deep-review

Coordinate parallel code review lenses with isolated contexts for structured findings.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/vr000m/skills.md --skill deep-review-vr000m
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-review
Source: https://github.com/vr000m/skills.md/tree/main/.claude/skills/deep-review
Command: npx skills add https://github.com/vr000m/skills.md --skill deep-review-vr000m

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Deep Review provides a rigorous, multi-lens code review by coordinating parallel subagents each with a fresh context to uncover issues missed by single-pass reviews.

Core Features & Use Cases

  • Parallel lens execution with isolated contexts for logic, security, spec compliance, architecture, and documentation.
  • Target-driven reviews that accept plan files or PR diffs and support progressive resumes via --continue or full replays.
  • Structured output including per-lens findings, statuses, and a unified report.

Quick Start

Provide a plan path or PR reference and run the deep-review workflow to start the multi-lens assessment.

Frequently Asked Questions about deep-review

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

FAQPage Schema
How does multi-lens code review with parallel subagents improve finding security and logic issues?

Parallel subagents run isolated contexts for logic, security, spec compliance, architecture, and documentation, uncovering issues missed by single-pass code reviews.

What is the best way to review a PR diff for architecture and spec compliance?

Provide a PR diff reference to a multi-lens code review workflow to generate structured architecture and spec compliance findings using parallel subagents.

Can I resume a code review workflow if it stops before finishing all lenses?

Yes, you can resume the code review workflow using the --continue flag for progressive resumes, or perform a full replay to restart all parallel lenses.

Does multi-lens code review work with plan files or only PR diffs?

Multi-lens code review accepts both plan files and PR diffs as target inputs to drive parallel subagent assessments and generate structured findings.

What dependencies do I need to run parallel AI-assisted code review lenses?

No dependencies are required to run parallel AI-assisted code review lenses; the workflow coordinates isolated subagents and per-lens state persistence internally.