review

Analyze git diffs against main for SQL safety and trust boundary issues.

5|2|Updated Mar 1, 2026
One-click install
npx skills add https://github.com/umairinayat/Specter-AI --skill review-umairinayat
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review
Source: https://github.com/umairinayat/Specter-AI/tree/main/.claude/skills/review
Command: npx skills add https://github.com/umairinayat/Specter-AI --skill review-umairinayat

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Pre-landing PR reviews often miss subtle SQL safety issues, data integrity risks, LLM prompt trust boundary violations, and unintended side effects in diffs. This Skill automates those checks against the main branch to reduce review toil and merge risk.

Core Features & Use Cases

  • Diff-based safety analysis: detects potential SQL injection risks, unsafe data writes, and to-be-changed dependencies across diffs.
  • LLM trust validation: flags prompt leakage, unsafe model configurations, and boundary violations before merge.
  • Structural consistency checks: identifies conditional side effects, missing validations, and risky code patterns in PRs.
  • Use Case: A developer submits a feature branch; reviewers receive a structured report highlighting critical issues to address before merging.

Quick Start

Run a pre-landing PR review on your feature branch to surface SQL, trust boundary, and side-effect issues before merging.

Frequently Asked Questions about review

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

FAQPage Schema
How do I check a git diff for SQL injection risks before merging a PR?

Pre-merge diff analysis checks the current git diff against the main branch to surface SQL injection risks and unsafe data writes before merging. It targets SQL safety by flagging potential injection vectors across the diff for actionable feedback.

How do I detect LLM prompt trust boundary violations in a pull request?

Diff-based safety analysis detects LLM trust boundary violations by flagging prompt leakage, unsafe model configurations, and boundary risks across the PR diff before merge. It validates trust boundaries to prevent unintended prompt exposure in production.

What is the best way to identify unintended side effects and data integrity risks in a feature branch?

Analyzing the feature branch diff against main identifies unintended side effects and data integrity risks by detecting conditional side effects, missing validations, and risky code patterns. This structural consistency check reduces merge risk and review toil.

Can I run pre-landing PR checks without adding external dependencies to my repository?

Pre-landing PR checks operate with no listed dependencies, running diff analysis directly against your main branch. You can integrate the structured feedback report into a standard review toolkit without installing external packages.

What types of code quality issues does diff-based PR analysis catch that manual reviews miss?

Diff-based PR analysis catches subtle SQL safety issues, data integrity risks, LLM prompt trust boundary violations, and unintended side effects across diffs that manual reviews often miss. It automates safety checks to reduce review toil and merge risk.

When should I avoid relying on automated diff analysis for pull request safety checks?

Automated diff analysis targets SQL safety, data integrity, and LLM trust boundaries, but does not replace human judgment for complex business logic or architectural decisions. Use it to surface critical issues, not as the sole merge gate for high-risk changes.