debate

Route artifacts across AI models for structured P1/P2/P3 critiques.

Updated Aug 27, 2026

Search Tags:#debate #code-review #artifact #cross-model #multi-round #AI-critique

Ready to run

npx skills add https://github.com/fancive/claude-skills --skill debate-fancive

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: debate
Source: https://github.com/fancive/claude-skills/tree/main/skills/debate
Command: npx skills add https://github.com/fancive/claude-skills --skill debate-fancive

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Orchestrates automated cross-model debates to surface issues, compare perspectives, and drive informed decisions on code, proposals, or documents.

Core Features & Use Cases

Multi-round critique routing and convergence between models.
Deterministic handling of mixed content (code + proposal) with phase-based evaluation.
Audit-ready session management with per-round inputs, critiques, and decisions.

Quick Start

Start a debate by typing /debate and providing an initial artifact or allowing auto-detection from the workspace.

Frequently Asked Questions about debate

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I run a cross-model debate to surface high-priority code review issues?▼

To run a cross-model debate, initiate the process and provide an initial artifact. The system auto-detects the artifact type and routes it through competing AI models to generate structured P1, P2, and P3 findings for your code review.

Can I use automated AI critique for mixed content containing both code and proposals?▼

Yes, automated AI critique supports mixed content with deterministic splitting. It separates code and proposal artifacts to evaluate them through phase-based evaluation, ensuring structured critique and policy-aligned improvements for both types.

What is multi-round critique routing and how does it help converge on document improvements?▼

Multi-round critique routing coordinates competing AI models to iteratively analyze documents. This process compares perspectives across rounds, enforcing a structured output rubric with a final decision to help teams converge on high-priority issues and improvements.

How do I start a multi-round critique session for a proposal artifact?▼

Start a critique session by invoking the debate command and providing your initial proposal artifact. The system supports auto-detection from the workspace, deterministically managing the artifact to produce audit-ready session inputs and decisions.

Does cross-model debate support audit-ready session management for code reviews?▼

Yes, cross-model debate enforces audit-ready session management. It tracks per-round inputs, critiques, and decisions, ensuring that code review sessions produce structured findings and policy-aligned improvements suitable for auditing.

What are the limitations of using automated AI critique for artifact evaluation?▼

Automated AI critique relies on deterministic splitting and phase-based evaluation for mixed content. While it enforces a structured P1/P2/P3 rubric, it requires clear artifact boundaries to effectively route and critique code, proposals, or documents.