access token to revoke
application_id: The application ID making the revocation request.
Only tokens belonging to this application will be revoked.
Returns:
OAuthAccessTokenInfo if token was found and revoked, None otherwise.
Note:
Always performs exactly 2 DB queries regardless of outcome to prevent
timing side-channel attacks that could reveal token existence.
"""
try:
token_hash = _hash_token(token)
# Use update_many to filter by both token and applicationId
updated_count = await PrismaOAuthAccessToken.prisma().update_many(
where={
"token": token_hash,
"applicationId": application_id,
"revokedAt": None,
},
data={"revokedAt": datetime.now(timezone.utc)},
)
# Always perform second query to ensure constant time
result = await PrismaOAuthAccessToken.prisma().find_unique(
where={"token": token_hash}
)
# Only return result if we actually revoked something
if updated_count == 0:
return None
return OAuthAccessTokenInfo.from_db(result) if result else None
except Exception as e:
logger.exception(f"Error revoking access token: {e}")
return None
# ============================================================================
# Refresh Token Management
# ============================================================================
async def create_refresh_token(
application_id: str, user_id: str, scopes: list[APIPermission]
) -> OAuthRefreshToken:
"""
Create a new refresh token.
Returns OAuthRefreshToken (with plaintext token).
"""
plaintext_token = REFRESH_TOKEN_PREFIX + _generate_token()
token_hash = _hash_token(plaintext_token)
now = datetime.now(timezone.utc)
expires_at = now + REFRESH_TOKEN_TTL
saved_token = await PrismaOAuthRefreshToken.prisma().create(
data={
"id": str(uuid.uuid4()),
"token": token_hash, # SHA256 hash for direct lookup
"expiresAt": expires_at,
"applicationId": application_id,
"userId": user_id,
"scopes": [s for s in scopes],
}
)
return OAuthRefreshToken.from_db(saved_token, plaintext_token=plaintext_token)
async def refresh_tokens(
refresh_token: str, application_id: str
) -> tuple[OAuthAccessToken, OAuthRefreshToken]:
"""
Use a refresh token to create new access and refresh tokens.
Returns (new_access_token, new_refresh_token) both with plaintext tokens included.
Raises:
InvalidGrantError: If refresh token is invalid, expired, or revoked
"""
token_hash = _hash_token(refresh_token)
# Direct lookup by hash
rt = await PrismaOAuthRefreshToken.prisma().find_unique(where={"token": token_hash})
if not rt:
raise InvalidGrantError("refresh token not found")
# NOTE: no need to check Application.isActive, this is checked by the token endpoint
if rt.revokedAt is not None:
raise InvalidGrantError("refresh token has been revoked")
# Validate application
if rt.applicationId != application_id:
raise InvalidGrantError("refresh token does not belong to this application")
# Check expiration
now = datetime.now(timezone.utc)
if rt.expiresAt < now:
raise InvalidGrantError("refresh token expired")
# Revoke old refresh token
await PrismaOAuthRefreshToken.prisma().update(
where={"token": token_hash},
data={"revokedAt": now},
)
# Create new access and refresh tokens with same scopes
scopes = [APIPermission(s) for s in rt.scopes]
new_access_token = await create_access_token(
rt.applicationId,
rt.userId,
scopes,
)
new_refresh_token = await create_refresh_token(
rt.applicationId,
rt.userId,
scopes,
)
return new_access_token, new_refresh_token
async def revoke_refresh_token(
token: str, application_id: str
) -> OAuthRefreshTokenInfo | None:
"""
Revoke a refresh token.
Args:
token: The plaintext refresh token to revoke
application_id: The application ID making the revocation request.
Only tokens belonging to this application will be revoked.
Returns:
OAuthRefreshTokenInfo if token was found and revoked, None otherwise.
Note:
Always performs exactly 2 DB queries regardless of outcome to prevent
timing side-channel attacks that could reveal token existence.
"""
try:
token_hash = _hash_token(token)
# Use update_many to filter by both token and applicationId
updated_count = await PrismaOAuthRefreshToken.prisma().update_many(
where={
"token": token_hash,
"applicationId": application_id,
"revokedAt": None,
},
data={"revokedAt": datetime.now(timezone.utc)},
)
# Always perform second query to ensure constant time
result = await PrismaOAuthRefreshToken.prisma().find_unique(
where={"token": token_hash}
)
# Only return result if we actually revoked something
if updated_count == 0:
return None
return OAuthRefreshTokenInfo.from_db(result) if result else None
except Exception as e:
logger.exception(f"Error revoking refresh token: {e}")
return None
# ============================================================================
# Token Introspection
# ============================================================================
async def introspect_token(
token: str,
token_type_hint: Optional[Literal["access_token", "refresh_token"]] = None,
) -> TokenIntrospectionResult:
"""
Introspect a token and return its metadata (RFC 7662).
Returns TokenIntrospectionResult with active=True and metadata if valid,
or active=False if the token is invalid/expired/revoked.
"""
# Try as access token first (or if hint says "access_token")
if token_type_hint != "refresh_token":
try:
token_info, app = await validate_access_token(token)
return TokenIntrospectionResult(
active=True,
scopes=list(s.value for s in token_info.scopes),
client_id=app.client_id if app else None,
user_id=token_info.user_id,
exp=int(token_info.expires_at.timestamp()),
token_type="access_token",
)
except InvalidTokenError:
pass # Try as refresh token
# Try as refresh token
token_hash = _hash_token(token)
refresh_token = await PrismaOAuthRefreshToken.prisma().find_unique(
where={"token": token_hash}
)
if refresh_token and refresh_token.revokedAt is None:
# Check if valid (not expired)
now = datetime.now(timezone.utc)
if refresh_token.expiresAt > now:
app = await get_oauth_application_by_id(refresh_token.applicationId)
return TokenIntrospectionResult(
active=True,
scopes=list(s for s in refresh_token.scopes),
client_id=app.client_id if app else None,
user_id=refresh_token.userId,
exp=int(refresh_token.expiresAt.timestamp()),
token_type="refresh_token",
)
# Token not found or inactive
return TokenIntrospectionResult(active=False)
async def get_oauth_application_by_id(app_id: str) -> Optional[OAuthApplicationInfo]:
"""Get OAuth application by ID"""
app = await PrismaOAuthApplication.prisma().find_unique(where={"id": app_id})
if not app:
return None
return OAuthApplicationInfo.from_db(app)
async def list_user_oauth_applications(user_id: str) -> list[OAuthApplicationInfo]:
"""Get all OAuth applications owned by a user"""
apps = await PrismaOAuthApplication.prisma().find_many(
where={"ownerId": user_id},
order={"createdAt": "desc"},
)
return [OAuthApplicationInfo.from_db(app) for app in apps]
async def update_oauth_application(
app_id: str,
*,
owner_id: str,
is_active: Optional[bool] = None,
logo_url: Optional[str] = None,
) -> Optional[OAuthApplicationInfo]:
"""
Update OAuth application active status.
Only the owner can update their app's status.
Returns the updated app info, or None if app not found or not owned by user.
"""
# First verify ownership
app = await PrismaOAuthApplication.prisma().find_first(
where={"id": app_id, "ownerId": owner_id}
)
if not app:
return None
patch: OAuthApplicationUpdateInput = {}
if is_active is not None:
patch["isActive"] = is_active
if logo_url:
patch["logoUrl"] = logo_url
if not patch:
return OAuthApplicationInfo.from_db(app) # return unchanged
updated_app = await PrismaOAuthApplication.prisma().update(
where={"id": app_id},
data=patch,
)
return OAuthApplicationInfo.from_db(updated_app) if updated_app else None
# ============================================================================
# Token Cleanup
# ============================================================================
async def cleanup_expired_oauth_tokens() -> dict[str, int]:
"""
Delete expired OAuth tokens from the database.
This removes:
- Expired authorization codes (10 min TTL)
- Expired access tokens (1 hour TTL)
- Expired refresh tokens (30 day TTL)
Returns a dict with counts of deleted tokens by type.
"""
now = datetime.now(timezone.utc)
# Delete expired authorization codes
codes_result = await PrismaOAuthAuthorizationCode.prisma().delete_many(
where={"expiresAt": {"lt": now}}
)
# Delete expired access tokens
access_result = await PrismaOAuthAccessToken.prisma().delete_many(
where={"expiresAt": {"lt": now}}
)
# Delete expired refresh tokens
refresh_result = await PrismaOAuthRefreshToken.prisma().delete_many(
where={"expiresAt": {"lt": now}}
)
deleted = {
"authorization_codes": codes_result,
"access_tokens": access_result,
"refresh_tokens": refresh_result,
}
total = sum(deleted.values())
if total > 0:
logger.info(f"Cleaned up {total} expired OAuth tokens: {deleted}")
return deleted

View File

@@ -59,13 +59,12 @@ from backend.integrations.credentials_store import (
MODEL_COST: dict[LlmModel, int] = {
LlmModel.O3: 4,
LlmModel.O3_MINI: 2,
LlmModel.O1: 16,
LlmModel.O3_MINI: 2, # $1.10 / $4.40
LlmModel.O1: 16, # $15 / $60
LlmModel.O1_MINI: 4,
# GPT-5 models
LlmModel.GPT5_2: 6,
LlmModel.GPT5_1: 5,
LlmModel.GPT5: 2,
LlmModel.GPT5_1: 5,
LlmModel.GPT5_MINI: 1,
LlmModel.GPT5_NANO: 1,
LlmModel.GPT5_CHAT: 5,
@@ -88,7 +87,7 @@ MODEL_COST: dict[LlmModel, int] = {
LlmModel.AIML_API_LLAMA3_3_70B: 1,
LlmModel.AIML_API_META_LLAMA_3_1_70B: 1,
LlmModel.AIML_API_LLAMA_3_2_3B: 1,
LlmModel.LLAMA3_3_70B: 1,
LlmModel.LLAMA3_3_70B: 1, # $0.59 / $0.79
LlmModel.LLAMA3_1_8B: 1,
LlmModel.OLLAMA_LLAMA3_3: 1,
LlmModel.OLLAMA_LLAMA3_2: 1,

View File

@@ -16,7 +16,6 @@ from prisma.models import CreditRefundRequest, CreditTransaction, User, UserBala
from prisma.types import CreditRefundRequestCreateInput, CreditTransactionWhereInput
from pydantic import BaseModel
from backend.api.features.admin.model import UserHistoryResponse
from backend.data.block_cost_config import BLOCK_COSTS
from backend.data.db import query_raw_with_schema
from backend.data.includes import MAX_CREDIT_REFUND_REQUESTS_FETCH
@@ -30,6 +29,7 @@ from backend.data.model import (
from backend.data.notifications import NotificationEventModel, RefundRequestData
from backend.data.user import get_user_by_id, get_user_email_by_id
from backend.notifications.notifications import queue_notification_async
from backend.server.v2.admin.model import UserHistoryResponse
from backend.util.exceptions import InsufficientBalanceError
from backend.util.feature_flag import Flag, is_feature_enabled
from backend.util.json import SafeJson, dumps
@@ -341,19 +341,6 @@ class UserCreditBase(ABC):
if result:
# UserBalance is already updated by the CTE
# Clear insufficient funds notification flags when credits are added
# so user can receive alerts again if they run out in the future.
if transaction.amount > 0 and transaction.type in [
CreditTransactionType.GRANT,
CreditTransactionType.TOP_UP,
]:
from backend.executor.manager import (
clear_insufficient_funds_notifications,
)
await clear_insufficient_funds_notifications(user_id)
return result[0]["balance"]
async def _add_transaction(
@@ -543,22 +530,6 @@ class UserCreditBase(ABC):
if result:
new_balance, tx_key = result[0]["balance"], result[0]["transactionKey"]
# UserBalance is already updated by the CTE
# Clear insufficient funds notification flags when credits are added
# so user can receive alerts again if they run out in the future.
if (
amount > 0
and is_active
and transaction_type
in [CreditTransactionType.GRANT, CreditTransactionType.TOP_UP]
):
# Lazy import to avoid circular dependency with executor.manager
from backend.executor.manager import (
clear_insufficient_funds_notifications,
)
await clear_insufficient_funds_notifications(user_id)
return new_balance, tx_key
# If no result, either user doesn't exist or insufficient balance

View File

@@ -111,7 +111,7 @@ def get_database_schema() -> str:
async def query_raw_with_schema(query_template: str, *args) -> list[dict]:
"""Execute raw SQL query with proper schema handling."""
schema = get_database_schema()
schema_prefix = f'"{schema}".' if schema != "public" else ""
schema_prefix = f"{schema}." if schema != "public" else ""
formatted_query = query_template.format(schema_prefix=schema_prefix)
import prisma as prisma_module

View File

@@ -6,14 +6,14 @@ import fastapi.exceptions
import pytest
from pytest_snapshot.plugin import Snapshot
import backend.api.features.store.model as store
from backend.api.model import CreateGraph
import backend.server.v2.store.model as store
from backend.blocks.basic import StoreValueBlock
from backend.blocks.io import AgentInputBlock, AgentOutputBlock
from backend.data.block import BlockSchema, BlockSchemaInput
from backend.data.graph import Graph, Link, Node
from backend.data.model import SchemaField
from backend.data.user import DEFAULT_USER_ID
from backend.server.model import CreateGraph
from backend.usecases.sample import create_test_user
from backend.util.test import SpinTestServer

View File

@@ -13,7 +13,7 @@ from prisma.models import PendingHumanReview
from prisma.types import PendingHumanReviewUpdateInput
from pydantic import BaseModel
from backend.api.features.executions.review.model import (
from backend.server.v2.executions.review.model import (
PendingHumanReviewModel,
SafeJsonData,
)

View File

@@ -23,7 +23,7 @@ from backend.util.exceptions import NotFoundError
from backend.util.json import SafeJson
if TYPE_CHECKING:
from backend.api.features.library.model import LibraryAgentPreset
from backend.server.v2.library.model import LibraryAgentPreset
from .db import BaseDbModel
from .graph import NodeModel
@@ -79,7 +79,7 @@ class WebhookWithRelations(Webhook):
# integrations.py → library/model.py → integrations.py (for Webhook)
# Runtime import is used in WebhookWithRelations.from_db() method instead
# Import at runtime to avoid circular dependency
from backend.api.features.library.model import LibraryAgentPreset
from backend.server.v2.library.model import LibraryAgentPreset
return WebhookWithRelations(
**Webhook.from_db(webhook).model_dump(),
@@ -285,8 +285,8 @@ async def unlink_webhook_from_graph(
user_id: The ID of the user (for authorization)
"""
# Avoid circular imports
from backend.api.features.library.db import set_preset_webhook
from backend.data.graph import set_node_webhook
from backend.server.v2.library.db import set_preset_webhook
# Find all nodes in this graph that use this webhook
nodes = await AgentNode.prisma().find_many(

