skills.rest
mor-duongmh
mor-duongmhCOMMUNITY
Explore all skills by this author →
GitHub

deep-review

Merge parallel specialist review findings from GitHub PRs or diffs into a Markdown report.

3|Updated Apr 20, 2026
Search Tags:#security#github#multi-agent#diff#code-review#report#risk
One-click install
npx skills add https://github.com/mor-duongmh/morkit-plugin --skill deep-review-mor-duongmh
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: deep-review
Source: https://github.com/mor-duongmh/morkit-plugin/tree/main/plugins/morkit/skills/deep-review
Command: npx skills add https://github.com/mor-duongmh/morkit-plugin --skill deep-review-mor-duongmh

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill coordinates a comprehensive, multi-agent review of a targeted PR or git diff, turning scattered findings into a single, coherent report.

Core Features & Use Cases

  • Dispatches five specialist subagents (risk, security, pattern, tests, convention) in parallel to assess changes across risk, architecture, tests, and compliance.
  • Ingests diffs from PRs, git diffs, or working-tree changes and synthesizes a Markdown matrix report suitable for code owners and stakeholders.
  • Honors CLAUDE.md as the highest source of truth when present and uses language-specific profiles to tailor checks.

Quick Start

Invoke the skill with a target PR or diff to generate a consolidated review report.

Frequently Asked Questions about deep-review

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

FAQPage Schema
How do I run a multi-agent code review on a GitHub PR?▼

To run a multi-agent code review on a GitHub PR, provide the PR or diff as input. The skill ingests the changes and dispatches five specialist agents in parallel to assess risk, security, patterns, tests, and conventions, synthesizing the results into a single Markdown report.

What does a multi-agent code review report cover?▼

A multi-agent code review report covers five core areas: risk, security, pattern, tests, and conventions. It merges findings from these specialist subagents into a coherent Markdown matrix suitable for code owners and stakeholders.

Can I use my own project conventions for automated code reviews?▼

Yes, you can use your own project conventions for automated code reviews. The skill honors CLAUDE.md as the highest source of truth when present and uses language-specific profiles to tailor its checks.

Does the deep-code review work with uncommitted working-tree changes?▼

Yes, the deep-code review works with uncommitted working-tree changes. It ingests diffs from GitHub PRs, git diffs, or working-tree changes to process and merge findings into a comprehensive report.

How to consolidate scattered security and risk findings from a git diff?▼

To consolidate scattered security and risk findings from a git diff, use a phase-based synthesis process. The skill ingests the diff, dispatches parallel specialist agents, and synthesizes their structured outputs into a single coherent report.

Related Skills

Explore More Software Engineering Skills→
windmill-labswindmill-labs
17.5k

pr

Create draft GitHub pull requests with conventional commit titles.

Official
Advanced
mahg-esmahg-es
3

cicd-quality

Design and enforce CI/CD quality gates for code, tests, security, build, E2E, and performance.

Official
Advanced
jayhjenkinsjayhjenkins

Jay-Dev QA Testing Skill

Automate three-tier QA testing with TEST_INDEX.md and RUNS/ reports.

Community
Intermediate
swannysecswannysec
2

security-vuln-analyzer

Orchestrates parallel security agents to validate vulnerabilities and generate remediation plans.

Community
Advanced