Multi-agent crew designer
Describe a goal and get a crewAI design: agents with roles, tasks with concrete outputs, tools, hand-offs and the YAML to run it.
Your key stays in your browser.It goes straight to the provider, never to our servers, and we neither log nor keep it. Use a dedicated key with a spending cap, and delete it in the provider's dashboard when you are done.
Input
Result
The result will appear here.Multi-agent systems fail in predictable ways: six agents where two would do, tasks whose 'expected output' is 'a comprehensive report', every agent given every tool, and nobody checking the result before it is acted on. crewAI's structure — agents with a role, goal and backstory, tasks with an explicit expected_output, a process that is sequential unless there is a reason — is a good antidote, and this designer applies it to your goal. You get the agents.yaml and tasks.yaml plus a minimal crew.py, or a framework-neutral design with a hand-off diagram, with a reviewer at the end whenever the output will be used.
How it works
- Each agent owns one kind of work a specialist would; agents that could be merged are merged, and the designer says why fewer are enough rather than padding to the requested count.
- Every task has a deliverable a reviewer can check — a table with these columns, a 400-word digest, a list of URLs with one-line notes — and the context it depends on.
- Tool access is minimal and named per agent, and the sequential or hierarchical process is chosen with a one-line reason.
Where your data goes
With your own key, what you type and the key go from your browser straight to the provider you choose; neither passes through Hysen Labs. In hosted mode, your text goes through our server to our provider (DeepSeek) and is billed in credits; we keep the token counts and cost of each run for billing, never the text itself or the answer. How the provider handles the text is governed by its own privacy policy.
This tool handles keys and credentials, so nothing about a run is saved, not even to your own history.
About your API key
We do not collect, store or leak your key: it lives only in this page's memory (unless you tick “remember in this tab”) and is gone when you close it. Still, treat any key you have pasted into a web page with care — create a dedicated key with a spending cap for use here, and delete or rotate it in your provider's dashboard when you are done.
What it costs
This tool is free, with no sign-in and no points.
Common questions
- Can I use the design with LangGraph, AutoGen or the OpenAI Agents SDK?
- Choose the framework-neutral format: the agents table, the ordered hand-offs, shared state and failure handling translate directly to a LangGraph graph or an Agents SDK handoff chain. The YAML format is specific to crewAI's project layout.
- Why does it keep adding a reviewer agent?
- Because an agent's output that nobody checks is where hallucinated links, wrong numbers and skipped steps reach your Slack or your database. The reviewer's task names what it verifies against — the sources, the roadmap file, a format — and is the cheapest place to catch errors. Remove it if the output is only a draft for a human.
- Does the output run as is?
- The YAML is complete and crew.py wires it with the current crewAI decorators, but you still provide the tools (search, fetcher, Slack webhook) and an API key, and you should tighten the backstories to your domain. Treat it as a reviewed first draft rather than a finished system.
The open-source behind it
This tool is a self-contained implementation. crewAIInc/crewAI (MIT) does the same job as a library — if you need this behaviour inside your own program, start there rather than calling a web page.
crewAIInc/crewAIAlso known as
- crewai agents yaml generator
- multi-agent system design
- crewai tasks.yaml example
- design ai agent workflow
- agent roles and tasks
- multi agent orchestration template