View File

@@ -4,8 +4,8 @@ from typing import AsyncGenerator
from pydantic import BaseModel, field_serializer
from backend.api.model import NotificationPayload
from backend.data.event_bus import AsyncRedisEventBus
from backend.server.model import NotificationPayload
from backend.util.settings import Settings

View File

@@ -9,8 +9,6 @@ from prisma.enums import OnboardingStep
from prisma.models import UserOnboarding
from prisma.types import UserOnboardingCreateInput, UserOnboardingUpdateInput
from backend.api.features.store.model import StoreAgentDetails
from backend.api.model import OnboardingNotificationPayload
from backend.data import execution as execution_db
from backend.data.credit import get_user_credit_model
from backend.data.notification_bus import (
@@ -18,6 +16,8 @@ from backend.data.notification_bus import (
NotificationEvent,
)
from backend.data.user import get_user_by_id
from backend.server.model import OnboardingNotificationPayload
from backend.server.v2.store.model import StoreAgentDetails
from backend.util.cache import cached
from backend.util.json import SafeJson
from backend.util.timezone_utils import get_user_timezone_or_utc
@@ -442,8 +442,6 @@ async def get_recommended_agents(user_id: str) -> list[StoreAgentDetails]:
runs=agent.runs,
rating=agent.rating,
versions=agent.versions,
agentGraphVersions=agent.agentGraphVersions,
agentGraphId=agent.agentGraphId,
last_updated=agent.updated_at,
)
for agent in recommended_agents

View File

@@ -1,429 +0,0 @@
"""Data models and access layer for user business understanding."""
import logging
from datetime import datetime
from typing import Any, Optional, cast
import pydantic
from prisma.models import UserBusinessUnderstanding
from prisma.types import (
UserBusinessUnderstandingCreateInput,
UserBusinessUnderstandingUpdateInput,
)
from backend.data.redis_client import get_redis_async
from backend.util.json import SafeJson
logger = logging.getLogger(__name__)
# Cache configuration
CACHE_KEY_PREFIX = "understanding"
CACHE_TTL_SECONDS = 48 * 60 * 60 # 48 hours
def _cache_key(user_id: str) -> str:
"""Generate cache key for user business understanding."""
return f"{CACHE_KEY_PREFIX}:{user_id}"
def _json_to_list(value: Any) -> list[str]:
"""Convert Json field to list[str], handling None."""
if value is None:
return []
if isinstance(value, list):
return cast(list[str], value)
return []
class BusinessUnderstandingInput(pydantic.BaseModel):
"""Input model for updating business understanding - all fields optional for incremental updates."""
# User info
user_name: Optional[str] = pydantic.Field(None, description="The user's name")
job_title: Optional[str] = pydantic.Field(None, description="The user's job title")
# Business basics
business_name: Optional[str] = pydantic.Field(
None, description="Name of the user's business"
)
industry: Optional[str] = pydantic.Field(None, description="Industry or sector")
business_size: Optional[str] = pydantic.Field(
None, description="Company size (e.g., '1-10', '11-50')"
)
user_role: Optional[str] = pydantic.Field(
None,
description="User's role in the organization (e.g., 'decision maker', 'implementer')",
)
# Processes & activities
key_workflows: Optional[list[str]] = pydantic.Field(
None, description="Key business workflows"
)
daily_activities: Optional[list[str]] = pydantic.Field(
None, description="Daily activities performed"
)
# Pain points & goals
pain_points: Optional[list[str]] = pydantic.Field(
None, description="Current pain points"
)
bottlenecks: Optional[list[str]] = pydantic.Field(
None, description="Process bottlenecks"
)
manual_tasks: Optional[list[str]] = pydantic.Field(
None, description="Manual/repetitive tasks"
)
automation_goals: Optional[list[str]] = pydantic.Field(
None, description="Desired automation goals"
)
# Current tools
current_software: Optional[list[str]] = pydantic.Field(
None, description="Software/tools currently used"
)
existing_automation: Optional[list[str]] = pydantic.Field(
None, description="Existing automations"
)
# Additional context
additional_notes: Optional[str] = pydantic.Field(
None, description="Any additional context"
)
class BusinessUnderstanding(pydantic.BaseModel):
"""Full business understanding model returned from database."""
id: str
user_id: str
created_at: datetime
updated_at: datetime
# User info
user_name: Optional[str] = None
job_title: Optional[str] = None
# Business basics
business_name: Optional[str] = None
industry: Optional[str] = None
business_size: Optional[str] = None
user_role: Optional[str] = None
# Processes & activities
key_workflows: list[str] = pydantic.Field(default_factory=list)
daily_activities: list[str] = pydantic.Field(default_factory=list)
# Pain points & goals
pain_points: list[str] = pydantic.Field(default_factory=list)
bottlenecks: list[str] = pydantic.Field(default_factory=list)
manual_tasks: list[str] = pydantic.Field(default_factory=list)
automation_goals: list[str] = pydantic.Field(default_factory=list)
# Current tools
current_software: list[str] = pydantic.Field(default_factory=list)
existing_automation: list[str] = pydantic.Field(default_factory=list)
# Additional context
additional_notes: Optional[str] = None
@classmethod
def from_db(cls, db_record: UserBusinessUnderstanding) -> "BusinessUnderstanding":
"""Convert database record to Pydantic model."""
return cls(
id=db_record.id,
user_id=db_record.userId,
created_at=db_record.createdAt,
updated_at=db_record.updatedAt,
user_name=db_record.userName,
job_title=db_record.jobTitle,
business_name=db_record.businessName,
industry=db_record.industry,
business_size=db_record.businessSize,
user_role=db_record.userRole,
key_workflows=_json_to_list(db_record.keyWorkflows),
daily_activities=_json_to_list(db_record.dailyActivities),
pain_points=_json_to_list(db_record.painPoints),
bottlenecks=_json_to_list(db_record.bottlenecks),
manual_tasks=_json_to_list(db_record.manualTasks),
automation_goals=_json_to_list(db_record.automationGoals),
current_software=_json_to_list(db_record.currentSoftware),
existing_automation=_json_to_list(db_record.existingAutomation),
additional_notes=db_record.additionalNotes,
)
def _merge_lists(existing: list | None, new: list | None) -> list | None:
"""Merge two lists, removing duplicates while preserving order."""
if new is None:
return existing
if existing is None:
return new
# Preserve order, add new items that don't exist
merged = list(existing)
for item in new:
if item not in merged:
merged.append(item)
return merged
async def _get_from_cache(user_id: str) -> Optional[BusinessUnderstanding]:
"""Get business understanding from Redis cache."""
try:
redis = await get_redis_async()
cached_data = await redis.get(_cache_key(user_id))
if cached_data:
return BusinessUnderstanding.model_validate_json(cached_data)
except Exception as e:
logger.warning(f"Failed to get understanding from cache: {e}")
return None
async def _set_cache(user_id: str, understanding: BusinessUnderstanding) -> None:
"""Set business understanding in Redis cache with TTL."""
try:
redis = await get_redis_async()
await redis.setex(
_cache_key(user_id),
CACHE_TTL_SECONDS,
understanding.model_dump_json(),
)
except Exception as e:
logger.warning(f"Failed to set understanding in cache: {e}")
async def _delete_cache(user_id: str) -> None:
"""Delete business understanding from Redis cache."""
try:
redis = await get_redis_async()
await redis.delete(_cache_key(user_id))
except Exception as e:
logger.warning(f"Failed to delete understanding from cache: {e}")
async def get_business_understanding(
user_id: str,
) -> Optional[BusinessUnderstanding]:
"""Get the business understanding for a user.
Checks cache first, falls back to database if not cached.
Results are cached for 48 hours.
"""
# Try cache first
cached = await _get_from_cache(user_id)
if cached:
logger.debug(f"Business understanding cache hit for user {user_id}")
return cached
# Cache miss - load from database
logger.debug(f"Business understanding cache miss for user {user_id}")
record = await UserBusinessUnderstanding.prisma().find_unique(
where={"userId": user_id}
)
if record is None:
return None
understanding = BusinessUnderstanding.from_db(record)
# Store in cache for next time
await _set_cache(user_id, understanding)
return understanding
async def upsert_business_understanding(
user_id: str,
data: BusinessUnderstandingInput,
) -> BusinessUnderstanding:
"""
Create or update business understanding with incremental merge strategy.
- String fields: new value overwrites if provided (not None)
- List fields: new items are appended to existing (deduplicated)
"""
# Get existing record for merge
existing = await UserBusinessUnderstanding.prisma().find_unique(
where={"userId": user_id}
)
# Build update data with merge strategy
update_data: UserBusinessUnderstandingUpdateInput = {}
create_data: dict[str, Any] = {"userId": user_id}
# String fields - overwrite if provided
if data.user_name is not None:
update_data["userName"] = data.user_name
create_data["userName"] = data.user_name
if data.job_title is not None:
update_data["jobTitle"] = data.job_title
create_data["jobTitle"] = data.job_title
if data.business_name is not None:
update_data["businessName"] = data.business_name
create_data["businessName"] = data.business_name
if data.industry is not None:
update_data["industry"] = data.industry
create_data["industry"] = data.industry
if data.business_size is not None:
update_data["businessSize"] = data.business_size
create_data["businessSize"] = data.business_size
if data.user_role is not None:
update_data["userRole"] = data.user_role
create_data["userRole"] = data.user_role
if data.additional_notes is not None:
update_data["additionalNotes"] = data.additional_notes
create_data["additionalNotes"] = data.additional_notes
# List fields - merge with existing
if data.key_workflows is not None:
existing_list = _json_to_list(existing.keyWorkflows) if existing else None
merged = _merge_lists(existing_list, data.key_workflows)
update_data["keyWorkflows"] = SafeJson(merged)
create_data["keyWorkflows"] = SafeJson(merged)
if data.daily_activities is not None:
existing_list = _json_to_list(existing.dailyActivities) if existing else None
merged = _merge_lists(existing_list, data.daily_activities)
update_data["dailyActivities"] = SafeJson(merged)
create_data["dailyActivities"] = SafeJson(merged)
if data.pain_points is not None:
existing_list = _json_to_list(existing.painPoints) if existing else None
merged = _merge_lists(existing_list, data.pain_points)
update_data["painPoints"] = SafeJson(merged)
create_data["painPoints"] = SafeJson(merged)
if data.bottlenecks is not None:
existing_list = _json_to_list(existing.bottlenecks) if existing else None
merged = _merge_lists(existing_list, data.bottlenecks)
update_data["bottlenecks"] = SafeJson(merged)
create_data["bottlenecks"] = SafeJson(merged)
if data.manual_tasks is not None:
existing_list = _json_to_list(existing.manualTasks) if existing else None
merged = _merge_lists(existing_list, data.manual_tasks)
update_data["manualTasks"] = SafeJson(merged)
create_data["manualTasks"] = SafeJson(merged)
if data.automation_goals is not None:
existing_list = _json_to_list(existing.automationGoals) if existing else None
merged = _merge_lists(existing_list, data.automation_goals)
update_data["automationGoals"] = SafeJson(merged)
create_data["automationGoals"] = SafeJson(merged)
if data.current_software is not None:
existing_list = _json_to_list(existing.currentSoftware) if existing else None
merged = _merge_lists(existing_list, data.current_software)
update_data["currentSoftware"] = SafeJson(merged)
create_data["currentSoftware"] = SafeJson(merged)
if data.existing_automation is not None:
existing_list = _json_to_list(existing.existingAutomation) if existing else None
merged = _merge_lists(existing_list, data.existing_automation)
update_data["existingAutomation"] = SafeJson(merged)
create_data["existingAutomation"] = SafeJson(merged)
# Upsert
record = await UserBusinessUnderstanding.prisma().upsert(
where={"userId": user_id},
data={
"create": UserBusinessUnderstandingCreateInput(**create_data),
"update": update_data,
},
)
understanding = BusinessUnderstanding.from_db(record)
# Update cache with new understanding
await _set_cache(user_id, understanding)
return understanding
async def clear_business_understanding(user_id: str) -> bool:
"""Clear/delete business understanding for a user from both DB and cache."""
# Delete from cache first
await _delete_cache(user_id)
try:
await UserBusinessUnderstanding.prisma().delete(where={"userId": user_id})
return True
except Exception:
# Record might not exist
return False
def format_understanding_for_prompt(understanding: BusinessUnderstanding) -> str:
"""Format business understanding as text for system prompt injection."""
sections = []
# User info section
user_info = []
if understanding.user_name:
user_info.append(f"Name: {understanding.user_name}")
if understanding.job_title:
user_info.append(f"Job Title: {understanding.job_title}")
if user_info:
sections.append("## User\n" + "\n".join(user_info))
# Business section
business_info = []
if understanding.business_name:
business_info.append(f"Company: {understanding.business_name}")
if understanding.industry:
business_info.append(f"Industry: {understanding.industry}")
if understanding.business_size:
business_info.append(f"Size: {understanding.business_size}")
if understanding.user_role:
business_info.append(f"Role Context: {understanding.user_role}")
if business_info:
sections.append("## Business\n" + "\n".join(business_info))
# Processes section
processes = []
if understanding.key_workflows:
processes.append(f"Key Workflows: {', '.join(understanding.key_workflows)}")
if understanding.daily_activities:
processes.append(
f"Daily Activities: {', '.join(understanding.daily_activities)}"
)
if processes:
sections.append("## Processes\n" + "\n".join(processes))
# Pain points section
pain_points = []
if understanding.pain_points:
pain_points.append(f"Pain Points: {', '.join(understanding.pain_points)}")
if understanding.bottlenecks:
pain_points.append(f"Bottlenecks: {', '.join(understanding.bottlenecks)}")
if understanding.manual_tasks:
pain_points.append(f"Manual Tasks: {', '.join(understanding.manual_tasks)}")
if pain_points:
sections.append("## Pain Points\n" + "\n".join(pain_points))
# Goals section
if understanding.automation_goals:
sections.append(
"## Automation Goals\n"
+ "\n".join(f"- {goal}" for goal in understanding.automation_goals)
)
# Current tools section
tools_info = []
if understanding.current_software:
tools_info.append(
f"Current Software: {', '.join(understanding.current_software)}"
)
if understanding.existing_automation:
tools_info.append(
f"Existing Automation: {', '.join(understanding.existing_automation)}"
)
if tools_info:
sections.append("## Current Tools\n" + "\n".join(tools_info))
# Additional notes
if understanding.additional_notes:
sections.append(f"## Additional Context\n{understanding.additional_notes}")
if not sections:
return ""
return "# User Business Context\n\n" + "\n\n".join(sections)

