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

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.

Prerequisites

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.

Installation

Requirements and caveats from upstream:

• Prerequisites: GitHub CLI (gh) installed and authenticated, open PR on current branch.

Basic usage or getting-started notes:

• The dashboard is the recommended way to run reviews, browse results, and manage your workflow. Launch it with ocr dashboard.

• Run reviews and maps

• <img src="assets/ocr-tool-example-translated-human-review.png" alt="Human-voice review posted to GitHub PR" width="700" />

• Source: https://github.com/spencermarx/open-code-review

• Extracted from upstream docs: https://raw.githubusercontent.com/spencermarx/open-code-review/HEAD/README.md

Documentation

• https://github.com/spencermarx/open-code-review#readme

Source

• Agent Skill Exchange