chat-assistant-backend

Status: Production FastAPI + Postgres backend deployed to Cloud Run and fronted by API Gateway. MongoDB has been removed; the Node OpenAuxilium/ service is the only legacy/experimental backend.

Production endpoints (Cloud Run + API Gateway)

• Public gateway (for frontend/Netlify): https://chat-assistant-backend-gw-3j4dip0k.uc.gateway.dev
• Backend service (private; use identity token if calling directly): https://chat-assistant-backend-276715928612.us-central1.run.app
• Frontend env var: set VITE_CHAT_API_BASE (or the legacy alias VITE_ASSISTANT_API_URL) to the gateway URL in the website project.

Tech Stack

• Language: Python
• Framework: FastAPI + Uvicorn
• Database: Postgres (Cloud SQL)
• AI Provider: OpenAI Chat Completions (configurable) with optional local LLM fallback via LOCAL_LLM_BASE_URL + LOCAL_LLM_MODEL
• Deployment: Cloud Run container with API Gateway in front; Heroku container workflow is kept for parity
• Services: web (API) + worker (background jobs)
• Telemetry: Lightweight logging gated by TELEMETRY_ENABLED (default true). Responses are trimmed to 2-3 sentences for consistency.

Local Development

1. Copy .env.example to backend/.env and set:
   • PORT=4000
   • DATABASE_URL=postgresql://app_user:password@127.0.0.1:5432/ravdevops
   • OPENAI_API_KEY=... (omit or set to disabled to exercise offline/local modes)
   • OPENAI_TIMEOUT_SECONDS and OPENAI_MAX_RETRIES as needed
   • ADMIN_API_TOKEN=... (required for /admin/* and write operations on /api/{collection})
   • CORS_ALLOW_ORIGINS=http://localhost:5173 plus any preview domains
2. Install dependencies and run the API:

   cd backend
   python -m venv .venv
   source .venv/bin/activate  # On Windows: .venv\Scripts\activate
   pip install -r requirements.txt
   uvicorn backend.main:app --reload --host 0.0.0.0 --port 4000

3. Point the frontend to your local instance by setting VITE_CHAT_API_BASE=http://localhost:4000 in the website .env.
4. Run the test suite (optional but recommended):

   pytest

CORS configuration

• Default dev origins: http://localhost:3000, http://localhost:5173 (see backend/config.py).
• Production setting is currently: CORS_ALLOW_ORIGINS="https://ravdevops.com,https://www.ravdevops.com".
• If you add preview domains, update CORS_ALLOW_ORIGINS in Cloud Run (or equivalent) to include them.

Docker & Cloud Run / Heroku

• Build and run locally with Docker:

  docker build -t chat-assistant-backend .
  docker run -p 4000:4000 --env-file backend/.env chat-assistant-backend

• The same image is used for Cloud Run; Heroku container deployment is retained for parity:
  • heroku stack:set container -a chat-assistant-backend
  • heroku container:push web worker -a chat-assistant-backend
  • heroku container:release web worker -a chat-assistant-backend

Documentation

• Roadmap: roadmap.md
• Detailed docs (API, deployment, architecture): docs/
• Project wiki & additional notes: wiki.md

CI / Automation

• GitHub Actions runs the test suite on each push and pull request to main (see .github/workflows/tests.yml).
• CodeQL code scanning is enabled for Python to detect security issues (see .github/workflows/codeql.yml).

Contributing / Pull Requests

1. Create a feature branch on your clone.
2. Ensure you have a local env file (see backend/.env, which is ignored by Git).
3. Make your code changes and run the backend locally (see "Local Development").
4. Commit your changes with a clear message.
5. Open a Pull Request (PR) against the main branch, describing:
   • What you changed.
   • How to test it.
   • Any new environment variables or config required.

License

This project is not open source. See LICENSE for details. All rights reserved.