Rehearse for $0

FAQ

Questions, answered.

Short answers. Where the answer is “not yet”, we say so.

Basics

What is yaaf?

Open-source software for building a company of AI agents. You describe the goal and hire a CEO; the CEO runs a team for weeks, and yaaf's server enforces the rules, not the prompt.

Is there a hosted version or a paid plan?

No, and there isn't going to be one to upsell you. yaaf is Apache-2.0 software you run yourself. Your agents pay your model provider directly. No account, no telemetry.

Who is it for?

Anyone handing off work that runs unattended for weeks and must be checked. Not for quick tasks; use a skill for those.

How it works

Models and cost

Which models does it run?

Claude, GPT, Gemini, Bedrock, Vertex, Azure, DeepSeek, Mistral, Kimi, Qwen, Grok, Groq, OpenRouter, a local Ollama or LM Studio, or any OpenAI-compatible server: 23 providers through 6 harnesses. Roles name a class (frontier, balanced, fast), so a team runs on whichever provider you pay. Harnesses and models →

How much does it cost?

yaaf itself costs nothing. The models cost what your provider charges. Nobody has published a real month's bill yet; the $0 rehearsal tells you the most a run can cost and still fit your budget, and every run is metered against a cap you set.

Can it run on local models?

Yes: Ollama, LM Studio or any OpenAI-compatible server. The prompts were tuned on Claude, so quality may vary.

Running it

Can I see a company run before I pay for any model?

Yes. Clone the repo, install it into a virtualenv (python3 -m venv .venv && .venv/bin/pip install -e .) and run yaaf rehearse examples/01-your-first-hour.json. A whole project runs on scripted agents: no key, no Docker, $0.

Can I run it from Claude?

Yes. yaaf is an MCP server. Claude Code, Claude Desktop and Cursor connect with a token. claude.ai needs a public address.

The project

Question not here? Ask on GitHub