Skip to main content

​
Build Sophisticated AI Agents in Minutes

Fast Agent is the first Python framework with complete, end-to-end tested MCP support including Sampling and Elicitations. Create multimodal agents and workflows with simple declarative syntax.

Get StartedView on GitHub

​
Key Features

Declarative Agents

Define agents with Python decorators and minimal boilerplate

Complete MCP Support

First framework with full MCP feature support including OAuth and sampling

Workflow Patterns

Chain, parallel, router, orchestrator, and more built-in patterns

Multimodal Ready

PDF, vision, audio, and video support out of the box

Multi-Provider

Support for Anthropic, OpenAI, Google, Azure, Ollama, and more

CLI-First Design

Interactive terminal UI with completions and live streaming

​
Quick Example

Get started with a basic agent in just a few lines of code:
agent.py
import asyncio
from fast_agent import FastAgent

# Create the application
fast = FastAgent("My Agent")

@fast.agent(
    instruction="You are a helpful AI assistant"
)
async def main():
    async with fast.run() as agent:
        await agent.interactive()

if __name__ == "__main__":
    asyncio.run(main())
Run your agent with: 
uv pip install fast-agent-mcp
uv run agent.py

​
Installation

1

Install uv package manager

Fast Agent uses the modern uv package manager for Python.
curl -LsSf https://astral.sh/uv/install.sh | sh
2

Install Fast Agent

Install fast-agent from PyPI:
uv pip install fast-agent-mcp
3

Start building

Create your first agent or explore examples:
fast-agent scaffold         # Create example agent
fast-agent quickstart workflow  # Create workflow examples
fast-agent go              # Start interactive session

Workflow Automation

Chain agents together to automate complex multi-step tasks

Data Analysis

Build data analysis agents with MCP server integrations

Content Generation

Create quality-controlled content with evaluator-optimizer workflows

Research Assistants

Build research agents that fetch and analyze information

​
Learn More

Core Concepts

Understand agents, workflows, and MCP integration

CLI Reference

Explore all available CLI commands

API Reference

Complete API documentation

​
Community

Join our community to get help, share projects, and stay updated: