Stats
Actions
Tags
From multi-model-debate
Start a multi-model debate on a software engineering topic. Use when the user invokes /debate:start with a topic to evaluate using Claude, Codex, and Gemini in parallel.
How this skill is triggered — by the user, by Claude, or both
Slash command
/multi-model-debate:debate-startThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Launch a structured multi-model debate with Claude, Codex, and Gemini.
Launch a structured multi-model debate with Claude, Codex, and Gemini.
Parse arguments:
topic (required): The debate topic or decision question--context <files>: Comma-separated file paths or glob patterns to include as context--perspectives <p1,p2,p3>: Custom perspective labels for each agent (Claude, Codex, Gemini)--rounds <N>: Number of debate rounds (default: 2)Verify prerequisites:
which codex via Bashwhich gemini via BashRead context files (if --context provided):
Create output directory:
YYYYMMDD-HHMMSS-<topic-slug> (slug: lowercase, hyphens, max 30 chars).debate/<debate-id>/Execute the debate-orchestration skill, passing:
/debate:start "Should we use REST or GraphQL for our API?"
/debate:start "Monorepo vs polyrepo" --context src/package.json,docs/architecture.md
/debate:start "Next.js vs Remix vs Astro" --perspectives "Performance,DX,Ecosystem" --rounds 3
npx claudepluginhub jaebit/claudemate --plugin multi-model-debateOrchestrates structured multi-provider AI debates between Claude and available advisors (Gemini, Codex, etc.) for critical decisions. Dispatches real providers via orchestrate.sh for diverse perspectives.
Multi-LLM craftsmanship council with live progress and debate mode for code review and design questions
Orchestrates multi-agent debates with 2-5 dynamic agents in Challenge (select best variant), Strategy (deep analysis with proposals), or Critic (find weaknesses) modes. Triggers on debate, challenge, compare, critique prompts.