review

Analyze pull request diffs for SQL safety, LLM trust boundaries, and side effects.

1|Updated Mar 28, 2026
One-click install
npx skills add https://github.com/shhubbh/flowstate --skill review-shhubbh
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review
Source: https://github.com/shhubbh/flowstate/tree/main/.claude/skills/gstack/review
Command: npx skills add https://github.com/shhubbh/flowstate --skill review-shhubbh

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Review processes often lack a consistent, thorough check of diffs against the base branch, leading to missed SQL safety issues, ambiguous LLM prompts, and overlooked side effects.

Core Features & Use Cases

  • Analyzes diffs against the base branch to surface SQL safety violations, prompt injection risks, and conditional side effects.
  • Provides structured, actionable feedback that reviewers can apply directly in code reviews or pre-landing PR reviews.
  • Proactively suggests changes or questions to raise before merging, improving overall code quality and governance.

Quick Start

Inspect the current PR diff with this skill to receive a structured review and actionable recommendations.

Frequently Asked Questions about review

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

FAQPage Schema
How do I check a pull request diff for SQL safety issues?

To check a pull request diff for SQL safety issues, you analyze the PR changes against the base branch to identify potential violations before merging. This process surfaces SQL risks and provides structured, actionable feedback for code reviewers.

What is the best way to review LLM trust boundary violations in a diff?

Reviewing LLM trust boundary violations involves analyzing pull request diffs to detect prompt injection risks and ambiguous LLM prompts. It surfaces these risks across AI-related changes, providing structured decision criteria and questions to raise before merging.

How do I perform a pre-landing code review on SQL-heavy repos?

Performing a pre-landing code review on SQL-heavy repos requires analyzing diffs against the base branch to catch unintended side effects and SQL safety issues. It applies deterministic checks and test hints to ensure code quality before changes land.

Can I use automated PR review to find unintended side effects?

Yes, automated PR review can find unintended side effects by analyzing the pull request diff against the base branch. It identifies conditional side effects and provides actionable recommendations to address them before merging the code.

Does static analysis catch prompt injection risks in AI-related changes?

Static analysis catches prompt injection risks in AI-related changes by applying structured checks to the pull request diff. It identifies LLM trust boundary violations and generates actionable questions to resolve them during the code review.

When do I need structured checks for pull request reviews?

You need structured checks for pull request reviews when working with SQL-heavy repos or AI-related changes to ensure SQL safety and LLM trust boundaries are maintained. It provides deterministic steps and decision criteria to catch issues before merging.