View File

@@ -2,11 +2,6 @@ import logging
from contextlib import asynccontextmanager
from typing import TYPE_CHECKING, Callable, Concatenate, ParamSpec, TypeVar, cast
from backend.api.features.library.db import (
add_store_agent_to_library,
list_library_agents,
)
from backend.api.features.store.db import get_store_agent_details, get_store_agents
from backend.data import db
from backend.data.analytics import (
get_accuracy_trends_and_alerts,
@@ -66,6 +61,8 @@ from backend.data.user import (
get_user_notification_preference,
update_user_integrations,
)
from backend.server.v2.library.db import add_store_agent_to_library, list_library_agents
from backend.server.v2.store.db import get_store_agent_details, get_store_agents
from backend.util.service import (
AppService,
AppServiceClient,

View File

@@ -48,8 +48,27 @@ from backend.data.notifications import (
ZeroBalanceData,
)
from backend.data.rabbitmq import SyncRabbitMQ
from backend.executor.activity_status_generator import (
generate_activity_status_for_execution,
)
from backend.executor.utils import (
GRACEFUL_SHUTDOWN_TIMEOUT_SECONDS,
GRAPH_EXECUTION_CANCEL_QUEUE_NAME,
GRAPH_EXECUTION_EXCHANGE,
GRAPH_EXECUTION_QUEUE_NAME,
GRAPH_EXECUTION_ROUTING_KEY,
CancelExecutionEvent,
ExecutionOutputEntry,
LogMetadata,
NodeExecutionProgress,
block_usage_cost,
create_execution_queue_config,
execution_usage_cost,
validate_exec,
)
from backend.integrations.creds_manager import IntegrationCredentialsManager
from backend.notifications.notifications import queue_notification
from backend.server.v2.AutoMod.manager import automod_manager
from backend.util import json
from backend.util.clients import (
get_async_execution_event_bus,
@@ -76,24 +95,7 @@ from backend.util.retry import (
)
from backend.util.settings import Settings
from .activity_status_generator import generate_activity_status_for_execution
from .automod.manager import automod_manager
from .cluster_lock import ClusterLock
from .utils import (
GRACEFUL_SHUTDOWN_TIMEOUT_SECONDS,
GRAPH_EXECUTION_CANCEL_QUEUE_NAME,
GRAPH_EXECUTION_EXCHANGE,
GRAPH_EXECUTION_QUEUE_NAME,
GRAPH_EXECUTION_ROUTING_KEY,
CancelExecutionEvent,
ExecutionOutputEntry,
LogMetadata,
NodeExecutionProgress,
block_usage_cost,
create_execution_queue_config,
execution_usage_cost,
validate_exec,
)
if TYPE_CHECKING:
from backend.executor import DatabaseManagerAsyncClient, DatabaseManagerClient
@@ -114,40 +116,6 @@ utilization_gauge = Gauge(
"Ratio of active graph runs to max graph workers",
)
# Redis key prefix for tracking insufficient funds Discord notifications.
# We only send one notification per user per agent until they top up credits.
INSUFFICIENT_FUNDS_NOTIFIED_PREFIX = "insufficient_funds_discord_notified"
# TTL for the notification flag (30 days) - acts as a fallback cleanup
INSUFFICIENT_FUNDS_NOTIFIED_TTL_SECONDS = 30 * 24 * 60 * 60
async def clear_insufficient_funds_notifications(user_id: str) -> int:
"""
Clear all insufficient funds notification flags for a user.
This should be called when a user tops up their credits, allowing
Discord notifications to be sent again if they run out of funds.
Args:
user_id: The user ID to clear notifications for.
Returns:
The number of keys that were deleted.
"""
try:
redis_client = await redis.get_redis_async()
pattern = f"{INSUFFICIENT_FUNDS_NOTIFIED_PREFIX}:{user_id}:*"
keys = [key async for key in redis_client.scan_iter(match=pattern)]
if keys:
return await redis_client.delete(*keys)
return 0
except Exception as e:
logger.warning(
f"Failed to clear insufficient funds notification flags for user "
f"{user_id}: {e}"
)
return 0
# Thread-local storage for ExecutionProcessor instances
_tls = threading.local()
@@ -1295,40 +1263,12 @@ class ExecutionProcessor:
graph_id: str,
e: InsufficientBalanceError,
):
# Check if we've already sent a notification for this user+agent combo.
# We only send one notification per user per agent until they top up credits.
redis_key = f"{INSUFFICIENT_FUNDS_NOTIFIED_PREFIX}:{user_id}:{graph_id}"
try:
redis_client = redis.get_redis()
# SET NX returns True only if the key was newly set (didn't exist)
is_new_notification = redis_client.set(
redis_key,
"1",
nx=True,
ex=INSUFFICIENT_FUNDS_NOTIFIED_TTL_SECONDS,
)
if not is_new_notification:
# Already notified for this user+agent, skip all notifications
logger.debug(
f"Skipping duplicate insufficient funds notification for "
f"user={user_id}, graph={graph_id}"
)
return
except Exception as redis_error:
# If Redis fails, log and continue to send the notification
# (better to occasionally duplicate than to never notify)
logger.warning(
f"Failed to check/set insufficient funds notification flag in Redis: "
f"{redis_error}"
)
shortfall = abs(e.amount) - e.balance
metadata = db_client.get_graph_metadata(graph_id)
base_url = (
settings.config.frontend_base_url or settings.config.platform_base_url
)
# Queue user email notification
queue_notification(
NotificationEventModel(
user_id=user_id,
@@ -1342,7 +1282,6 @@ class ExecutionProcessor:
)
)
# Send Discord system alert
try:
user_email = db_client.get_user_email_by_id(user_id)

