review

Reviews pull requests for SQL safety, LLM trust boundary violations, conditional side effects, and structural issues.

1|Updated May 8, 2026
One-click install
npx skills add https://github.com/npc-chris/dfn-discovery --skill review-npc-chris
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review
Source: https://github.com/npc-chris/dfn-discovery/tree/main/.agents/skills/gstack/review
Command: npx skills add https://github.com/npc-chris/dfn-discovery --skill review-npc-chris

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires git, grep, python, ruby, node.js, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill performs a comprehensive review of pull requests before merging, identifying potential issues such as SQL safety, LLM trust boundary violations, conditional side effects, and other structural issues.

Core Features & Use Cases

  • PR Review: Analyzes the diff against the base branch for potential issues.
  • SQL Safety: Checks for string interpolation in SQL, TOCTOU races, bypassing model validations, and N+1 queries.
  • LLM Trust Boundary: Validates format validation of LLM-generated values and structured tool output.
  • Race Conditions & Concurrency: Detects read-check-write without uniqueness constraints, find-or-create without unique DB index, and unsafe HTML rendering.
  • Use Case: Use this Skill when you need to ensure the safety and reliability of your code before merging into the main branch.

Quick Start

Run the review skill to analyze the current PR's diff against the base branch.

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 for SQL injection and N+1 query issues before merging?

To check a pull request for SQL safety, run a pre-landing PR review that analyzes the codebase diff to detect string interpolation in SQL, TOCTOU races, and N+1 queries. It identifies bypassed model validations to prevent database vulnerabilities before merging.

What is an LLM trust boundary violation in code review?

An LLM trust boundary violation occurs when LLM-generated values lack proper format validation or structured tool output is unvalidated. A pre-landing PR review detects these violations by analyzing the diff to ensure LLM outputs are safely constrained before merging into the main branch.

How do I detect race conditions and conditional side effects in a git diff?

To detect race conditions and conditional side effects in a git diff, run a PR review that checks for read-check-write operations without uniqueness constraints and find-or-create logic without unique database indexes. It also flags unsafe HTML rendering within the codebase changes.

Do I need git and Python installed to run an automated PR review for structural code issues?

Yes, you need git installed to analyze the diff against the base branch. You also need Python, Ruby, Node.js, and grep available in your environment to allow the PR review process to thoroughly analyze the codebase for structural issues.

What is the best way to automate code reviews for unsafe HTML rendering and database validations?

The best way to automate code reviews for unsafe HTML rendering and bypassed database validations is running a pre-landing PR review script. It evaluates the pull request diff against the base branch to catch these structural issues before code merges.