sudheerdunga

FastAPI Langgraph Starter — AI skill for Claude Code

AI community

A production-ready, async AI agent boilerplate built with FastAPI, LangGraph, and uv.

How to install FastAPI Langgraph Starter

This entry records only its repository, not the path inside it, so there is no exact command to give. Open sudheerdunga/fastapi-langgraph-starter and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What FastAPI Langgraph Starter does

A production-ready, async AI agent boilerplate built with FastAPI, LangGraph, and uv. Features multi-provider LLM support (Anthropic Claude, Google Gemini, Ollama), stateful multi-turn conversations with SQLite persistence, dynamic tool execution, and real-time Server-Sent Events (SSE) streaming.

Alternatives in AI

  • Opencodex — Universal provider proxy for OpenAI Codex & Claude Code — use any LLM (Claude, Gemini, Grok, DeepSeek, Ollama… 12.3k ★
  • Loushang — AI-native agent harness for coding workflows by python: multi-model LLM orchestration, stateful sessions, tool 1.3k ★
  • Agor — Orchestrate Claude Code, Codex, and Gemini sessions on a multiplayer canvas 1.2k ★

README

FastAPI LangGraph Starter 🚀

A modern, production-grade template for building stateful AI agents using **FastAPI**, **LangGraph**, and **uv**.

🌟 Key Highlights

  • FastAPI Backend: Async, high-performance web service with automated OpenAPI Swagger UI.
  • LangGraph Agent Workflow: Dynamic cyclic graph workflows with conditional tool routing and SQLite session checkpointer persistence (thread_id).
  • Multi-LLM Provider Support: Out-of-the-box support for Anthropic Claude, Google Gemini, and local Ollama models.
  • Real-Time Streaming: Server-Sent Events (SSE) emitting token streams and tool lifecycle events (tool_start, tool_end).
  • Tool Registry: Dynamic tool discovery, registration, and per-request tool filtering.
  • Modern Python Tooling: Fast dependency resolution with uv, code linting with ruff, and testing with pytest.

📂 Quick Links

🚀 Getting Started

Navigate to the `backend` folder and get started:

cd backend
cp .env.example .env
uv sync --all-groups
uv run app

Visit the interactive Swagger API documentation at: [http://localhost:8000/docs](http://localhost:8000/docs).