View File

@@ -1,560 +0,0 @@
from unittest.mock import AsyncMock, MagicMock, patch
import pytest
from prisma.enums import NotificationType
from backend.data.notifications import ZeroBalanceData
from backend.executor.manager import (
INSUFFICIENT_FUNDS_NOTIFIED_PREFIX,
ExecutionProcessor,
clear_insufficient_funds_notifications,
)
from backend.util.exceptions import InsufficientBalanceError
from backend.util.test import SpinTestServer
async def async_iter(items):
"""Helper to create an async iterator from a list."""
for item in items:
yield item
@pytest.mark.asyncio(loop_scope="session")
async def test_handle_insufficient_funds_sends_discord_alert_first_time(
server: SpinTestServer,
):
"""Test that the first insufficient funds notification sends a Discord alert."""
execution_processor = ExecutionProcessor()
user_id = "test-user-123"
graph_id = "test-graph-456"
error = InsufficientBalanceError(
message="Insufficient balance",
user_id=user_id,
balance=72, # $0.72
amount=-714, # Attempting to spend $7.14
)
with patch(
"backend.executor.manager.queue_notification"
) as mock_queue_notif, patch(
"backend.executor.manager.get_notification_manager_client"
) as mock_get_client, patch(
"backend.executor.manager.settings"
) as mock_settings, patch(
"backend.executor.manager.redis"
) as mock_redis_module:
# Setup mocks
mock_client = MagicMock()
mock_get_client.return_value = mock_client
mock_settings.config.frontend_base_url = "https://test.com"
# Mock Redis to simulate first-time notification (set returns True)
mock_redis_client = MagicMock()
mock_redis_module.get_redis.return_value = mock_redis_client
mock_redis_client.set.return_value = True # Key was newly set
# Create mock database client
mock_db_client = MagicMock()
mock_graph_metadata = MagicMock()
mock_graph_metadata.name = "Test Agent"
mock_db_client.get_graph_metadata.return_value = mock_graph_metadata
mock_db_client.get_user_email_by_id.return_value = "test@example.com"
# Test the insufficient funds handler
execution_processor._handle_insufficient_funds_notif(
db_client=mock_db_client,
user_id=user_id,
graph_id=graph_id,
e=error,
)
# Verify notification was queued
mock_queue_notif.assert_called_once()
notification_call = mock_queue_notif.call_args[0][0]
assert notification_call.type == NotificationType.ZERO_BALANCE
assert notification_call.user_id == user_id
assert isinstance(notification_call.data, ZeroBalanceData)
assert notification_call.data.current_balance == 72
# Verify Redis was checked with correct key pattern
expected_key = f"{INSUFFICIENT_FUNDS_NOTIFIED_PREFIX}:{user_id}:{graph_id}"
mock_redis_client.set.assert_called_once()
call_args = mock_redis_client.set.call_args
assert call_args[0][0] == expected_key
assert call_args[1]["nx"] is True
# Verify Discord alert was sent
mock_client.discord_system_alert.assert_called_once()
discord_message = mock_client.discord_system_alert.call_args[0][0]
assert "Insufficient Funds Alert" in discord_message
assert "test@example.com" in discord_message
assert "Test Agent" in discord_message
@pytest.mark.asyncio(loop_scope="session")
async def test_handle_insufficient_funds_skips_duplicate_notifications(
server: SpinTestServer,
):
"""Test that duplicate insufficient funds notifications skip both email and Discord."""
execution_processor = ExecutionProcessor()
user_id = "test-user-123"
graph_id = "test-graph-456"
error = InsufficientBalanceError(
message="Insufficient balance",
user_id=user_id,
balance=72,
amount=-714,
)
with patch(
"backend.executor.manager.queue_notification"
) as mock_queue_notif, patch(
"backend.executor.manager.get_notification_manager_client"
) as mock_get_client, patch(
"backend.executor.manager.settings"
) as mock_settings, patch(
"backend.executor.manager.redis"
) as mock_redis_module:
# Setup mocks
mock_client = MagicMock()
mock_get_client.return_value = mock_client
mock_settings.config.frontend_base_url = "https://test.com"
# Mock Redis to simulate duplicate notification (set returns False/None)
mock_redis_client = MagicMock()
mock_redis_module.get_redis.return_value = mock_redis_client
mock_redis_client.set.return_value = None # Key already existed
# Create mock database client
mock_db_client = MagicMock()
mock_db_client.get_graph_metadata.return_value = MagicMock(name="Test Agent")
# Test the insufficient funds handler
execution_processor._handle_insufficient_funds_notif(
db_client=mock_db_client,
user_id=user_id,
graph_id=graph_id,
e=error,
)
# Verify email notification was NOT queued (deduplication worked)
mock_queue_notif.assert_not_called()
# Verify Discord alert was NOT sent (deduplication worked)
mock_client.discord_system_alert.assert_not_called()
@pytest.mark.asyncio(loop_scope="session")
async def test_handle_insufficient_funds_different_agents_get_separate_alerts(
server: SpinTestServer,
):
"""Test that different agents for the same user get separate Discord alerts."""
execution_processor = ExecutionProcessor()
user_id = "test-user-123"
graph_id_1 = "test-graph-111"
graph_id_2 = "test-graph-222"
error = InsufficientBalanceError(
message="Insufficient balance",
user_id=user_id,
balance=72,
amount=-714,
)
with patch("backend.executor.manager.queue_notification"), patch(
"backend.executor.manager.get_notification_manager_client"
) as mock_get_client, patch(
"backend.executor.manager.settings"
) as mock_settings, patch(
"backend.executor.manager.redis"
) as mock_redis_module:
mock_client = MagicMock()
mock_get_client.return_value = mock_client
mock_settings.config.frontend_base_url = "https://test.com"
mock_redis_client = MagicMock()
mock_redis_module.get_redis.return_value = mock_redis_client
# Both calls return True (first time for each agent)
mock_redis_client.set.return_value = True
mock_db_client = MagicMock()
mock_graph_metadata = MagicMock()
mock_graph_metadata.name = "Test Agent"
mock_db_client.get_graph_metadata.return_value = mock_graph_metadata
mock_db_client.get_user_email_by_id.return_value = "test@example.com"
# First agent notification
execution_processor._handle_insufficient_funds_notif(
db_client=mock_db_client,
user_id=user_id,
graph_id=graph_id_1,
e=error,
)
# Second agent notification
execution_processor._handle_insufficient_funds_notif(
db_client=mock_db_client,
user_id=user_id,
graph_id=graph_id_2,
e=error,
)
# Verify Discord alerts were sent for both agents
assert mock_client.discord_system_alert.call_count == 2
# Verify Redis was called with different keys
assert mock_redis_client.set.call_count == 2
calls = mock_redis_client.set.call_args_list
assert (
calls[0][0][0]
== f"{INSUFFICIENT_FUNDS_NOTIFIED_PREFIX}:{user_id}:{graph_id_1}"
)
assert (
calls[1][0][0]
== f"{INSUFFICIENT_FUNDS_NOTIFIED_PREFIX}:{user_id}:{graph_id_2}"
)
@pytest.mark.asyncio(loop_scope="session")
async def test_clear_insufficient_funds_notifications(server: SpinTestServer):
"""Test that clearing notifications removes all keys for a user."""
user_id = "test-user-123"
with patch("backend.executor.manager.redis") as mock_redis_module:
mock_redis_client = MagicMock()
# get_redis_async is an async function, so we need AsyncMock for it
mock_redis_module.get_redis_async = AsyncMock(return_value=mock_redis_client)
# Mock scan_iter to return some keys as an async iterator
mock_keys = [
f"{INSUFFICIENT_FUNDS_NOTIFIED_PREFIX}:{user_id}:graph-1",
f"{INSUFFICIENT_FUNDS_NOTIFIED_PREFIX}:{user_id}:graph-2",
f"{INSUFFICIENT_FUNDS_NOTIFIED_PREFIX}:{user_id}:graph-3",
]
mock_redis_client.scan_iter.return_value = async_iter(mock_keys)
# delete is awaited, so use AsyncMock
mock_redis_client.delete = AsyncMock(return_value=3)
# Clear notifications
result = await clear_insufficient_funds_notifications(user_id)
# Verify correct pattern was used
expected_pattern = f"{INSUFFICIENT_FUNDS_NOTIFIED_PREFIX}:{user_id}:*"
mock_redis_client.scan_iter.assert_called_once_with(match=expected_pattern)
# Verify delete was called with all keys
mock_redis_client.delete.assert_called_once_with(*mock_keys)
# Verify return value
assert result == 3
@pytest.mark.asyncio(loop_scope="session")
async def test_clear_insufficient_funds_notifications_no_keys(server: SpinTestServer):
"""Test clearing notifications when there are no keys to clear."""
user_id = "test-user-no-notifications"
with patch("backend.executor.manager.redis") as mock_redis_module:
mock_redis_client = MagicMock()
# get_redis_async is an async function, so we need AsyncMock for it
mock_redis_module.get_redis_async = AsyncMock(return_value=mock_redis_client)
# Mock scan_iter to return no keys as an async iterator
mock_redis_client.scan_iter.return_value = async_iter([])
# Clear notifications
result = await clear_insufficient_funds_notifications(user_id)
# Verify delete was not called
mock_redis_client.delete.assert_not_called()
# Verify return value
assert result == 0
@pytest.mark.asyncio(loop_scope="session")
async def test_clear_insufficient_funds_notifications_handles_redis_error(
server: SpinTestServer,
):
"""Test that clearing notifications handles Redis errors gracefully."""
user_id = "test-user-redis-error"
with patch("backend.executor.manager.redis") as mock_redis_module:
# Mock get_redis_async to raise an error
mock_redis_module.get_redis_async = AsyncMock(
side_effect=Exception("Redis connection failed")
)
# Clear notifications should not raise, just return 0
result = await clear_insufficient_funds_notifications(user_id)
# Verify it returned 0 (graceful failure)
assert result == 0
@pytest.mark.asyncio(loop_scope="session")
async def test_handle_insufficient_funds_continues_on_redis_error(
server: SpinTestServer,
):
"""Test that both email and Discord notifications are still sent when Redis fails."""
execution_processor = ExecutionProcessor()
user_id = "test-user-123"
graph_id = "test-graph-456"
error = InsufficientBalanceError(
message="Insufficient balance",
user_id=user_id,
balance=72,
amount=-714,
)
with patch(
"backend.executor.manager.queue_notification"
) as mock_queue_notif, patch(
"backend.executor.manager.get_notification_manager_client"
) as mock_get_client, patch(
"backend.executor.manager.settings"
) as mock_settings, patch(
"backend.executor.manager.redis"
) as mock_redis_module:
mock_client = MagicMock()
mock_get_client.return_value = mock_client
mock_settings.config.frontend_base_url = "https://test.com"
# Mock Redis to raise an error
mock_redis_client = MagicMock()
mock_redis_module.get_redis.return_value = mock_redis_client
mock_redis_client.set.side_effect = Exception("Redis connection error")
mock_db_client = MagicMock()
mock_graph_metadata = MagicMock()
mock_graph_metadata.name = "Test Agent"
mock_db_client.get_graph_metadata.return_value = mock_graph_metadata
mock_db_client.get_user_email_by_id.return_value = "test@example.com"
# Test the insufficient funds handler
execution_processor._handle_insufficient_funds_notif(
db_client=mock_db_client,
user_id=user_id,
graph_id=graph_id,
e=error,
)
# Verify email notification was still queued despite Redis error
mock_queue_notif.assert_called_once()
# Verify Discord alert was still sent despite Redis error
mock_client.discord_system_alert.assert_called_once()
@pytest.mark.asyncio(loop_scope="session")
async def test_add_transaction_clears_notifications_on_grant(server: SpinTestServer):
"""Test that _add_transaction clears notification flags when adding GRANT credits."""
from prisma.enums import CreditTransactionType
from backend.data.credit import UserCredit
user_id = "test-user-grant-clear"
with patch("backend.data.credit.query_raw_with_schema") as mock_query, patch(
"backend.executor.manager.redis"
) as mock_redis_module:
# Mock the query to return a successful transaction
mock_query.return_value = [{"balance": 1000, "transactionKey": "test-tx-key"}]
# Mock async Redis for notification clearing
mock_redis_client = MagicMock()
mock_redis_module.get_redis_async = AsyncMock(return_value=mock_redis_client)
mock_redis_client.scan_iter.return_value = async_iter(
[f"{INSUFFICIENT_FUNDS_NOTIFIED_PREFIX}:{user_id}:graph-1"]
)
mock_redis_client.delete = AsyncMock(return_value=1)
# Create a concrete instance
credit_model = UserCredit()
# Call _add_transaction with GRANT type (should clear notifications)
await credit_model._add_transaction(
user_id=user_id,
amount=500, # Positive amount
transaction_type=CreditTransactionType.GRANT,
is_active=True, # Active transaction
)
# Verify notification clearing was called
mock_redis_module.get_redis_async.assert_called_once()
mock_redis_client.scan_iter.assert_called_once_with(
match=f"{INSUFFICIENT_FUNDS_NOTIFIED_PREFIX}:{user_id}:*"
)
@pytest.mark.asyncio(loop_scope="session")
async def test_add_transaction_clears_notifications_on_top_up(server: SpinTestServer):
"""Test that _add_transaction clears notification flags when adding TOP_UP credits."""
from prisma.enums import CreditTransactionType
from backend.data.credit import UserCredit
user_id = "test-user-topup-clear"
with patch("backend.data.credit.query_raw_with_schema") as mock_query, patch(
"backend.executor.manager.redis"
) as mock_redis_module:
# Mock the query to return a successful transaction
mock_query.return_value = [{"balance": 2000, "transactionKey": "test-tx-key-2"}]
# Mock async Redis for notification clearing
mock_redis_client = MagicMock()
mock_redis_module.get_redis_async = AsyncMock(return_value=mock_redis_client)
mock_redis_client.scan_iter.return_value = async_iter([])
mock_redis_client.delete = AsyncMock(return_value=0)
credit_model = UserCredit()
# Call _add_transaction with TOP_UP type (should clear notifications)
await credit_model._add_transaction(
user_id=user_id,
amount=1000, # Positive amount
transaction_type=CreditTransactionType.TOP_UP,
is_active=True,
)
# Verify notification clearing was attempted
mock_redis_module.get_redis_async.assert_called_once()
@pytest.mark.asyncio(loop_scope="session")
async def test_add_transaction_skips_clearing_for_inactive_transaction(
server: SpinTestServer,
):
"""Test that _add_transaction does NOT clear notifications for inactive transactions."""
from prisma.enums import CreditTransactionType
from backend.data.credit import UserCredit
user_id = "test-user-inactive"
with patch("backend.data.credit.query_raw_with_schema") as mock_query, patch(
"backend.executor.manager.redis"
) as mock_redis_module:
# Mock the query to return a successful transaction
mock_query.return_value = [{"balance": 500, "transactionKey": "test-tx-key-3"}]
# Mock async Redis
mock_redis_client = MagicMock()
mock_redis_module.get_redis_async = AsyncMock(return_value=mock_redis_client)
credit_model = UserCredit()
# Call _add_transaction with is_active=False (should NOT clear notifications)
await credit_model._add_transaction(
user_id=user_id,
amount=500,
transaction_type=CreditTransactionType.TOP_UP,
is_active=False, # Inactive - pending Stripe payment
)
# Verify notification clearing was NOT called
mock_redis_module.get_redis_async.assert_not_called()
@pytest.mark.asyncio(loop_scope="session")
async def test_add_transaction_skips_clearing_for_usage_transaction(
server: SpinTestServer,
):
"""Test that _add_transaction does NOT clear notifications for USAGE transactions."""
from prisma.enums import CreditTransactionType
from backend.data.credit import UserCredit
user_id = "test-user-usage"
with patch("backend.data.credit.query_raw_with_schema") as mock_query, patch(
"backend.executor.manager.redis"
) as mock_redis_module:
# Mock the query to return a successful transaction
mock_query.return_value = [{"balance": 400, "transactionKey": "test-tx-key-4"}]
# Mock async Redis
mock_redis_client = MagicMock()
mock_redis_module.get_redis_async = AsyncMock(return_value=mock_redis_client)
credit_model = UserCredit()
# Call _add_transaction with USAGE type (spending, should NOT clear)
await credit_model._add_transaction(
user_id=user_id,
amount=-100, # Negative - spending credits
transaction_type=CreditTransactionType.USAGE,
is_active=True,
)
# Verify notification clearing was NOT called
mock_redis_module.get_redis_async.assert_not_called()
@pytest.mark.asyncio(loop_scope="session")
async def test_enable_transaction_clears_notifications(server: SpinTestServer):
"""Test that _enable_transaction clears notification flags when enabling a TOP_UP."""
from prisma.enums import CreditTransactionType
from backend.data.credit import UserCredit
user_id = "test-user-enable"
with patch("backend.data.credit.CreditTransaction") as mock_credit_tx, patch(
"backend.data.credit.query_raw_with_schema"
) as mock_query, patch("backend.executor.manager.redis") as mock_redis_module:
# Mock finding the pending transaction
mock_transaction = MagicMock()
mock_transaction.amount = 1000
mock_transaction.type = CreditTransactionType.TOP_UP
mock_credit_tx.prisma.return_value.find_first = AsyncMock(
return_value=mock_transaction
)
# Mock the query to return updated balance
mock_query.return_value = [{"balance": 1500}]
# Mock async Redis for notification clearing
mock_redis_client = MagicMock()
mock_redis_module.get_redis_async = AsyncMock(return_value=mock_redis_client)
mock_redis_client.scan_iter.return_value = async_iter(
[f"{INSUFFICIENT_FUNDS_NOTIFIED_PREFIX}:{user_id}:graph-1"]
)
mock_redis_client.delete = AsyncMock(return_value=1)
credit_model = UserCredit()
# Call _enable_transaction (simulates Stripe checkout completion)
from backend.util.json import SafeJson
await credit_model._enable_transaction(
transaction_key="cs_test_123",
user_id=user_id,
metadata=SafeJson({"payment": "completed"}),
)
# Verify notification clearing was called
mock_redis_module.get_redis_async.assert_called_once()
mock_redis_client.scan_iter.assert_called_once_with(
match=f"{INSUFFICIENT_FUNDS_NOTIFIED_PREFIX}:{user_id}:*"
)

View File

@@ -3,16 +3,16 @@ import logging
import fastapi.responses
import pytest
import backend.api.features.library.model
import backend.api.features.store.model
from backend.api.model import CreateGraph
from backend.api.rest_api import AgentServer
import backend.server.v2.library.model
import backend.server.v2.store.model
from backend.blocks.basic import StoreValueBlock
from backend.blocks.data_manipulation import FindInDictionaryBlock
from backend.blocks.io import AgentInputBlock
from backend.blocks.maths import CalculatorBlock, Operation
from backend.data import execution, graph
from backend.data.model import User
from backend.server.model import CreateGraph
from backend.server.rest_api import AgentServer
from backend.usecases.sample import create_test_graph, create_test_user
from backend.util.test import SpinTestServer, wait_execution
@@ -356,7 +356,7 @@ async def test_execute_preset(server: SpinTestServer):
test_graph = await create_graph(server, test_graph, test_user)
# Create preset with initial values
preset = backend.api.features.library.model.LibraryAgentPresetCreatable(
preset = backend.server.v2.library.model.LibraryAgentPresetCreatable(
name="Test Preset With Clash",
description="Test preset with clashing input values",
graph_id=test_graph.id,
@@ -444,7 +444,7 @@ async def test_execute_preset_with_clash(server: SpinTestServer):
test_graph = await create_graph(server, test_graph, test_user)
# Create preset with initial values
preset = backend.api.features.library.model.LibraryAgentPresetCreatable(
preset = backend.server.v2.library.model.LibraryAgentPresetCreatable(
name="Test Preset With Clash",
description="Test preset with clashing input values",
graph_id=test_graph.id,
@@ -485,7 +485,7 @@ async def test_store_listing_graph(server: SpinTestServer):
test_user = await create_test_user()
test_graph = await create_graph(server, create_test_graph(), test_user)
store_submission_request = backend.api.features.store.model.StoreSubmissionRequest(
store_submission_request = backend.server.v2.store.model.StoreSubmissionRequest(
agent_id=test_graph.id,
agent_version=test_graph.version,
slug=test_graph.id,
@@ -514,7 +514,7 @@ async def test_store_listing_graph(server: SpinTestServer):
admin_user = await create_test_user(alt_user=True)
await server.agent_server.test_review_store_listing(
backend.api.features.store.model.ReviewSubmissionRequest(
backend.server.v2.store.model.ReviewSubmissionRequest(
store_listing_version_id=slv_id,
is_approved=True,
comments="Test comments",
@@ -523,7 +523,7 @@ async def test_store_listing_graph(server: SpinTestServer):
)
# Add the approved store listing to the admin user's library so they can execute it
from backend.api.features.library.db import add_store_agent_to_library
from backend.server.v2.library.db import add_store_agent_to_library
await add_store_agent_to_library(
store_listing_version_id=slv_id, user_id=admin_user.id

View File

@@ -23,7 +23,6 @@ from dotenv import load_dotenv
from pydantic import BaseModel, Field, ValidationError
from sqlalchemy import MetaData, create_engine
from backend.data.auth.oauth import cleanup_expired_oauth_tokens
from backend.data.block import BlockInput
from backend.data.execution import GraphExecutionWithNodes
from backend.data.model import CredentialsMetaInput
@@ -243,12 +242,6 @@ def cleanup_expired_files():
run_async(cleanup_expired_files_async())
def cleanup_oauth_tokens():
"""Clean up expired OAuth tokens from the database."""
# Wait for completion
run_async(cleanup_expired_oauth_tokens())
def execution_accuracy_alerts():
"""Check execution accuracy and send alerts if drops are detected."""
return report_execution_accuracy_alerts()
@@ -453,17 +446,6 @@ class Scheduler(AppService):
jobstore=Jobstores.EXECUTION.value,
)
# OAuth Token Cleanup - configurable interval
self.scheduler.add_job(
cleanup_oauth_tokens,
id="cleanup_oauth_tokens",
trigger="interval",
replace_existing=True,
seconds=config.oauth_token_cleanup_interval_hours
* 3600, # Convert hours to seconds
jobstore=Jobstores.EXECUTION.value,
)
# Execution Accuracy Monitoring - configurable interval
self.scheduler.add_job(
execution_accuracy_alerts,
@@ -622,11 +604,6 @@ class Scheduler(AppService):
"""Manually trigger cleanup of expired cloud storage files."""
return cleanup_expired_files()
@expose
def execute_cleanup_oauth_tokens(self):
"""Manually trigger cleanup of expired OAuth tokens."""
return cleanup_oauth_tokens()
@expose
def execute_report_execution_accuracy_alerts(self):
"""Manually trigger execution accuracy alert checking."""

