ComposioHQ/composio

Composio powers 1000+ toolkits, tool search, context management, authentication, and a sandboxed workbench to help you build AI agents that turn intent into action.

What it solves

Composio provides a unified interface for AI agents to interact with external applications. It solves the problem of manually integrating dozens of different APIs, managing user-specific authentication, and handling the complex setup required to turn an AI's intent into actual actions across various software tools.

How it works

Composio acts as a middleware layer between the AI agent and external apps. It provides over 1,000 pre-authenticated toolkits. Instead of loading all tool definitions into an agent's context window, it uses "meta tools" that allow agents to discover and execute tools at runtime.

Key components include:

  • SDKs: Available in TypeScript and Python for easy integration into agent frameworks.
  • Sessions: User-scoped sessions that manage authentication and tool access for specific users.
  • Providers: Adapters that translate Composio tools into the native format required by frameworks like OpenAI, LangChain, and Claude.
  • MCP Support: Every session can be exposed as a Model Context Protocol (MCP) endpoint for clients like Cursor or Claude.
  • CLI: A command-line interface for searching, executing, and scripting tools directly from the shell.

Who it’s for

Developers building AI agents that need to perform real-world tasks across multiple third-party applications (e.g., automating email summaries, managing calendars, or interacting with SaaS tools).

Highlights

  • 1,000+ Pre-authenticated Toolkits: Immediate access to a vast library of app integrations.
  • Per-user Sessions: Securely manages authentication and tool access on a per-user basis.
  • Broad Framework Support: Native adapters for OpenAI, Anthropic, Vercel AI SDK, LangChain, LlamaIndex, CrewAI, and AutoGen.
  • MCP Integration: Supports Model Context Protocol for seamless connection to MCP-compatible clients.
  • Coded-Agent Friendly: The TypeScript SDK is designed to be inspectable by coding agents.

Related

  • Project
    compozy/compozy

    An operating system for AI agents that provides a daemon-based runtime to manage persistent sessions, automation, and memory for ACP-compatible agent CLIs.

  • Project
    UsefulSoftwareCo/executor

    An open-source integration layer that centralizes tool configurations, authentication, and policies for AI agents via the Model Context Protocol (MCP).

  • Project
    aipotheosis-labs/aci

    ACI.dev is an open-source tool-calling platform that provides AI agents with unified, secure access to over 600 pre-built integrations via an MCP server or SDK.

  • Project
    mondaycom/mcp

    An open framework and MCP server that enables AI agents to securely interact with monday.com to manage projects, track tasks, and execute workflows.

  • Project
    codeany-ai/open-agent-sdk-typescript

    An open-source TypeScript SDK for running autonomous AI agents in-process, featuring a vast library of built-in tools and support for multiple LLM providers.