deep-review

Run four-pass code review with parallel security, quality, and structural analysis.

5|Updated Mar 16, 2026
One-click install
npx skills add https://github.com/RobinNorberg/oh-my-copilot --skill deep-review-robinnorberg
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-review
Source: https://github.com/RobinNorberg/oh-my-copilot/tree/main/skills/deep-review
Command: npx skills add https://github.com/RobinNorberg/oh-my-copilot --skill deep-review-robinnorberg

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Deep Review structures code review as 4 specialized passes: Security (pass 1), Quality (pass 2), Structural (pass 3), and Validation (pass 4). Passes 1-3 run in parallel for independent analysis. The result is a consolidated, severity-rated review with validated findings.

Core Features & Use Cases

  • Parallel security, quality, and structural analysis passes
  • Validation pass to filter false positives
  • Consolidated markdown report with severity ratings
  • Customizable scope from diffs/PRs/files
  • Output includes validationStatus for each finding

Quick Start

Initiate a deep-review by invoking /deep-review on your PR or changeset.

Frequently Asked Questions about deep-review

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

FAQPage Schema
How do I perform a multi-pass code review for a large PR with 10+ files?

Multi-pass code review for large PRs analyzes security, quality, and structural dimensions in parallel, followed by a validation pass. This structured approach filters false positives and outputs a consolidated markdown report with severity-rated findings.

What is the best way to automate security and quality validation for significant code changes?

Automated security and quality validation for significant code changes runs specialized parallel analysis passes. It applies architecture and security checks across diffs or files, ultimately producing validated findings and a markdown report.

Can I get a consolidated markdown report with severity ratings from an automated code review?

Yes, automated code review can generate a consolidated markdown report with severity ratings. The process executes parallel analysis passes and a final validation step to output structured findings and a validation status for each issue.

Does multi-pass code review help reduce false positives in automated security analysis?

Multi-pass code review reduces false positives in automated security analysis by applying a dedicated validation pass. This final step filters the initial parallel security, quality, and structural findings before generating the final markdown report.

When do I need to use a structured multi-pass approach for reviewing code diffs?

A structured multi-pass approach for reviewing code diffs is needed when handling significant changes or large PRs with 10+ files. It enables parallel security, quality, and structural analysis to consolidate findings effectively without missing critical issues.