View File

@@ -1,7 +1,7 @@
import pytest
from backend.api.model import CreateGraph
from backend.data import db
from backend.server.model import CreateGraph
from backend.usecases.sample import create_test_graph, create_test_user
from backend.util.clients import get_scheduler_client
from backend.util.test import SpinTestServer

View File

@@ -0,0 +1,157 @@
"""
Embedding service for generating text embeddings using OpenAI.
Used for vector-based semantic search in the store.
"""
import functools
import logging
from typing import Optional
import openai
from backend.util.settings import Settings
logger = logging.getLogger(__name__)
# Model configuration
# Using text-embedding-3-small (1536 dimensions) for compatibility with pgvector indexes
# pgvector IVFFlat/HNSW indexes have dimension limits (2000 for IVFFlat, varies for HNSW)
EMBEDDING_MODEL = "text-embedding-3-small"
EMBEDDING_DIMENSIONS = 1536
# Input validation limits
# OpenAI text-embedding-3-large supports up to 8191 tokens (~32k chars)
# We set a conservative limit to prevent abuse
MAX_TEXT_LENGTH = 10000 # characters
MAX_BATCH_SIZE = 100 # maximum texts per batch request
class EmbeddingService:
"""Service for generating text embeddings using OpenAI.
The service can be created without an API key - the key is validated
only when the client property is first accessed. This allows the service
to be instantiated at module load time without requiring configuration.
"""
def __init__(self, api_key: Optional[str] = None):
settings = Settings()
self.api_key = (
api_key
or settings.secrets.openai_internal_api_key
or settings.secrets.openai_api_key
)
@functools.cached_property
def client(self) -> openai.AsyncOpenAI:
"""Lazily create the OpenAI client, raising if no API key is configured."""
if not self.api_key:
raise ValueError(
"OpenAI API key not configured. "
"Set OPENAI_API_KEY or OPENAI_INTERNAL_API_KEY environment variable."
)
return openai.AsyncOpenAI(api_key=self.api_key)
async def generate_embedding(self, text: str) -> list[float]:
"""
Generate embedding for a single text string.
Args:
text: The text to generate an embedding for.
Returns:
A list of floats representing the embedding vector.
Raises:
ValueError: If the text is empty or exceeds maximum length.
openai.APIError: If the OpenAI API call fails.
"""
# Input validation
if not text or not text.strip():
raise ValueError("Text cannot be empty")
if len(text) > MAX_TEXT_LENGTH:
raise ValueError(
f"Text exceeds maximum length of {MAX_TEXT_LENGTH} characters"
)
response = await self.client.embeddings.create(
model=EMBEDDING_MODEL,
input=text,
dimensions=EMBEDDING_DIMENSIONS,
)
if not response.data:
raise ValueError("OpenAI API returned empty embedding data")
return response.data[0].embedding
async def generate_embeddings(self, texts: list[str]) -> list[list[float]]:
"""
Generate embeddings for multiple texts (batch).
Args:
texts: List of texts to generate embeddings for.
Returns:
List of embedding vectors, one per input text.
Raises:
ValueError: If any text is invalid or batch size exceeds limit.
openai.APIError: If the OpenAI API call fails.
"""
# Input validation
if not texts:
raise ValueError("Texts list cannot be empty")
if len(texts) > MAX_BATCH_SIZE:
raise ValueError(f"Batch size exceeds maximum of {MAX_BATCH_SIZE} texts")
for i, text in enumerate(texts):
if not text or not text.strip():
raise ValueError(f"Text at index {i} cannot be empty")
if len(text) > MAX_TEXT_LENGTH:
raise ValueError(
f"Text at index {i} exceeds maximum length of {MAX_TEXT_LENGTH} characters"
)
response = await self.client.embeddings.create(
model=EMBEDDING_MODEL,
input=texts,
dimensions=EMBEDDING_DIMENSIONS,
)
# Sort by index to ensure correct ordering
sorted_data = sorted(response.data, key=lambda x: x.index)
return [item.embedding for item in sorted_data]
def create_search_text(name: str, sub_heading: str, description: str) -> str:
"""
Combine fields into searchable text for embedding.
This creates a single text string from the agent's name, sub-heading,
and description, which is then converted to an embedding vector.
Args:
name: The agent name.
sub_heading: The agent sub-heading/tagline.
description: The agent description.
Returns:
A single string combining all non-empty fields.
"""
parts = [name or "", sub_heading or "", description or ""]
# filter(None, parts) removes empty strings since empty string is falsy
return " ".join(filter(None, parts)).strip()
@functools.cache
def get_embedding_service() -> EmbeddingService:
"""
Get or create the embedding service singleton.
Uses functools.cache for thread-safe lazy initialization.
Returns:
The shared EmbeddingService instance.
Raises:
ValueError: If OpenAI API key is not configured (when generating embeddings).
"""
return EmbeddingService()

View File

@@ -0,0 +1,235 @@
"""Tests for the embedding service.
This module tests:
- create_search_text utility function
- EmbeddingService input validation
- EmbeddingService API interaction (mocked)
"""
from unittest.mock import AsyncMock, MagicMock, patch
import pytest
from backend.integrations.embeddings import (
EMBEDDING_DIMENSIONS,
MAX_BATCH_SIZE,
MAX_TEXT_LENGTH,
EmbeddingService,
create_search_text,
)
class TestCreateSearchText:
"""Tests for the create_search_text utility function."""
def test_combines_all_fields(self):
result = create_search_text("Agent Name", "A cool agent", "Does amazing things")
assert result == "Agent Name A cool agent Does amazing things"
def test_handles_empty_name(self):
result = create_search_text("", "Sub heading", "Description")
assert result == "Sub heading Description"
def test_handles_empty_sub_heading(self):
result = create_search_text("Name", "", "Description")
assert result == "Name Description"
def test_handles_empty_description(self):
result = create_search_text("Name", "Sub heading", "")
assert result == "Name Sub heading"
def test_handles_all_empty(self):
result = create_search_text("", "", "")
assert result == ""
def test_handles_none_values(self):
# The function expects strings but should handle None gracefully
result = create_search_text(None, None, None) # type: ignore
assert result == ""
def test_preserves_content_strips_outer_whitespace(self):
# The function joins parts and strips the outer result
# Internal whitespace in each part is preserved
result = create_search_text(" Name ", " Sub ", " Desc ")
# Each part is joined with space, then outer strip applied
assert result == "Name Sub Desc"
def test_handles_only_whitespace(self):
# Parts that are only whitespace become empty after filter
result = create_search_text(" ", " ", " ")
assert result == ""
class TestEmbeddingServiceValidation:
"""Tests for EmbeddingService input validation."""
@pytest.fixture
def mock_settings(self):
"""Mock settings with a test API key."""
with patch("backend.integrations.embeddings.Settings") as mock:
mock_instance = MagicMock()
mock_instance.secrets.openai_internal_api_key = "test-api-key"
mock_instance.secrets.openai_api_key = ""
mock.return_value = mock_instance
yield mock
@pytest.fixture
def service(self, mock_settings):
"""Create an EmbeddingService instance with mocked settings."""
service = EmbeddingService()
# Inject a mock client by setting the cached_property directly
service.__dict__["client"] = MagicMock()
return service
def test_client_access_requires_api_key(self):
"""Test that accessing client fails without an API key."""
with patch("backend.integrations.embeddings.Settings") as mock:
mock_instance = MagicMock()
mock_instance.secrets.openai_internal_api_key = ""
mock_instance.secrets.openai_api_key = ""
mock.return_value = mock_instance
# Service creation should succeed
service = EmbeddingService()
# But accessing client should fail
with pytest.raises(ValueError, match="OpenAI API key not configured"):
_ = service.client
def test_init_accepts_explicit_api_key(self):
"""Test that explicit API key overrides settings."""
with patch("backend.integrations.embeddings.Settings") as mock:
mock_instance = MagicMock()
mock_instance.secrets.openai_internal_api_key = ""
mock_instance.secrets.openai_api_key = ""
mock.return_value = mock_instance
with patch("backend.integrations.embeddings.openai.AsyncOpenAI"):
service = EmbeddingService(api_key="explicit-key")
assert service.api_key == "explicit-key"
@pytest.mark.asyncio
async def test_generate_embedding_empty_text(self, service):
"""Test that empty text raises ValueError."""
with pytest.raises(ValueError, match="Text cannot be empty"):
await service.generate_embedding("")
@pytest.mark.asyncio
async def test_generate_embedding_whitespace_only(self, service):
"""Test that whitespace-only text raises ValueError."""
with pytest.raises(ValueError, match="Text cannot be empty"):
await service.generate_embedding(" ")
@pytest.mark.asyncio
async def test_generate_embedding_exceeds_max_length(self, service):
"""Test that text exceeding max length raises ValueError."""
long_text = "a" * (MAX_TEXT_LENGTH + 1)
with pytest.raises(ValueError, match="exceeds maximum length"):
await service.generate_embedding(long_text)
@pytest.mark.asyncio
async def test_generate_embeddings_empty_list(self, service):
"""Test that empty list raises ValueError."""
with pytest.raises(ValueError, match="Texts list cannot be empty"):
await service.generate_embeddings([])
@pytest.mark.asyncio
async def test_generate_embeddings_exceeds_batch_size(self, service):
"""Test that batch exceeding max size raises ValueError."""
texts = ["text"] * (MAX_BATCH_SIZE + 1)
with pytest.raises(ValueError, match="Batch size exceeds maximum"):
await service.generate_embeddings(texts)
@pytest.mark.asyncio
async def test_generate_embeddings_empty_text_in_batch(self, service):
"""Test that empty text in batch raises ValueError with index."""
with pytest.raises(ValueError, match="Text at index 1 cannot be empty"):
await service.generate_embeddings(["valid", "", "also valid"])
@pytest.mark.asyncio
async def test_generate_embeddings_long_text_in_batch(self, service):
"""Test that long text in batch raises ValueError with index."""
long_text = "a" * (MAX_TEXT_LENGTH + 1)
with pytest.raises(ValueError, match="Text at index 2 exceeds maximum length"):
await service.generate_embeddings(["short", "also short", long_text])
class TestEmbeddingServiceAPI:
"""Tests for EmbeddingService API interaction."""
@pytest.fixture
def mock_openai_client(self):
"""Create a mock OpenAI client."""
mock_client = MagicMock()
mock_client.embeddings = MagicMock()
return mock_client
@pytest.fixture
def service_with_mock_client(self, mock_openai_client):
"""Create an EmbeddingService with a mocked OpenAI client."""
with patch("backend.integrations.embeddings.Settings") as mock_settings:
mock_instance = MagicMock()
mock_instance.secrets.openai_internal_api_key = "test-key"
mock_instance.secrets.openai_api_key = ""
mock_settings.return_value = mock_instance
service = EmbeddingService()
# Inject mock client by setting the cached_property directly
service.__dict__["client"] = mock_openai_client
return service, mock_openai_client
@pytest.mark.asyncio
async def test_generate_embedding_success(self, service_with_mock_client):
"""Test successful embedding generation."""
service, mock_client = service_with_mock_client
# Create mock response
mock_embedding = [0.1] * EMBEDDING_DIMENSIONS
mock_response = MagicMock()
mock_response.data = [MagicMock(embedding=mock_embedding)]
mock_client.embeddings.create = AsyncMock(return_value=mock_response)
result = await service.generate_embedding("test text")
assert result == mock_embedding
mock_client.embeddings.create.assert_called_once()
@pytest.mark.asyncio
async def test_generate_embeddings_success(self, service_with_mock_client):
"""Test successful batch embedding generation."""
service, mock_client = service_with_mock_client
# Create mock response with multiple embeddings
mock_embeddings = [[0.1] * EMBEDDING_DIMENSIONS, [0.2] * EMBEDDING_DIMENSIONS]
mock_response = MagicMock()
mock_response.data = [
MagicMock(embedding=mock_embeddings[0], index=0),
MagicMock(embedding=mock_embeddings[1], index=1),
]
mock_client.embeddings.create = AsyncMock(return_value=mock_response)
result = await service.generate_embeddings(["text1", "text2"])
assert result == mock_embeddings
mock_client.embeddings.create.assert_called_once()
@pytest.mark.asyncio
async def test_generate_embeddings_preserves_order(self, service_with_mock_client):
"""Test that batch embeddings are returned in correct order even if API returns out of order."""
service, mock_client = service_with_mock_client
# Create mock response with embeddings out of order
mock_embeddings = [[0.1] * EMBEDDING_DIMENSIONS, [0.2] * EMBEDDING_DIMENSIONS]
mock_response = MagicMock()
# Return in reverse order
mock_response.data = [
MagicMock(embedding=mock_embeddings[1], index=1),
MagicMock(embedding=mock_embeddings[0], index=0),
]
mock_client.embeddings.create = AsyncMock(return_value=mock_response)
result = await service.generate_embeddings(["text1", "text2"])
# Should be sorted by index
assert result[0] == mock_embeddings[0]
assert result[1] == mock_embeddings[1]

View File

@@ -149,10 +149,10 @@ async def setup_webhook_for_block(
async def migrate_legacy_triggered_graphs():
from prisma.models import AgentGraph
from backend.api.features.library.db import create_preset
from backend.api.features.library.model import LibraryAgentPresetCreatable
from backend.data.graph import AGENT_GRAPH_INCLUDE, GraphModel, set_node_webhook
from backend.data.model import is_credentials_field_name
from backend.server.v2.library.db import create_preset
from backend.server.v2.library.model import LibraryAgentPresetCreatable
triggered_graphs = [
GraphModel.from_db(_graph)

View File

@@ -1,5 +1,5 @@
from backend.api.rest_api import AgentServer
from backend.app import run_processes
from backend.server.rest_api import AgentServer
def main():

