review

Analyze diffs against the base branch for SQL safety and trust boundary risks.

Updated Mar 29, 2026
One-click install
npx skills add https://github.com/filipnyquist/caidos --skill review-filipnyquist
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review
Source: https://github.com/filipnyquist/caidos/tree/main/.claude/skills/gstack/review
Command: npx skills add https://github.com/filipnyquist/caidos --skill review-filipnyquist

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Pre-landing PR reviews are error-prone and time-consuming. This skill streamlines the process by analyzing diffs for critical risks such as SQL safety, LLM trust boundary violations, and conditional side effects, helping reviewers catch issues early.

Core Features & Use Cases

  • Automated diff analysis for SQL safety to prevent unsafe interpolations and injections.
  • LLM trust boundary checks to ensure prompts and outputs stay within safe, auditable boundaries.
  • Clear risk flags and remediation suggestions to guide reviewers toward safe, timely decisions.
  • Use Case: When reviewing a PR with database changes, the skill highlights potential SQL injection risks and prompts for additional tests.

Quick Start

Provide the current diff against the base branch to begin an automated pre-landing review.

Frequently Asked Questions about review

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

FAQPage Schema
How do I automate PR review for SQL safety and LLM trust boundary issues?

Automated PR review for SQL safety and LLM trust boundaries works by analyzing your diff against the base branch to surface injection risks, trust breaches, and side effects, generating a remediation plan for reviewers.

What is pre-landing diff analysis and how does it detect code risks?

Pre-landing diff analysis evaluates code changes before merging to detect risks like unsafe SQL interpolations, LLM trust boundary violations, and conditional side effects. It provides clear risk flags to guide reviewers through remediation and decision points.

How do I check a diff for SQL injection risks before landing a PR?

To check a diff for SQL injection risks before landing a PR, provide the current diff against the base branch. The analysis highlights potential unsafe interpolations and prompts for additional tests to ensure database query safety.

Does automated code review work for catching unintended side effects in diffs?

Yes, automated code review works for catching unintended side effects by analyzing diffs against the base branch. It targets conditional side effects and trust boundary breaches, providing clear signals for potential risks and improvement opportunities.

Can I use diff analysis to enforce LLM trust boundaries in code reviews?

Yes, you can use diff analysis to enforce LLM trust boundaries in code reviews. The skill checks that prompts and outputs stay within safe, auditable boundaries, flagging trust boundary violations for remediation before landing.

Do I need to provide the full repository to run a pre-landing code review?

No, you do not need to provide the full repository to run a pre-landing code review. You only need to provide the current diff against the base branch to begin the automated analysis and receive risk flags.