# Agent Review Panel

> Orchestrate a multi-agent adversarial review panel: subagents with different perspectives independently review, debate and discuss, reach or fail consensus, then a supreme judge renders the verdict. Trigger on "review panel", "multi-agent review", "adversarial review", "panel review", "have agents debate this", "get different opinions on this code/plan/doc", "multiple reviewers", "independently evaluate", "thorough feedback from different angles", "stress-test this design", "red team this", "devil's advocate perspective", "fresh eyes on this", "a second/third opinion", "argue pros and cons", "every angle covered", "critical look from security and performance angles", "high-stakes", or /agent-review-panel. NOT for single-reviewer review, quick sanity checks, bug fixes, deployment, existing PR comments, skill improvement, peer review, code explanation, or writing tests. Modes, also /agent-review-panel arguments: deep research ("deep review", "thorough review", "research review") researches domain best practices; multi-run union ("multi-run review", "run twice", "run 3 times", "maximum coverage") repeats with rotated personas, merged by stability; data-flow trace tiers ("exhaustive review", "trace everything", "catch all bugs") add a pre-review data-path trace; budget ("budget mode", "budget review", "cheap review", "low-cost panel", "token-efficient review", "keep the cost down", "without burning tokens") is the same protocol at reduced cost. Uses specialist domain-expert reviewers.

## Facts
- Page: https://tashan.sh/capability/skill-wan-huiyan-agent-review-panel
- tashan id: skill:wan-huiyan/agent-review-panel
- Source: https://github.com/wan-huiyan/agent-review-panel
- Type: skill
- Category: other
- tashan score: not scored (catalogued only — too little public evidence)
- Adoption: 9.0
- Upkeep: 99.0
- Freshness: 97.0
- Evidence coverage: 84% of the inputs this score can use
- Health: active
- Instruction depth: not yet graded
- License: MIT
- Official: no

## Install

```sh
cp -r agent-review-panel ~/.claude/skills/
```

## Security audit
Not scanned. We audit npm-published capabilities; this one has no npm package we can resolve, or has not reached the queue. This is not a clean bill of health.

---
Measured 2026-08-14 by tashan (https://tashan.sh) from public evidence. Scorer s5.