Full toolkit for running an AI agent service built with LangGraph, FastAPI and Streamlit. Python-based, MIT-licensed, and actively updated since 2024-08-04 with 4406 stars.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
A full toolkit for running an AI agent service built with LangGraph, FastAPI and Streamlit. It includes a LangGraph agent, a FastAPI service to serve it, a client to interact with the service, and a Streamlit app that uses the client to provide a chat interface. Data structures and settings are built with Pydantic.
How it works
The project provides multiple components:
- LangGraph Agent with features like interrupt handling, flow control, long-term memory, and supervisor integration.
- FastAPI service to expose the agent with streaming and non-streaming endpoints.
- A client module to interact with the service (synchronous, asynchronous, streaming and non-streaming supported).
- A Streamlit app for a chat UI, including voice input/output.
- Asynchronous design across components and a RAG agent using ChromaDB.
- Docker support with a compose setup for local development and deployment.
Getting started
Quickstart and setup steps are provided in the README. Key commands include: "echo 'OPENAI_API_KEY=your_openai_api_key' >> .env" and the recommended uv installation and sync, followed by activation and running the service: "# At least one LLM API key is required" (from README). The README also shows docker-based setup: "docker compose watch". Local development without Docker uses a Python virtual environment with commands like: "uv sync --frozen" and "source .venv/bin/activate" as part of the workflow, followed by running the server with: "python src/run_service.py" and then "streamlit run src/streamlit_app.py".
Recent releases
RELEASES (latest 0): - none
Traction
Stars: 4406. Forks: 747. Open issues: 14. Language: Python. License: MIT. Created: 2024-08-04. Last push: 2026-08-04.
Behind the repo
Not applicable: No linked startup/company section in the provided facts.
Caveats
License: MIT. Created: 2024-08-04. Last push: 2026-08-04. Open issues: 14.