View File

@@ -3,12 +3,12 @@ from typing import Dict, Set
from fastapi import WebSocket
from backend.api.model import NotificationPayload, WSMessage, WSMethod
from backend.data.execution import (
ExecutionEventType,
GraphExecutionEvent,
NodeExecutionEvent,
)
from backend.server.model import NotificationPayload, WSMessage, WSMethod
_EVENT_TYPE_TO_METHOD_MAP: dict[ExecutionEventType, WSMethod] = {
ExecutionEventType.GRAPH_EXEC_UPDATE: WSMethod.GRAPH_EXECUTION_EVENT,

View File

@@ -4,13 +4,13 @@ from unittest.mock import AsyncMock
import pytest
from fastapi import WebSocket
from backend.api.conn_manager import ConnectionManager
from backend.api.model import NotificationPayload, WSMessage, WSMethod
from backend.data.execution import (
ExecutionStatus,
GraphExecutionEvent,
NodeExecutionEvent,
)
from backend.server.conn_manager import ConnectionManager
from backend.server.model import NotificationPayload, WSMessage, WSMethod
@pytest.fixture

View File

@@ -0,0 +1,29 @@
from fastapi import FastAPI
from backend.monitoring.instrumentation import instrument_fastapi
from backend.server.middleware.security import SecurityHeadersMiddleware
from .routes.integrations import integrations_router
from .routes.tools import tools_router
from .routes.v1 import v1_router
external_app = FastAPI(
title="AutoGPT External API",
description="External API for AutoGPT integrations",
docs_url="/docs",
version="1.0",
)
external_app.add_middleware(SecurityHeadersMiddleware)
external_app.include_router(v1_router, prefix="/v1")
external_app.include_router(tools_router, prefix="/v1")
external_app.include_router(integrations_router, prefix="/v1")
# Add Prometheus instrumentation
instrument_fastapi(
external_app,
service_name="external-api",
expose_endpoint=True,
endpoint="/metrics",
include_in_schema=True,
)

View File

@@ -0,0 +1,36 @@
from fastapi import HTTPException, Security
from fastapi.security import APIKeyHeader
from prisma.enums import APIKeyPermission
from backend.data.api_key import APIKeyInfo, has_permission, validate_api_key
api_key_header = APIKeyHeader(name="X-API-Key", auto_error=False)
async def require_api_key(api_key: str | None = Security(api_key_header)) -> APIKeyInfo:
"""Base middleware for API key authentication"""
if api_key is None:
raise HTTPException(status_code=401, detail="Missing API key")
api_key_obj = await validate_api_key(api_key)
if not api_key_obj:
raise HTTPException(status_code=401, detail="Invalid API key")
return api_key_obj
def require_permission(permission: APIKeyPermission):
"""Dependency function for checking specific permissions"""
async def check_permission(
api_key: APIKeyInfo = Security(require_api_key),
) -> APIKeyInfo:
if not has_permission(api_key, permission):
raise HTTPException(
status_code=403,
detail=f"API key lacks the required permission '{permission}'",
)
return api_key
return check_permission

View File

@@ -16,9 +16,7 @@ from fastapi import APIRouter, Body, HTTPException, Path, Security, status
from prisma.enums import APIKeyPermission
from pydantic import BaseModel, Field, SecretStr
from backend.api.external.middleware import require_permission
from backend.api.features.integrations.models import get_all_provider_names
from backend.data.auth.base import APIAuthorizationInfo
from backend.data.api_key import APIKeyInfo
from backend.data.model import (
APIKeyCredentials,
Credentials,
@@ -30,6 +28,8 @@ from backend.data.model import (
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.server.external.middleware import require_permission
from backend.server.integrations.models import get_all_provider_names
from backend.util.settings import Settings
if TYPE_CHECKING:
@@ -255,7 +255,7 @@ def _get_oauth_handler_for_external(
@integrations_router.get("/providers", response_model=list[ProviderInfo])
async def list_providers(
auth: APIAuthorizationInfo = Security(
api_key: APIKeyInfo = Security(
require_permission(APIKeyPermission.READ_INTEGRATIONS)
),
) -> list[ProviderInfo]:
@@ -319,7 +319,7 @@ async def list_providers(
async def initiate_oauth(
provider: Annotated[str, Path(title="The OAuth provider")],
request: OAuthInitiateRequest,
auth: APIAuthorizationInfo = Security(
api_key: APIKeyInfo = Security(
require_permission(APIKeyPermission.MANAGE_INTEGRATIONS)
),
) -> OAuthInitiateResponse:
@@ -337,10 +337,7 @@ async def initiate_oauth(
if not validate_callback_url(request.callback_url):
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=(
f"Callback URL origin is not allowed. "
f"Allowed origins: {settings.config.external_oauth_callback_origins}",
),
detail=f"Callback URL origin is not allowed. Allowed origins: {settings.config.external_oauth_callback_origins}",
)
# Validate provider
@@ -362,15 +359,13 @@ async def initiate_oauth(
)
# Store state token with external flow metadata
# Note: initiated_by_api_key_id is only available for API key auth, not OAuth
api_key_id = getattr(auth, "id", None) if auth.type == "api_key" else None
state_token, code_challenge = await creds_manager.store.store_state_token(
user_id=auth.user_id,
user_id=api_key.user_id,
provider=provider if isinstance(provider_name, str) else provider_name.value,
scopes=request.scopes,
callback_url=request.callback_url,
state_metadata=request.state_metadata,
initiated_by_api_key_id=api_key_id,
initiated_by_api_key_id=api_key.id,
)
# Build login URL
@@ -398,7 +393,7 @@ async def initiate_oauth(
async def complete_oauth(
provider: Annotated[str, Path(title="The OAuth provider")],
request: OAuthCompleteRequest,
auth: APIAuthorizationInfo = Security(
api_key: APIKeyInfo = Security(
require_permission(APIKeyPermission.MANAGE_INTEGRATIONS)
),
) -> OAuthCompleteResponse:
@@ -411,7 +406,7 @@ async def complete_oauth(
"""
# Verify state token
valid_state = await creds_manager.store.verify_state_token(
auth.user_id, request.state_token, provider
api_key.user_id, request.state_token, provider
)
if not valid_state:
@@ -458,7 +453,7 @@ async def complete_oauth(
)
# Store credentials
await creds_manager.create(auth.user_id, credentials)
await creds_manager.create(api_key.user_id, credentials)
logger.info(f"Successfully completed external OAuth for provider {provider}")
@@ -475,7 +470,7 @@ async def complete_oauth(
@integrations_router.get("/credentials", response_model=list[CredentialSummary])
async def list_credentials(
auth: APIAuthorizationInfo = Security(
api_key: APIKeyInfo = Security(
require_permission(APIKeyPermission.READ_INTEGRATIONS)
),
) -> list[CredentialSummary]:
@@ -484,7 +479,7 @@ async def list_credentials(
Returns metadata about each credential without exposing sensitive tokens.
"""
credentials = await creds_manager.store.get_all_creds(auth.user_id)
credentials = await creds_manager.store.get_all_creds(api_key.user_id)
return [
CredentialSummary(
id=cred.id,
@@ -504,7 +499,7 @@ async def list_credentials(
)
async def list_credentials_by_provider(
provider: Annotated[str, Path(title="The provider to list credentials for")],
auth: APIAuthorizationInfo = Security(
api_key: APIKeyInfo = Security(
require_permission(APIKeyPermission.READ_INTEGRATIONS)
),
) -> list[CredentialSummary]:
@@ -512,7 +507,7 @@ async def list_credentials_by_provider(
List credentials for a specific provider.
"""
credentials = await creds_manager.store.get_creds_by_provider(
auth.user_id, provider
api_key.user_id, provider
)
return [
CredentialSummary(
@@ -541,7 +536,7 @@ async def create_credential(
CreateUserPasswordCredentialRequest,
CreateHostScopedCredentialRequest,
] = Body(..., discriminator="type"),
auth: APIAuthorizationInfo = Security(
api_key: APIKeyInfo = Security(
require_permission(APIKeyPermission.MANAGE_INTEGRATIONS)
),
) -> CreateCredentialResponse:
@@ -596,7 +591,7 @@ async def create_credential(
# Store credentials
try:
await creds_manager.create(auth.user_id, credentials)
await creds_manager.create(api_key.user_id, credentials)
except Exception as e:
logger.error(f"Failed to store credentials: {e}")
raise HTTPException(
@@ -628,7 +623,7 @@ class DeleteCredentialResponse(BaseModel):
async def delete_credential(
provider: Annotated[str, Path(title="The provider")],
cred_id: Annotated[str, Path(title="The credential ID to delete")],
auth: APIAuthorizationInfo = Security(
api_key: APIKeyInfo = Security(
require_permission(APIKeyPermission.DELETE_INTEGRATIONS)
),
) -> DeleteCredentialResponse:
@@ -639,7 +634,7 @@ async def delete_credential(
use the main API's delete endpoint which handles webhook cleanup and
token revocation.
"""
creds = await creds_manager.store.get_creds_by_id(auth.user_id, cred_id)
creds = await creds_manager.store.get_creds_by_id(api_key.user_id, cred_id)
if not creds:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND, detail="Credentials not found"
@@ -650,6 +645,6 @@ async def delete_credential(
detail="Credentials do not match the specified provider",
)
await creds_manager.delete(auth.user_id, cred_id)
await creds_manager.delete(api_key.user_id, cred_id)
return DeleteCredentialResponse(deleted=True, credentials_id=cred_id)

View File

@@ -14,19 +14,19 @@ from fastapi import APIRouter, Security
from prisma.enums import APIKeyPermission
from pydantic import BaseModel, Field
from backend.api.external.middleware import require_permission
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
from backend.data.api_key import APIKeyInfo
from backend.server.external.middleware import require_permission
from backend.server.v2.chat.model import ChatSession
from backend.server.v2.chat.tools import find_agent_tool, run_agent_tool
from backend.server.v2.chat.tools.models import ToolResponseBase
logger = logging.getLogger(__name__)
tools_router = APIRouter(prefix="/tools", tags=["tools"])
# Note: We use Security() as a function parameter dependency (auth: APIAuthorizationInfo = Security(...))
# Note: We use Security() as a function parameter dependency (api_key: APIKeyInfo = Security(...))
# rather than in the decorator's dependencies= list. This avoids duplicate permission checks
# while still enforcing auth AND giving us access to auth for extracting user_id.
# while still enforcing auth AND giving us access to the api_key for extracting user_id.
# Request models
@@ -80,9 +80,7 @@ def _create_ephemeral_session(user_id: str | None) -> ChatSession:
)
async def find_agent(
request: FindAgentRequest,
auth: APIAuthorizationInfo = Security(
require_permission(APIKeyPermission.USE_TOOLS)
),
api_key: APIKeyInfo = Security(require_permission(APIKeyPermission.USE_TOOLS)),
) -> dict[str, Any]:
"""
Search for agents in the marketplace based on capabilities and user needs.
@@ -93,9 +91,9 @@ async def find_agent(
Returns:
List of matching agents or no results response
"""
session = _create_ephemeral_session(auth.user_id)
session = _create_ephemeral_session(api_key.user_id)
result = await find_agent_tool._execute(
user_id=auth.user_id,
user_id=api_key.user_id,
session=session,
query=request.query,
)
@@ -107,9 +105,7 @@ async def find_agent(
)
async def run_agent(
request: RunAgentRequest,
auth: APIAuthorizationInfo = Security(
require_permission(APIKeyPermission.USE_TOOLS)
),
api_key: APIKeyInfo = Security(require_permission(APIKeyPermission.USE_TOOLS)),
) -> dict[str, Any]:
"""
Run or schedule an agent from the marketplace.
@@ -133,9 +129,9 @@ async def run_agent(
- execution_started: If agent was run or scheduled successfully
- error: If something went wrong
"""
session = _create_ephemeral_session(auth.user_id)
session = _create_ephemeral_session(api_key.user_id)
result = await run_agent_tool._execute(
user_id=auth.user_id,
user_id=api_key.user_id,
session=session,
username_agent_slug=request.username_agent_slug,
inputs=request.inputs,

View File

@@ -5,60 +5,46 @@ from typing import Annotated, Any, Literal, Optional, Sequence
from fastapi import APIRouter, Body, HTTPException, Security
from prisma.enums import AgentExecutionStatus, APIKeyPermission
from pydantic import BaseModel, Field
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.data.block
from backend.api.external.middleware import require_permission
import backend.server.v2.store.cache as store_cache
import backend.server.v2.store.model as store_model
from backend.data import execution as execution_db
from backend.data import graph as graph_db
from backend.data import user as user_db
from backend.data.auth.base import APIAuthorizationInfo
from backend.data.api_key import APIKeyInfo
from backend.data.block import BlockInput, CompletedBlockOutput
from backend.executor.utils import add_graph_execution
from backend.server.external.middleware import require_permission
from backend.util.settings import Settings
from .integrations import integrations_router
from .tools import tools_router
settings = Settings()
logger = logging.getLogger(__name__)
v1_router = APIRouter()
v1_router.include_router(integrations_router)
v1_router.include_router(tools_router)
class NodeOutput(TypedDict):
key: str
value: Any
class UserInfoResponse(BaseModel):
id: str
name: Optional[str]
email: str
timezone: str = Field(
description="The user's last known timezone (e.g. 'Europe/Amsterdam'), "
"or 'not-set' if not set"
)
class ExecutionNode(TypedDict):
node_id: str
input: Any
output: dict[str, Any]
@v1_router.get(
path="/me",
tags=["user", "meta"],
)
async def get_user_info(
auth: APIAuthorizationInfo = Security(
require_permission(APIKeyPermission.IDENTITY)
),
) -> UserInfoResponse:
user = await user_db.get_user_by_id(auth.user_id)
class ExecutionNodeOutput(TypedDict):
node_id: str
outputs: list[NodeOutput]
return UserInfoResponse(
id=user.id,
name=user.name,
email=user.email,
timezone=user.timezone,
)
class GraphExecutionResult(TypedDict):
execution_id: str
status: str
nodes: list[ExecutionNode]
output: Optional[list[dict[str, str]]]
@v1_router.get(
@@ -79,9 +65,7 @@ async def get_graph_blocks() -> Sequence[dict[Any, Any]]:
async def execute_graph_block(
block_id: str,
data: BlockInput,
auth: APIAuthorizationInfo = Security(
require_permission(APIKeyPermission.EXECUTE_BLOCK)
),
api_key: APIKeyInfo = Security(require_permission(APIKeyPermission.EXECUTE_BLOCK)),
) -> CompletedBlockOutput:
obj = backend.data.block.get_block(block_id)
if not obj:
@@ -101,14 +85,12 @@ async def execute_graph(
graph_id: str,
graph_version: int,
node_input: Annotated[dict[str, Any], Body(..., embed=True, default_factory=dict)],
auth: APIAuthorizationInfo = Security(
require_permission(APIKeyPermission.EXECUTE_GRAPH)
),
api_key: APIKeyInfo = Security(require_permission(APIKeyPermission.EXECUTE_GRAPH)),
) -> dict[str, Any]:
try:
graph_exec = await add_graph_execution(
graph_id=graph_id,
user_id=auth.user_id,
user_id=api_key.user_id,
inputs=node_input,
graph_version=graph_version,
)
@@ -118,19 +100,6 @@ async def execute_graph(
raise HTTPException(status_code=400, detail=msg)
class ExecutionNode(TypedDict):
node_id: str
input: Any
output: dict[str, Any]
class GraphExecutionResult(TypedDict):
execution_id: str
status: str
nodes: list[ExecutionNode]
output: Optional[list[dict[str, str]]]
@v1_router.get(
path="/graphs/{graph_id}/executions/{graph_exec_id}/results",
tags=["graphs"],
@@ -138,12 +107,10 @@ class GraphExecutionResult(TypedDict):
async def get_graph_execution_results(
graph_id: str,
graph_exec_id: str,
auth: APIAuthorizationInfo = Security(
require_permission(APIKeyPermission.READ_GRAPH)
),
api_key: APIKeyInfo = Security(require_permission(APIKeyPermission.READ_GRAPH)),
) -> GraphExecutionResult:
graph_exec = await execution_db.get_graph_execution(
user_id=auth.user_id,
user_id=api_key.user_id,
execution_id=graph_exec_id,
include_node_executions=True,
)
@@ -155,7 +122,7 @@ async def get_graph_execution_results(
if not await graph_db.get_graph(
graph_id=graph_exec.graph_id,
version=graph_exec.graph_version,
user_id=auth.user_id,
user_id=api_key.user_id,
):
raise HTTPException(status_code=404, detail=f"Graph #{graph_id} not found.")

