What is multi-agent-debate-review?

A codebase review process that uses several specialized agents to inspect a project, compare their findings, and debate disagreements before producing one report. The review can cover code quality, architecture, and security.

Comprehensive reviews before releases, audits after large changes, security checks, architecture assessments, and prioritized follow-up actions.

It reduces the chance that one reviewer misses an important problem or reaches a conclusion without checking other viewpoints.

multi-agent-debate-review — Skill by mvilrokx/claude-skills

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent

npx skills add mvilrokx/claude-skills --skill multi-agent-debate-review

Clone the repo

git clone --depth 1 https://github.com/mvilrokx/claude-skills

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/mvilrokx/claude-skills/multi-agent-debate-review/github.svg)](https://agentmods.dev/skills/mvilrokx/claude-skills/multi-agent-debate-review)

Your own site

<a href="https://agentmods.dev/skills/mvilrokx/claude-skills/multi-agent-debate-review"><img src="https://agentmods.dev/badge/skills/mvilrokx/claude-skills/multi-agent-debate-review/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/mvilrokx/claude-skills/multi-agent-debate-review"><img src="https://agentmods.dev/badge/skills/mvilrokx/claude-skills/multi-agent-debate-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 104 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,825 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00104	$0.01825
Opus 5.5	$0.00042	$0.00730
Sonnet 5.5	$0.00021	$0.00365
Haiku 4.5	$0.00010	$0.00183

Measured 2mo ago against content hash cbeecf34be33, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

multi-agent-debate-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

multi-agent-debate-review/SKILL.md · 195 lines

How it starts

The opening of the file, as written. The whole thing — 195 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Multi-Agent Debate Review

Run 6+ specialized agents in parallel against a codebase, compare findings, re-invoke agents on points of disagreement, and produce a consolidated report with prioritized actions.

Mode Selection

Ask the user which mode to use, or infer from context:

Direct mode (default): You coordinate — launch agents, collect results, run debates, write report. More interactive, allows user to discuss findings between phases.
Delegated mode: Launch a single general-purpose agent that coordinates the entire debate autonomously. Hands-off — user gets a final report. Use when user says "run it hands-off", "do it all automatically", or wants minimal interaction.

Direct Mode

Step 1: Load Standards

Load relevant skills to give agents better baselines. Select based on project language:

Go: coding-standards, golang-patterns, security-review, backend-patterns
Python: coding-standards, python-patterns, security-review
TypeScript/JS: coding-standards, frontend-patterns, security-review

Step 2: Launch Agents in Parallel

Launch all agents with mode: "background". Each agent gets a focused prompt with explicit instructions to provide severity ratings, file references, and suggested fixes.

Select agents based on the project. See references/agent-configs.md for the full prompt templates for each agent type.

Go projects: go-reviewer, go-security-reviewer, go-refactor-cleaner, code-reviewer, database-reviewer, architect
Python projects: python-reviewer, python-security-reviewer, python-refactor-cleaner, code-reviewer, database-reviewer, architect
JS/TS projects: code-reviewer, js-security-reviewer, js-refactor-cleaner, database-reviewer, architect

Always include code-reviewer, database-reviewer, and architect regardless of language.

Model Selection

Use the model parameter on each agent to assign different models. Two strategies:

Cost-optimized (default): Match model capability to task complexity.

Read the full file on GitHub · 195 lines

Files

What ships with it

2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2mo ago First seen · 195 lines · 104 tokens per session scan A cbeecf34be33

Subscribe to this mod's changes

multi-agent-debate-review is a skill published in the GitHub repository mvilrokx/claude-skills (2 stars, last pushed 8mo ago), licensed MIT. It adds 104 tokens to every session and 1,825 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Code review

multi-agent-debate-review next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
multi-agent-debate-reviewmvilrokx/claude-skills	Skill	2	8mo ago	MIT	A	104	Claude Code, Codex
autoreviewopenclaw/clawhub	Skill	9.5k	today	MIT	A	17	Claude Code, Codex
omh-code-reviewrlaope/oh-my-hermes	Skill	5.2k	today	MIT	A	48	Claude Code, Codex
agent-teams-simplify-and-hardenLeoYeAI/openclaw-master-skills	Skill	2.2k	6d ago	MIT	A	115	Claude Code, Codex
revdiff-planumputun/revdiff	Skill	947	2d ago	MIT	A	84	Claude Code
explore-unknownsdzhng/skills	Skill	1.4k	today	MIT	A	162	Claude Code, Codex
re0-mergeLilMGenius/paperthin	Skill	1.4k	4d ago	MIT	A	70	Claude Code