Multi-agent orchestration
How to use
CrewAI.
Building collaborative agent crews for autonomous tasks and structured, stateful Flows for reliable event-driven AI automation.
CrewAI is a framework and managed platform for multi-agent crews and event-driven AI flows. Building collaborative agent crews for autonomous tasks and structured, stateful Flows for reliable event-driven AI automation. This guide covers the whole path in one place: official access, a first session that produces something reviewable, the checks that make output trustworthy, and the permissions worth limiting before you connect real work.
AI should make the work easier to inspect. If the workflow removes the source, the owner, or the review step, redesign the workflow.
Access & setup
Find, install, and sign in to CrewAI
Get into the official CrewAI experience with the right account and a setup you understand.
- 01
Start at https://www.crewai.com/ and confirm the domain before entering account or payment information.
- 02
Availability: CrewAI is available as an open-source Python framework and CLI, with CrewAI AMP for managed deployment, monitoring, triggers, and team controls.
- 03
A supported Python environment for the framework or CrewAI account for AMP, configured model credentials, and securely implemented tools and data sources.
- 04
Sign in with the account you intend to keep using, then review plan, data, notification, and permission settings.
- 05
Run one low-risk test task before connecting sensitive files, repositories, or workspace data.
- Use official download pages.
- Review permissions during setup.
- Keep installers and applications updated.
First session
Your first useful CrewAI session
Learn the interaction loop using a small task with a clear outcome.
- 01
Define a small benchmark task and confirm that multiple specialized agents are more useful than one model call.
- 02
Install CrewAI in an isolated project and store model credentials outside code.
- 03
Create two narrow agents and sequential tasks with structured outputs and no external side effects.
- 04
Run the benchmark while inspecting delegation, loops, errors, latency, token use, and result quality.
- State the outcome before the background.
- Provide the real source material.
- Review the result before expanding the task.
Quality control
Check the quality of CrewAI output
Establish that an autonomous run did the right thing, not merely that it finished.
- 01
Define what the run should achieve and what it must never touch before granting it a single tool.
- 02
Read the full execution trace: which tools were called, with what arguments, and in what order.
- 03
Verify the side effects directly in the target system rather than trusting the agent's own report of success.
- 04
Confirm failures surfaced as failures — a silent retry loop or a swallowed error is more dangerous than a crash.
- 05
Re-run the same task and compare: an agent that behaves differently across identical runs is not yet production-ready.
- Verify side effects in the system of record, not in the agent's summary.
- Require human approval for any irreversible or outward-facing action.
- Log every tool call so a run can be reconstructed afterwards.
Privacy & permissions
Use CrewAI safely
Bound what an autonomous system can reach before you let it run unattended.
- 01
Enumerate every tool, credential, and system the agent can reach, and remove the ones it does not need.
- 02
Require explicit human approval for irreversible actions: sending, publishing, paying, deleting, or deploying.
- 03
Run against non-production data until behaviour is predictable across repeated runs.
- 04
Set hard limits on spend, iterations, and runtime so a failure loop cannot run unbounded.
- 05
Treat anything the agent reads from the web or a document as data, never as instructions it may follow.
- Multi-agent designs can multiply nondeterminism, context exposure, tool authority, loops, latency, and cost. Justify every agent, cap iterations and budgets, constrain delegation and memory, validate structured task outputs, sandbox custom tools, require approval for side effects, make Flow state and retries idempotent, trace complete runs, compare against simpler baselines, and pin framework, model, and dependency versions.
- Follow your organisation's approved-use policy.
- Never treat fluent output as authorization to act.
Core workflows
Step-by-step ways to use CrewAI for the work it does best.
Each workflow is a separate guide with its own steps and review checkpoints.
Assign narrow roles, tasks, tools, outputs, and guardrails without multiplying agents beyond the problem's needs.
↗ 02 WorkflowBuild a recoverable CrewAI FlowUse explicit state, routes, persistence, crews, and human input for long-running automation that can fail and resume safely.
↗Official references
Check the current product documentation.
Features, plan limits, availability, and data controls change. These official pages are the starting points used for this guide.