#9999th of 118 in Agent frameworks and SDKs
roast-room
Claude Code skill that runs a council of AI advisors on business decisions
- Stars
- 2.1k
- License
- custom license
- Last commit
- Oct 2026
- Language
- JavaScript
9 on 6 Oct 20262,066 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Roast Room is a set of Claude Code skills and sub-agents that debate a business decision. Twelve advisor personas, each built from a person's public work, score your options in rounds, and the Chair writes a verdict to verdict.md plus a PDF. A zero-dependency Node app on localhost:4730 replays the sessions as a table view, and /newroaster builds additional advisors.
Who it is for: Founders and operators who use Claude Code and want a structured second opinion
Strengths
- Runs on a Claude plan through Claude Code, no API key needed by default
- Quotes are checked against each advisor's sources with check_quotes.py
- /newroaster builds custom advisors from public or personal material
- Web app has no npm dependencies and saves every session with PDF export
Weaknesses
- Full council uses 21 sub-agent runs for seven seats, a heavy bite of plan usage
- Advisors are simulations of real people, not endorsed by them
- Portrait generation needs a kie.ai key and opencv-python
- License is reported as NOASSERTION; README says code is MIT
What it needs
- no GPU
- Needs Claude Code, Node.js 20+, kie.ai API key (optional, portraits), yt-dlp (optional), pdftotext (optional), Anthropic API key (optional, api mode)
- Models: Claude (via Claude Code), Anthropic API (optional)
- port 4730
Quick answers
Is roast-room open source?
Its code is public on GitHub under its own license terms; read them before you use it at work.
Does roast-room need a GPU?
No, it runs without one.
Also in Agent frameworks and SDKs
See all 118| Rank | Project | Score |
|---|---|---|
| 1 | ppt-masterAgent skill that turns documents into natively editable PowerPoint decks | 95 out of 100 |
| 2 | openai-agents-pythonPython SDK for multi-agent workflows with handoffs, guardrails and tracing | 91 out of 100 |
| 3 | mastraTypeScript framework for building AI agents, workflows and MCP servers | 91 out of 100 |
| #4 | browser-usePython library that lets an LLM agent control a web browser | 90 out of 100 |
| #5 | copilotkitSDK for adding agent chat, generative UI and shared state to apps | 90 out of 100 |