ClaudeMap

CrewAI

Role-playing multi-agent orchestration framework — crews of agents collaborating on tasks.

Plugins & Agentsmulti-agentorchestrationpythonframework
Visit projectcrewAIInc/crewAI

CrewAI (57k+ stars) orchestrates role-playing autonomous AI agents: define agents with roles, goals, and backstories, assemble them into crews with tasks and tools, and the framework manages delegation and collaboration. Python-first (pip install crewai), and runs on Claude models via the ANTHROPIC_API_KEY path. The highest-level multi-agent framework on the map — the opposite design point from LangGraph's explicit graphs.

What it does

  • Python framework for orchestrating multi-agent 'crews' plus event-driven 'flows'
  • Requires Python >=3.10 and <3.14; uses uv for dependency management
  • CLI installed as a uv tool; agent skills also installable via npx skills
  • Large ecosystem: templates, docs at docs.crewai.com, and a skills repo

When to use it

For Python teams building structured multi-agent pipelines — research crews, content pipelines, data processing — where roles, tasks and process order need explicit definition.

Getting started

Requires Python >=3.10 <3.14 and uv. Then:

uv tool install crewai

Agent skills for coding agents: npx skills add crewaiinc/skills

Things to know

Python version window is strict (<3.14). Framework-level abstraction means a learning curve before crews behave predictably.

Repository

crewAIInc/crewAI

Charted

Related on the map

  • clawd-conduit

    Plugin / Agent

    Wire all 15 Claude Code hook events into the Clawd on Desk desktop pet — and turn permission requests into a blocking desktop card.

    • hooks
    • desktop
    • notifications
    • claude-code
    View

  • CometixCode

    Plugin / Agent

    An unofficial Rust 1:1 reimplementation of the Claude Code terminal UI, ported file-for-file onto an iocraft TUI framework.

    • terminal-ui
    • rust
    • reimplementation
    • unofficial
    View

  • Abide

    Plugin / Agent

    Hook-level enforcement for the natural-language rules in AGENTS.md and CLAUDE.md — every edit is judged and violations are repaired in the same turn.

    • hooks
    • code-review
    • rules
    • agents-md
    View