Skip to main content

Quick Start

Get up and running in minutes

API Reference

Explore the complete API

Authentication

Learn about JWT and OAuth

MCP Integration

Connect external tools via MCP

​
What is Agent Chassis?

Agent Chassis is a modular, scalable foundation for building AI agents with FastAPI, OpenAI, and MCP (Model Context Protocol). It serves as a “chassis” or scaffolding for building robust agentic applications, handling the infrastructure—server management, protocol translation, tool execution—so you can focus on the agent’s logic and capabilities.

​
Key Features

  • Modular Architecture: Clean separation of concerns (Routes, Services, Schemas)
  • MCP Support: Native integration with Model Context Protocol servers (Stdio, SSE, Streamable HTTP)
  • Hybrid Tooling: Mix remote MCP tools and local Python functions seamlessly
  • Streaming Support: Real-time feedback via Server-Sent Events (SSE)
  • Session Persistence: Optional Redis + PostgreSQL storage for conversations
  • Authentication: JWT-based user auth with Google OAuth support (OSP-14)
  • Access Control: Ownership-based session access control (OSP-12)
  • Async & Scalable: Fully asynchronous design using asyncio and FastAPI
  • Modern Tooling: Built with uv for lightning-fast dependency management

​
Architecture

Agent Chassis follows a clean, modular architecture: 
app/
├── api/v1/endpoints/    # API route handlers
├── core/                # Configuration, security, settings
├── models/              # Database models (User, Conversation)
├── schemas/             # Pydantic request/response models
└── services/            # Business logic (Agent, MCP, Auth, etc.)

​
Use Cases

  • AI Agent Development: Build autonomous agents with tool-calling capabilities
  • MCP Integration: Connect to external data sources and tools via MCP
  • Session Management: Persistent conversations with Redis and PostgreSQL
  • Multi-User Systems: JWT authentication with Google OAuth support
  • API Development: FastAPI-based REST API with OpenAPI documentation

​
Getting Started

Ready to build your agent? Start with the Quick Start Guide or dive into the API Reference.