FastAPI LangGraph Agent Production Template

Use it when

• Your LangGraph agent runs fine as a local prototype but you need JWT auth, rate limiting, migrations, and monitoring before exposing it as a hosted API
• Your agent forgets users between sessions and you want per-user semantic long-term memory plus tracing on every LLM call without wiring mem0, Langfuse, and Prometheus by hand

What it solves

Not the fit when

• Frontend or chat UI; this is a backend API template only
• Non-LangGraph agent frameworks; the scaffold is built around LangGraph checkpointing
• Out-of-the-box multi-provider LLM support; the LLMRegistry currently targets OpenAI-compatible endpoints only
• context-window-overflow
• structured-output-drift

Install

git clone <repo-url> my-agent && cd my-agent && cp .env.example .env.development && make install

Invoke

make docker-up then open http://localhost:8000/docs for the interactive API; add tools by dropping @tool functions into app/core/langgraph/tools/

Alternatives

No reviewed alternatives recorded yet.