View File

@@ -17,8 +17,6 @@ from fastapi import (
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
from backend.api.features.library.model import LibraryAgentPreset
from backend.data.graph import NodeModel, get_graph, set_node_webhook
from backend.data.integrations import (
WebhookEvent,
@@ -47,6 +45,13 @@ 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
from backend.server.integrations.models import (
ProviderConstants,
ProviderNamesResponse,
get_all_provider_names,
)
from backend.server.v2.library.db import set_preset_webhook, update_preset
from backend.server.v2.library.model import LibraryAgentPreset
from backend.util.exceptions import (
GraphNotInLibraryError,
MissingConfigError,
@@ -55,8 +60,6 @@ from backend.util.exceptions import (
)
from backend.util.settings import Settings
from .models import ProviderConstants, ProviderNamesResponse, get_all_provider_names
if TYPE_CHECKING:
from backend.integrations.oauth import BaseOAuthHandler

View File

@@ -3,7 +3,7 @@ from fastapi import FastAPI
from fastapi.testclient import TestClient
from starlette.applications import Starlette
from backend.api.middleware.security import SecurityHeadersMiddleware
from backend.server.middleware.security import SecurityHeadersMiddleware
@pytest.fixture

View File

@@ -4,7 +4,7 @@ from typing import Any, Literal, Optional
import pydantic
from prisma.enums import OnboardingStep
from backend.data.auth.api_key import APIKeyInfo, APIKeyPermission
from backend.data.api_key import APIKeyInfo, APIKeyPermission
from backend.data.graph import Graph
from backend.util.timezone_name import TimeZoneName

View File

@@ -16,33 +16,35 @@ from fastapi.middleware.gzip import GZipMiddleware
from fastapi.routing import APIRoute
from prisma.errors import PrismaError
import backend.api.features.admin.credit_admin_routes
import backend.api.features.admin.execution_analytics_routes
import backend.api.features.admin.store_admin_routes
import backend.api.features.builder
import backend.api.features.builder.routes
import backend.api.features.chat.routes as chat_routes
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.oauth
import backend.api.features.otto.routes
import backend.api.features.postmark.postmark
import backend.api.features.store.model
import backend.api.features.store.routes
import backend.api.features.v1
import backend.data.block
import backend.data.db
import backend.data.graph
import backend.data.user
import backend.integrations.webhooks.utils
import backend.server.routers.postmark.postmark
import backend.server.routers.v1
import backend.server.v2.admin.credit_admin_routes
import backend.server.v2.admin.execution_analytics_routes
import backend.server.v2.admin.store_admin_routes
import backend.server.v2.builder
import backend.server.v2.builder.routes
import backend.server.v2.chat.routes as chat_routes
import backend.server.v2.executions.review.routes
import backend.server.v2.library.db
import backend.server.v2.library.model
import backend.server.v2.library.routes
import backend.server.v2.otto.routes
import backend.server.v2.store.model
import backend.server.v2.store.routes
import backend.util.service
import backend.util.settings
from backend.blocks.llm import DEFAULT_LLM_MODEL
from backend.blocks.llm import LlmModel
from backend.data.model import Credentials
from backend.integrations.providers import ProviderName
from backend.monitoring.instrumentation import instrument_fastapi
from backend.server.external.api import external_app
from backend.server.middleware.security import SecurityHeadersMiddleware
from backend.server.utils.cors import build_cors_params
from backend.util import json
from backend.util.cloud_storage import shutdown_cloud_storage_handler
from backend.util.exceptions import (
@@ -53,13 +55,6 @@ from backend.util.exceptions import (
from backend.util.feature_flag import initialize_launchdarkly, shutdown_launchdarkly
from backend.util.service import UnhealthyServiceError
from .external.fastapi_app import external_api
from .features.analytics import router as analytics_router
from .features.integrations.router import router as integrations_router
from .middleware.security import SecurityHeadersMiddleware
from .utils.cors import build_cors_params
from .utils.openapi import sort_openapi
settings = backend.util.settings.Settings()
logger = logging.getLogger(__name__)
@@ -113,7 +108,7 @@ async def lifespan_context(app: fastapi.FastAPI):
await backend.data.user.migrate_and_encrypt_user_integrations()
await backend.data.graph.fix_llm_provider_credentials()
await backend.data.graph.migrate_llm_models(DEFAULT_LLM_MODEL)
await backend.data.graph.migrate_llm_models(LlmModel.GPT4O)
await backend.integrations.webhooks.utils.migrate_legacy_triggered_graphs()
with launch_darkly_context():
@@ -180,9 +175,6 @@ app.add_middleware(GZipMiddleware, minimum_size=50_000) # 50KB threshold
# Add 401 responses to authenticated endpoints in OpenAPI spec
add_auth_responses_to_openapi(app)
# Sort OpenAPI schema to eliminate diff on refactors
sort_openapi(app)
# Add Prometheus instrumentation
instrument_fastapi(
app,
@@ -261,52 +253,42 @@ app.add_exception_handler(MissingConfigError, handle_internal_http_error(503))
app.add_exception_handler(ValueError, handle_internal_http_error(400))
app.add_exception_handler(Exception, handle_internal_http_error(500))
app.include_router(backend.api.features.v1.v1_router, tags=["v1"], prefix="/api")
app.include_router(backend.server.routers.v1.v1_router, tags=["v1"], prefix="/api")
app.include_router(
integrations_router,
prefix="/api/integrations",
tags=["v1", "integrations"],
backend.server.v2.store.routes.router, tags=["v2"], prefix="/api/store"
)
app.include_router(
analytics_router,
prefix="/api/analytics",
tags=["analytics"],
backend.server.v2.builder.routes.router, tags=["v2"], prefix="/api/builder"
)
app.include_router(
backend.api.features.store.routes.router, tags=["v2"], prefix="/api/store"
)
app.include_router(
backend.api.features.builder.routes.router, tags=["v2"], prefix="/api/builder"
)
app.include_router(
backend.api.features.admin.store_admin_routes.router,
backend.server.v2.admin.store_admin_routes.router,
tags=["v2", "admin"],
prefix="/api/store",
)
app.include_router(
backend.api.features.admin.credit_admin_routes.router,
backend.server.v2.admin.credit_admin_routes.router,
tags=["v2", "admin"],
prefix="/api/credits",
)
app.include_router(
backend.api.features.admin.execution_analytics_routes.router,
backend.server.v2.admin.execution_analytics_routes.router,
tags=["v2", "admin"],
prefix="/api/executions",
)
app.include_router(
backend.api.features.executions.review.routes.router,
backend.server.v2.executions.review.routes.router,
tags=["v2", "executions", "review"],
prefix="/api/review",
)
app.include_router(
backend.api.features.library.routes.router, tags=["v2"], prefix="/api/library"
backend.server.v2.library.routes.router, tags=["v2"], prefix="/api/library"
)
app.include_router(
backend.api.features.otto.routes.router, tags=["v2", "otto"], prefix="/api/otto"
backend.server.v2.otto.routes.router, tags=["v2", "otto"], prefix="/api/otto"
)
app.include_router(
backend.api.features.postmark.postmark.router,
backend.server.routers.postmark.postmark.router,
tags=["v1", "email"],
prefix="/api/email",
)
@@ -315,13 +297,8 @@ app.include_router(
tags=["v2", "chat"],
prefix="/api/chat",
)
app.include_router(
backend.api.features.oauth.router,
tags=["oauth"],
prefix="/api/oauth",
)
app.mount("/external-api", external_api)
app.mount("/external-api", external_app)
@app.get(path="/health", tags=["health"], dependencies=[])
@@ -374,7 +351,7 @@ class AgentServer(backend.util.service.AppProcess):
graph_version: Optional[int] = None,
node_input: Optional[dict[str, Any]] = None,
):
return await backend.api.features.v1.execute_graph(
return await backend.server.routers.v1.execute_graph(
user_id=user_id,
graph_id=graph_id,
graph_version=graph_version,
@@ -389,16 +366,16 @@ class AgentServer(backend.util.service.AppProcess):
user_id: str,
for_export: bool = False,
):
return await backend.api.features.v1.get_graph(
return await backend.server.routers.v1.get_graph(
graph_id, user_id, graph_version, for_export
)
@staticmethod
async def test_create_graph(
create_graph: backend.api.features.v1.CreateGraph,
create_graph: backend.server.routers.v1.CreateGraph,
user_id: str,
):
return await backend.api.features.v1.create_new_graph(create_graph, user_id)
return await backend.server.routers.v1.create_new_graph(create_graph, user_id)
@staticmethod
async def test_get_graph_run_status(graph_exec_id: str, user_id: str):
@@ -414,45 +391,45 @@ class AgentServer(backend.util.service.AppProcess):
@staticmethod
async def test_delete_graph(graph_id: str, user_id: str):
"""Used for clean-up after a test run"""
await backend.api.features.library.db.delete_library_agent_by_graph_id(
await backend.server.v2.library.db.delete_library_agent_by_graph_id(
graph_id=graph_id, user_id=user_id
)
return await backend.api.features.v1.delete_graph(graph_id, user_id)
return await backend.server.routers.v1.delete_graph(graph_id, user_id)
@staticmethod
async def test_get_presets(user_id: str, page: int = 1, page_size: int = 10):
return await backend.api.features.library.routes.presets.list_presets(
return await backend.server.v2.library.routes.presets.list_presets(
user_id=user_id, page=page, page_size=page_size
)
@staticmethod
async def test_get_preset(preset_id: str, user_id: str):
return await backend.api.features.library.routes.presets.get_preset(
return await backend.server.v2.library.routes.presets.get_preset(
preset_id=preset_id, user_id=user_id
)
@staticmethod
async def test_create_preset(
preset: backend.api.features.library.model.LibraryAgentPresetCreatable,
preset: backend.server.v2.library.model.LibraryAgentPresetCreatable,
user_id: str,
):
return await backend.api.features.library.routes.presets.create_preset(
return await backend.server.v2.library.routes.presets.create_preset(
preset=preset, user_id=user_id
)
@staticmethod
async def test_update_preset(
preset_id: str,
preset: backend.api.features.library.model.LibraryAgentPresetUpdatable,
preset: backend.server.v2.library.model.LibraryAgentPresetUpdatable,
user_id: str,
):
return await backend.api.features.library.routes.presets.update_preset(
return await backend.server.v2.library.routes.presets.update_preset(
preset_id=preset_id, preset=preset, user_id=user_id
)
@staticmethod
async def test_delete_preset(preset_id: str, user_id: str):
return await backend.api.features.library.routes.presets.delete_preset(
return await backend.server.v2.library.routes.presets.delete_preset(
preset_id=preset_id, user_id=user_id
)
@@ -462,7 +439,7 @@ class AgentServer(backend.util.service.AppProcess):
user_id: str,
inputs: Optional[dict[str, Any]] = None,
):
return await backend.api.features.library.routes.presets.execute_preset(
return await backend.server.v2.library.routes.presets.execute_preset(
preset_id=preset_id,
user_id=user_id,
inputs=inputs or {},
@@ -471,20 +448,18 @@ class AgentServer(backend.util.service.AppProcess):
@staticmethod
async def test_create_store_listing(
request: backend.api.features.store.model.StoreSubmissionRequest, user_id: str
request: backend.server.v2.store.model.StoreSubmissionRequest, user_id: str
):
return await backend.api.features.store.routes.create_submission(
request, user_id
)
return await backend.server.v2.store.routes.create_submission(request, user_id)
### ADMIN ###
@staticmethod
async def test_review_store_listing(
request: backend.api.features.store.model.ReviewSubmissionRequest,
request: backend.server.v2.store.model.ReviewSubmissionRequest,
user_id: str,
):
return await backend.api.features.admin.store_admin_routes.review_submission(
return await backend.server.v2.admin.store_admin_routes.review_submission(
request.store_listing_version_id, request, user_id
)
@@ -494,7 +469,10 @@ class AgentServer(backend.util.service.AppProcess):
provider: ProviderName,
credentials: Credentials,
) -> Credentials:
from .features.integrations.router import create_credentials, get_credential
from backend.server.integrations.router import (
create_credentials,
get_credential,
)
try:
return await create_credentials(

