The open source, multi-agent platform

Crews of AI agents can take advantage of a specialized planning agent that creates a step-by-step plan for all tasks and shares it with the crew. The new plan-execute pattern lets agents draft, refine, and follow structured plans that adapt as work unfolds.

CrewAI provides AI agents with hundreds of open-source tools out of the box — search the Internet, interact with websites, query vector databases, run code, and much more. With first-class support for MCP (Model Context Protocol), Custom MCP Servers, and native sandbox tools for E2B and Daytona, agents can safely execute code and call any tool you need, anywhere.

CrewAI implements a sophisticated memory-management system that gives AI agents access to shared short-term, long-term, entity, and contextual memory. Choose from storage backends including Qdrant Edge, ChromaDB, Mem0, and SQLite, with hierarchical memory isolation via automatic root_scope so multi-tenant crews stay safely separated.

Agentic RAG combines a broad range of knowledge sources, files, websites, and vector databases — with intelligent query rewriting to optimize retrieval. Path and URL validation plus SSRF protections ship by default, so agents pull from the right sources without putting your stack at risk.

CrewAI transforms a set of AI agents into a crew of AI agents that collaborate via context sharing and delegation to perform complex tasks. Native Agent-to-Agent (A2A) protocol support lets crews coordinate across processes, services, and clouds — with poll, stream, and push update mechanisms baked in.

Pause, resume, fork, and replay any crew or flow. CrewAI's checkpointing system captures runtime state at every step, with a tree-view TUI for browsing checkpoints, editable inputs and outputs, and lineage tracking across forks. Bring crashed runs back to life or branch a successful run into multiple variations — all without rewriting your code.

Run crews, flows, tasks, tools, knowledge, memory, and LLM calls asynchronously with native async/await support. Streaming results flow back in real time from flows and crews, and async context managers make resource cleanup automatic. Write the orchestration you want — CrewAI handles the concurrency.

Run crews, flows, tasks, tools, knowledge, memory, and LLM calls asynchronously with native async/await support. Streaming results flow back in real time from flows and crews, and async context managers make resource cleanup automatic. Write the orchestration you want — CrewAI handles the concurrency.