Welcome to python-agents documentation!

python-agents is an easy to use and powerful AI Agent development framework that implements the REACT (Reasoning and Acting) pattern for building agents with tool-calling capabilities.

The library provides a minimal but powerful abstraction for creating agents that can reason step-by-step and use tools to accomplish complex tasks.

Contents:

• Quick Start Guide
  • Installation
  • Basic Usage
  • Key Concepts
  • Next Steps
• How-To Guides
  • Tool Calling with Python Functions
  • Using MCP Servers
  • Configure LLMClient for Different Providers
• API Reference
  • Client
  • Agents
  • Tools
  • Message
  • Memory
  • Utilities

Features

• REACT Agent Pattern: Step-by-step reasoning with tool calling

• Simple Tool Integration: Convert Python functions to agent tools automatically

• MCP Server Support: Connect to Model Context Protocol servers for external integrations

• Async-First Design: Built on async/await for efficient concurrent operations

• OpenRouter Compatible: Works with any OpenAI-compatible API endpoint

Indices and tables

• Index

• Module Index

• Search Page