review

Analyze branch diffs for SQL safety, trust boundary, and side-effect issues.

Updated Apr 3, 2026
One-click install
npx skills add https://github.com/lyh2/gstack-deepseek --skill review-lyh2
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review
Source: https://github.com/lyh2/gstack-deepseek/tree/main/review
Command: npx skills add https://github.com/lyh2/gstack-deepseek --skill review-lyh2

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The Review Skill provides a comprehensive, automated analysis of a branch's diff against the base to surface structural issues, data risks, and policy violations that tests may miss, helping teams merge confidently.

Core Features & Use Cases

  • Two-pass workflow for critical checks and informational checks to improve code quality and safety.
  • Detect scope drift, missing requirements, and architectural concerns by comparing the diff to the intended plan.
  • Integrate telemetry, checklists, and design-review steps to create a traceable pre-landing review process.

Quick Start

Run /review on a feature branch to generate a pre-landing assessment of the diff.

Frequently Asked Questions about review

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

FAQPage Schema
How do I automate pull request reviews to catch SQL safety issues?

Automated pull request review analyzes a branch diff against its base to detect SQL safety issues, LLM trust boundary violations, and conditional side effects before merging. It runs a two-pass workflow to surface critical and informational code quality gaps.

What is diff analysis in code review and when do I need it?

Diff analysis compares a feature branch against the base to surface structural issues, data risks, and policy violations that automated tests may miss. You need it during pre-landing PR reviews to catch scope drift, missing requirements, and architectural concerns.

How do I check my diff for missing requirements and scope drift before merging?

Run a pre-landing review on your feature branch to compare the diff against the intended plan. This detects scope drift and missing requirements, providing proactive recommendations and flagging gaps in tests or design before you merge.

Can I integrate telemetry and checklists into my pre-landing PR review workflow?

Yes, you can integrate telemetry, checklists, and design-review steps into the pre-landing PR review workflow. This creates a traceable review process that provides proactive recommendations and flags gaps in tests or design as configured.

Does automated code review work for LLM trust boundary violations?

Automated code review detects LLM trust boundary violations by analyzing the branch diff for potential problems before merging. It applies a two-pass workflow to check critical safety issues and informational code quality concerns.

What are the limitations of automated diff analysis for code review?

Automated diff analysis focuses on structural issues, data risks, and policy violations found in the branch diff. It supplements but does not replace human review, acting as a proactive check to surface potential problems before merging.