OpenAI Agents SDK

by OpenAI

Lightweight framework for multi-agent workflows with handoffs and tracing

The OpenAI Agents SDK is a lightweight Python and TypeScript framework for building multi-agent workflows. Agents combine instructions, tools, guardrails, and handoffs, with built-in sessions, human-in-the-loop, MCP tools, and tracing. It is provider-agnostic and supports the OpenAI APIs and many other models.

Key features

  • Handoffs and agents-as-tools
  • Input and output guardrails
  • Sessions and human-in-the-loop
  • MCP and hosted tools
  • Built-in tracing
  • Voice, realtime, and sandbox agents

Similar tools

  • Claude Agent SDK

    Build custom agents on the same agent loop and tools as Claude Code

    by AnthropicFreeCLI
  • Vercel AI SDK

    Provider-agnostic TypeScript toolkit for AI apps and agents

    by VercelFreeOpen sourceCLI
  • FastMCP

    Fast, Pythonic framework for building MCP servers and clients

    by PrefectFreeOpen sourceCLI
  • MCP Python SDK

    Official Python SDK for Model Context Protocol servers and clients

    FreeOpen sourceCLI
  • MCP TypeScript SDK

    Official TypeScript SDK for building MCP servers and clients

    FreeOpen sourceCLI
  • MCP Inspector

    Official visual testing tool for Model Context Protocol servers

    FreeOpen sourceWeb, CLI

Built a tool like OpenAI Agents SDK? Submit it for review