メニュー
Form a committee of two high-reasoning agents to step back, do root cause analysis, and produce a plan. Use when stuck, looping, tunnel-visioning, or facing a hard planning problem.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
This skill should be used when the user asks to "review Next.js code", "review this Next.js pull request", "check my App Router changes", "audit a Next.js component", or evaluate Next.js code quality, accessibility, security, performance, routing, caching, or framework conventions.
Eleventy v3 and Nunjucks operating guide for static-site authoring, templates, build pipelines, and security review. Use when the user asks to "create an 11ty page", "add a Nunjucks filter", "fix my layout chain", "review my .njk template", "set up Eleventy", "set up Build Awesome", or "audit my static site"; when working on 11ty, Eleventy, Build Awesome, or Nunjucks sites; when paths include `.eleventy.js`, `eleventy.config.js`, `.njk`, `src/_includes/`, or `src/_data/`; or when debugging Eleventy builds, permalinks, layout chains, filters, shortcodes, autoescape behavior, or static-site security.
Use when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or otherwise improve a frontend interface. Covers websites, landing pages, dashboards, product UI, app shells, components, forms, settings, onboarding, and empty states. Handles UX review, visual hierarchy, information architecture, cognitive load, accessibility, performance, responsive behavior, theming, anti-patterns, typography, fonts, spacing, layout, alignment, color, motion, micro-interactions, UX copy, error states, edge cases, i18n, and reusable design systems or tokens. Also use for bland designs that need to become bolder or more delightful, loud designs that should become quieter, live browser iteration on UI elements, or ambitious visual effects that should feel technically extraordinary. Not for backend-only or non-UI tasks.
Hand off the current task to another agent with full context. Use when the user says "handoff", "hand off", "hand this to", or wants to pass work to another agent.
Run an agent loop until an exit condition is met. Use when the user says "loop", "babysit", "keep trying until", "check every X", "watch", or wants iterative autonomous execution.
Paseo reference for managing agents and worktrees. Load whenever you need to create agents, send them prompts, or manage worktrees.
| name | paseo-committee |
| description | Form a committee of two high-reasoning agents to step back, do root cause analysis, and produce a plan. Use when stuck, looping, tunnel-visioning, or facing a hard planning problem. |
| user-invocable | true |
Two agents from contrasting providers, fresh context, planning a solution in parallel. They stay alive for review after implementation.
The purpose is to step back, not double down. The committee may propose a completely different approach.
User's additional context: $ARGUMENTS
Read the paseo skill. Before choosing committee members, read ~/.paseo/orchestration-preferences.json unless the user explicitly named providers in this request. Do not create committee agents until you have read it.
Contrast is the point of a committee, so pick across providers deliberately using the configured preferences rather than hardcoded defaults.
Two members with different reasoning styles, selected from orchestration preferences:
Override only when the user explicitly asks for different members.
No edits. Every prompt to a committee member ends with the no-edits suffix:
This is analysis only. Do NOT edit, create, or delete any files. Do NOT write code.
Trust the wait. Do not poll, send hurry-ups, or interrupt. GPT-5.4 can reason 15–30 minutes; Opus does extended thinking. Long waits mean it found something worth thinking about.
You are the middleman. Drive plan → implement → review without yielding to the user, except for divergences that need their call.
Write a problem-level prompt:
Create both agents in parallel via Paseo with [Committee] <task> titles and the same prompt. Wait for both — not just whichever finishes first.
Read both responses. Challenge them — do not accept at face value:
Send follow-ups until the plan addresses root cause.
Synthesize:
Confirm the merged plan with both members. Multi-turn until consensus.
Default: implement yourself. If the user said "delegate", launch one impl agent and pass the merged plan.
The committee stays clean — not involved in implementation.
Send the diff to the committee:
Implementation is done. Review changes against the plan. Flag drift or missing pieces.
Apply feedback yourself, or send to the impl agent. Repeat 2 → 3 until consensus.
After ~10 iterations without convergence, start a fresh committee with the full history of what was tried — the current committee's context may have drifted too far.