CrewAI

Overview

CrewAI is an open-source Python framework for building multi-agent workflows in which multiple AI agents split roles and work together. It provides both high-level abstractions and low-level APIs.

It has two approaches: Crews, in which role-based agents collaborate autonomously, and Flows, which give precise, event-driven control over the flow. The commercial enterprise product, CrewAI AMP Suite, adds managed deployment, real-time tracing and observability, governance and security features, and also provides a no-code visual editor.

It suits developers who want to automate work with agents, and enterprise platform teams that need to govern this centrally. The enterprise product provides permission management, audit logs, human approval steps, and the ability to mask personal information on every LLM and tool call.

What it is used for

• Have multiple agents with assigned roles handle complex tasks together.
• Build event-driven agent automation with defined steps.
• Trace agent runs in real time and check costs.
• Add steps where a person approves or intervenes during a run.

Pricing and license

The open-source framework is MIT License. For the commercial CrewAI AMP Suite, see the official website

Getting started

1. Check installation instructions in the GitHub repository and official documentation (docs.crewai.com).
2. Build your first agent workflow with Crews or Flows.
3. If you need operational management, try Crew Control Plane for free.

Links

Official site ↗Repository ↗

Related terms and companies

Multi-agent AI agent Agentic workflow Human-in-the-loop LLM observability Personal data de-identification

Sources

• crewAIInc/crewAI README
• CrewAI

Compiled from public sources (official pages, READMEs). Pricing and features change, so check the official site.