Skill

A Claude Code skill that runs a multi-agent discussion room

Install

npx -y skills add OrYamin/room --skill skill

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Start a multi-agent discussion room where several Claude agents with different expert personas collaborate, debate, and plan together. Use when the user wants multiple perspectives on a decision, wants to explore trade-offs from different viewpoints, or wants a simulated planning meeting / design review. Triggers include phrases like "get multiple opinions", "discuss with a team", "simulate a planning meeting", "have experts debate", "what would different stakeholders think", or any /room invocation.

SKILL.md

2.4 KB, 499 tokens by cl100k_base, as published. Nobody here has run it

Room skill

You have been invoked as /room. A compiled Node.js bundle does all the actual multi-agent orchestration. Your job is just to translate the user's natural language arguments into one JSON config line and exec the bundle via Bash.

Step 1 — parse `$ARGUMENTS`

Build a single-line JSON object with these fields (omit any field the user did not specify so defaults apply):

goal (string, required) — the discussion topic. Take everything that isn't a flag. If the user typed only flags, ask them for the goal first.
method ("round-robin" | "popcorn" | "orchestrator", default round-robin). Accept synonyms: rr→round-robin, pop→popcorn, orch/boss→orchestrator.
agents (integer, default 3) — number of persona agents.
maxTurns (integer, default 12).
threshold (float 0..1, default 0.85) — consensus confidence required to stop early.
model (string, optional) — override the heavy model for persona turns.
personas (array of {name, description}, optional) — if omitted, the bundle auto-generates personas tailored to the goal.

Step 2 — run the bundle

Execute via the Bash tool, passing the JSON as a single argv arg. Use single quotes around the JSON and escape any inner single quotes:

node ~/.claude/skills/room/room.js '<json-config>'

The bundle streams persona turns, router/orchestrator/decision events, and a final synthesis to stdout. Relay the output back to the user verbatim — do not paraphrase the discussion.

Step 3 — working directory

Do NOT cd before invoking. The bundle reads process.cwd() at startup and injects it into each persona's prompt so they can ground their points in the user's current repo via Read/Glob/Grep.

Notes

Persona agents have read-only tools only (Read, Glob, Grep, WebSearch, WebFetch). They cannot modify files.
If the bundle errors out, surface the stderr to the user — do not retry silently.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.