View File

@@ -6,11 +6,10 @@ from typing import Annotated
import fastapi
import pydantic
from autogpt_libs.auth import get_user_id
from autogpt_libs.auth.dependencies import requires_user
import backend.data.analytics
router = fastapi.APIRouter(dependencies=[fastapi.Security(requires_user)])
router = fastapi.APIRouter()
logger = logging.getLogger(__name__)

View File

@@ -0,0 +1,150 @@
"""Example of analytics tests with improved error handling and assertions."""
import json
from unittest.mock import AsyncMock, Mock
import fastapi
import fastapi.testclient
import pytest
import pytest_mock
from pytest_snapshot.plugin import Snapshot
import backend.server.routers.analytics as analytics_routes
from backend.server.test_helpers import (
assert_error_response_structure,
assert_mock_called_with_partial,
assert_response_status,
safe_parse_json,
)
app = fastapi.FastAPI()
app.include_router(analytics_routes.router)
client = fastapi.testclient.TestClient(app)
@pytest.fixture(autouse=True)
def setup_app_auth(mock_jwt_user):
"""Setup auth overrides for all tests in this module"""
from autogpt_libs.auth.jwt_utils import get_jwt_payload
app.dependency_overrides[get_jwt_payload] = mock_jwt_user["get_jwt_payload"]
yield
app.dependency_overrides.clear()
def test_log_raw_metric_success_improved(
mocker: pytest_mock.MockFixture,
configured_snapshot: Snapshot,
test_user_id: str,
) -> None:
"""Test successful raw metric logging with improved assertions."""
# Mock the analytics function
mock_result = Mock(id="metric-123-uuid")
mock_log_metric = mocker.patch(
"backend.data.analytics.log_raw_metric",
new_callable=AsyncMock,
return_value=mock_result,
)
request_data = {
"metric_name": "page_load_time",
"metric_value": 2.5,
"data_string": "/dashboard",
}
response = client.post("/log_raw_metric", json=request_data)
# Improved assertions with better error messages
assert_response_status(response, 200, "Metric logging should succeed")
response_data = safe_parse_json(response, "Metric response parsing")
assert response_data == "metric-123-uuid", f"Unexpected response: {response_data}"
# Verify the function was called with correct parameters
assert_mock_called_with_partial(
mock_log_metric,
user_id=test_user_id,
metric_name="page_load_time",
metric_value=2.5,
data_string="/dashboard",
)
# Snapshot test the response
configured_snapshot.assert_match(
json.dumps({"metric_id": response_data}, indent=2, sort_keys=True),
"analytics_log_metric_success_improved",
)
def test_log_raw_metric_invalid_request_improved() -> None:
"""Test invalid metric request with improved error assertions."""
# Test missing required fields
response = client.post("/log_raw_metric", json={})
error_data = assert_error_response_structure(
response, expected_status=422, expected_error_fields=["loc", "msg", "type"]
)
# Verify specific error details
detail = error_data["detail"]
assert isinstance(detail, list), "Error detail should be a list"
assert len(detail) > 0, "Should have at least one error"
# Check that required fields are mentioned in errors
error_fields = [error["loc"][-1] for error in detail if "loc" in error]
assert "metric_name" in error_fields, "Should report missing metric_name"
assert "metric_value" in error_fields, "Should report missing metric_value"
assert "data_string" in error_fields, "Should report missing data_string"
def test_log_raw_metric_type_validation_improved(
mocker: pytest_mock.MockFixture,
) -> None:
"""Test metric type validation with improved assertions."""
# Mock the analytics function to avoid event loop issues
mocker.patch(
"backend.data.analytics.log_raw_metric",
new_callable=AsyncMock,
return_value=Mock(id="test-id"),
)
invalid_requests = [
{
"data": {
"metric_name": "test",
"metric_value": "not_a_number", # Invalid type
"data_string": "test",
},
"expected_error": "Input should be a valid number",
},
{
"data": {
"metric_name": "", # Empty string
"metric_value": 1.0,
"data_string": "test",
},
"expected_error": "String should have at least 1 character",
},
{
"data": {
"metric_name": "test",
"metric_value": 123, # Valid number
"data_string": "", # Empty data_string
},
"expected_error": "String should have at least 1 character",
},
]
for test_case in invalid_requests:
response = client.post("/log_raw_metric", json=test_case["data"])
error_data = assert_error_response_structure(response, expected_status=422)
# Check that expected error is in the response
error_text = json.dumps(error_data)
assert (
test_case["expected_error"] in error_text
or test_case["expected_error"].lower() in error_text.lower()
), f"Expected error '{test_case['expected_error']}' not found in: {error_text}"

View File

@@ -0,0 +1,115 @@
"""Example of parametrized tests for analytics endpoints."""
import json
from unittest.mock import AsyncMock, Mock
import fastapi
import fastapi.testclient
import pytest
import pytest_mock
from pytest_snapshot.plugin import Snapshot
import backend.server.routers.analytics as analytics_routes
app = fastapi.FastAPI()
app.include_router(analytics_routes.router)
client = fastapi.testclient.TestClient(app)
@pytest.fixture(autouse=True)
def setup_app_auth(mock_jwt_user):
"""Setup auth overrides for all tests in this module"""
from autogpt_libs.auth.jwt_utils import get_jwt_payload
app.dependency_overrides[get_jwt_payload] = mock_jwt_user["get_jwt_payload"]
yield
app.dependency_overrides.clear()
@pytest.mark.parametrize(
"metric_value,metric_name,data_string,test_id",
[
(100, "api_calls_count", "external_api", "integer_value"),
(0, "error_count", "no_errors", "zero_value"),
(-5.2, "temperature_delta", "cooling", "negative_value"),
(1.23456789, "precision_test", "float_precision", "float_precision"),
(999999999, "large_number", "max_value", "large_number"),
(0.0000001, "tiny_number", "min_value", "tiny_number"),
],
)
def test_log_raw_metric_values_parametrized(
mocker: pytest_mock.MockFixture,
configured_snapshot: Snapshot,
metric_value: float,
metric_name: str,
data_string: str,
test_id: str,
) -> None:
"""Test raw metric logging with various metric values using parametrize."""
# Mock the analytics function
mock_result = Mock(id=f"metric-{test_id}-uuid")
mocker.patch(
"backend.data.analytics.log_raw_metric",
new_callable=AsyncMock,
return_value=mock_result,
)
request_data = {
"metric_name": metric_name,
"metric_value": metric_value,
"data_string": data_string,
}
response = client.post("/log_raw_metric", json=request_data)
# Better error handling
assert response.status_code == 200, f"Failed for {test_id}: {response.text}"
response_data = response.json()
# Snapshot test the response
configured_snapshot.assert_match(
json.dumps(
{"metric_id": response_data, "test_case": test_id}, indent=2, sort_keys=True
),
f"analytics_metric_{test_id}",
)
@pytest.mark.parametrize(
"invalid_data,expected_error",
[
({}, "Field required"), # Missing all fields
({"metric_name": "test"}, "Field required"), # Missing metric_value
(
{"metric_name": "test", "metric_value": "not_a_number"},
"Input should be a valid number",
), # Invalid type
(
{"metric_name": "", "metric_value": 1.0, "data_string": "test"},
"String should have at least 1 character",
), # Empty name
],
)
def test_log_raw_metric_invalid_requests_parametrized(
mocker: pytest_mock.MockFixture,
invalid_data: dict,
expected_error: str,
) -> None:
"""Test invalid metric requests with parametrize."""
# Mock the analytics function to avoid event loop issues
mocker.patch(
"backend.data.analytics.log_raw_metric",
new_callable=AsyncMock,
return_value=Mock(id="test-id"),
)
response = client.post("/log_raw_metric", json=invalid_data)
assert response.status_code == 422
error_detail = response.json()
assert "detail" in error_detail
# Verify error message contains expected error
error_text = json.dumps(error_detail)
assert expected_error in error_text or expected_error.lower() in error_text.lower()

View File

@@ -0,0 +1,284 @@
import json
from unittest.mock import AsyncMock, Mock
import fastapi
import fastapi.testclient
import pytest
import pytest_mock
from pytest_snapshot.plugin import Snapshot
import backend.server.routers.analytics as analytics_routes
app = fastapi.FastAPI()
app.include_router(analytics_routes.router)
client = fastapi.testclient.TestClient(app)
@pytest.fixture(autouse=True)
def setup_app_auth(mock_jwt_user):
"""Setup auth overrides for all tests in this module"""
from autogpt_libs.auth.jwt_utils import get_jwt_payload
app.dependency_overrides[get_jwt_payload] = mock_jwt_user["get_jwt_payload"]
yield
app.dependency_overrides.clear()
def test_log_raw_metric_success(
mocker: pytest_mock.MockFixture,
configured_snapshot: Snapshot,
test_user_id: str,
) -> None:
"""Test successful raw metric logging"""
# Mock the analytics function
mock_result = Mock(id="metric-123-uuid")
mock_log_metric = mocker.patch(
"backend.data.analytics.log_raw_metric",
new_callable=AsyncMock,
return_value=mock_result,
)
request_data = {
"metric_name": "page_load_time",
"metric_value": 2.5,
"data_string": "/dashboard",
}
response = client.post("/log_raw_metric", json=request_data)
assert response.status_code == 200
response_data = response.json()
assert response_data == "metric-123-uuid"
# Verify the function was called with correct parameters
mock_log_metric.assert_called_once_with(
user_id=test_user_id,
metric_name="page_load_time",
metric_value=2.5,
data_string="/dashboard",
)
# Snapshot test the response
configured_snapshot.assert_match(
json.dumps({"metric_id": response.json()}, indent=2, sort_keys=True),
"analytics_log_metric_success",
)
def test_log_raw_metric_various_values(
mocker: pytest_mock.MockFixture,
configured_snapshot: Snapshot,
) -> None:
"""Test raw metric logging with various metric values"""
# Mock the analytics function
mock_result = Mock(id="metric-456-uuid")
mocker.patch(
"backend.data.analytics.log_raw_metric",
new_callable=AsyncMock,
return_value=mock_result,
)
# Test with integer value
request_data = {
"metric_name": "api_calls_count",
"metric_value": 100,
"data_string": "external_api",
}
response = client.post("/log_raw_metric", json=request_data)
assert response.status_code == 200
# Test with zero value
request_data = {
"metric_name": "error_count",
"metric_value": 0,
"data_string": "no_errors",
}
response = client.post("/log_raw_metric", json=request_data)
assert response.status_code == 200
# Test with negative value
request_data = {
"metric_name": "temperature_delta",
"metric_value": -5.2,
"data_string": "cooling",
}
response = client.post("/log_raw_metric", json=request_data)
assert response.status_code == 200
# Snapshot the last response
configured_snapshot.assert_match(
json.dumps({"metric_id": response.json()}, indent=2, sort_keys=True),
"analytics_log_metric_various_values",
)
def test_log_raw_analytics_success(
mocker: pytest_mock.MockFixture,
configured_snapshot: Snapshot,
test_user_id: str,
) -> None:
"""Test successful raw analytics logging"""
# Mock the analytics function
mock_result = Mock(id="analytics-789-uuid")
mock_log_analytics = mocker.patch(
"backend.data.analytics.log_raw_analytics",
new_callable=AsyncMock,
return_value=mock_result,
)
request_data = {
"type": "user_action",
"data": {
"action": "button_click",
"button_id": "submit_form",
"timestamp": "2023-01-01T00:00:00Z",
"metadata": {
"form_type": "registration",
"fields_filled": 5,
},
},
"data_index": "button_click_submit_form",
}
response = client.post("/log_raw_analytics", json=request_data)
assert response.status_code == 200
response_data = response.json()
assert response_data == "analytics-789-uuid"
# Verify the function was called with correct parameters
mock_log_analytics.assert_called_once_with(
test_user_id,
"user_action",
request_data["data"],
"button_click_submit_form",
)
# Snapshot test the response
configured_snapshot.assert_match(
json.dumps({"analytics_id": response_data}, indent=2, sort_keys=True),
"analytics_log_analytics_success",
)
def test_log_raw_analytics_complex_data(
mocker: pytest_mock.MockFixture,
configured_snapshot: Snapshot,
) -> None:
"""Test raw analytics logging with complex nested data"""
# Mock the analytics function
mock_result = Mock(id="analytics-complex-uuid")
mocker.patch(
"backend.data.analytics.log_raw_analytics",
new_callable=AsyncMock,
return_value=mock_result,
)
request_data = {
"type": "agent_execution",
"data": {
"agent_id": "agent_123",
"execution_id": "exec_456",
"status": "completed",
"duration_ms": 3500,
"nodes_executed": 15,
"blocks_used": [
{"block_id": "llm_block", "count": 3},
{"block_id": "http_block", "count": 5},
{"block_id": "code_block", "count": 2},
],
"errors": [],
"metadata": {
"trigger": "manual",
"user_tier": "premium",
"environment": "production",
},
},
"data_index": "agent_123_exec_456",
}
response = client.post("/log_raw_analytics", json=request_data)
assert response.status_code == 200
response_data = response.json()
# Snapshot test the complex data structure
configured_snapshot.assert_match(
json.dumps(
{
"analytics_id": response_data,
"logged_data": request_data["data"],
},
indent=2,
sort_keys=True,
),
"analytics_log_analytics_complex_data",
)
def test_log_raw_metric_invalid_request() -> None:
"""Test raw metric logging with invalid request data"""
# Missing required fields
response = client.post("/log_raw_metric", json={})
assert response.status_code == 422
# Invalid metric_value type
response = client.post(
"/log_raw_metric",
json={
"metric_name": "test",
"metric_value": "not_a_number",
"data_string": "test",
},
)
assert response.status_code == 422
# Missing data_string
response = client.post(
"/log_raw_metric",
json={
"metric_name": "test",
"metric_value": 1.0,
},
)
assert response.status_code == 422
def test_log_raw_analytics_invalid_request() -> None:
"""Test raw analytics logging with invalid request data"""
# Missing required fields
response = client.post("/log_raw_analytics", json={})
assert response.status_code == 422
# Invalid data type (should be dict)
response = client.post(
"/log_raw_analytics",
json={
"type": "test",
"data": "not_a_dict",
"data_index": "test",
},
)
assert response.status_code == 422
# Missing data_index
response = client.post(
"/log_raw_analytics",
json={
"type": "test",
"data": {"key": "value"},
},
)
assert response.status_code == 422

Some files were not shown because too many files have changed in this diff Show More