Marketplace for trusted AI agent skills

Run multi-agent code review rounds with structured reviewer discourse before human approval

Skill Detail

Run multi-agent code review rounds with structured reviewer discourse before human approval

<p>Use Open Code Review when an agent needs several reviewer personas to inspect a diff, debate findings, and synthesize review output before a human approves, posts, or acts on the review.</p>

Code Quality & ReviewMulti-Framework
Code Quality & Review Multi-Framework Security Reviewed
⭐ 131 GitHub stars ⬇ 1.1k/wk npm
INSTALL WITH ANY AGENT
npx skills add agentskillexchange/skills --skill run-multi-agent-code-review-rounds-with-structured-reviewer-discourse-before-human-approval Copy
Works best when you want a reusable capability, not another fragile one-off prompt.
View source
At a glance
Tools required
Node.js 20+, Git, and a supported AI coding assistant such as Claude Code, Cursor, Windsurf, or another OCR-supported tool. GitHub CLI auth is also needed if you want to post reviews back to pull requests.
Install & setup
<p>Install the CLI with <code>npm install -g @open-code-review/cli</code>, run <code>ocr init</code> inside the target repository, then open <code>ocr dashboard</code> or invoke the review from your coding assistant with <code>/ocr:review</code> or <code>/ocr-review</code>. Use <code>ocr progress</code> to watch live execution and connect <code>gh</code> if you want OCR to post findings to pull requests.</p>
Author
Spencer Marx
Publisher
Individual
Last updated
Apr 14, 2026
Quick brief

Open Code Review gives an agent a narrow code-review workflow rather than a generic coding platform. It initializes a review project, analyzes a diff or requirements context, runs several reviewer agents in parallel, preserves their discourse, and produces structured findings plus a synthesized review that can be browsed in the dashboard or posted back to a pull request.

How it works

What this skill actually does

The scope stays tight: this is for multi-agent review rounds on code changes. It is not a generic IDE, not a broad AI coding assistant listing, and not a general CI platform card. Invoke it when the missing step is deeper review coverage and reviewer debate before human sign-off, not when the user simply wants another code editor or single-agent coding tool.