review

Identify SQL safety, LLM trust boundaries, and side effects in branch diffs.

107|19|Updated Mar 23, 2026
One-click install
npx skills add https://github.com/mr-daedalium/ostack --skill review-mr-daedalium
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review
Source: https://github.com/mr-daedalium/ostack/tree/main/review
Command: npx skills add https://github.com/mr-daedalium/ostack --skill review-mr-daedalium

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Pre-landing PR review automates the detection of SQL safety issues, LLM trust boundary violations, conditional side effects, and other structural risks in diffs between the feature branch and the base branch. Proactively surface guidance for merge readiness and provide concrete remediation suggestions.

Core Features & Use Cases

  • Diff-scoped analysis of PR changes to surface high-risk patterns before landing.
  • SQL safety checks to validate queries and data modifications against best practices.
  • LLM trust boundary validation to prevent leakage of prompts or sensitive information.
  • Detection of conditional side effects and potential race conditions across touched files.
  • Proactive escalation prompts and actionable recommendations for merge decisions.
  • Supports a two-pass workflow with actionable findings and fixed-first remediation suggestions.

Quick Start

Run the review skill on a PR branch to generate a pre-landing review report.

Frequently Asked Questions about review

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

FAQPage Schema
How do I automate pre-landing PR review for SQL safety and side effects?

Automating pre-landing PR review involves analyzing the diff between your current branch and base branch to identify SQL safety issues, LLM trust boundary violations, and conditional side effects. The process generates an actionable report with proposed fixes and escalation prompts.

What are LLM trust boundary violations in a pull request?

LLM trust boundary violations involve the potential leakage of prompts or sensitive information within code changes. Validating these boundaries during PR review detects structural risks early, ensuring sensitive data remains protected before code lands.

How do I check for race conditions and side effects across modified files?

Checking for race conditions and side effects requires diff-scoped analysis of touched files to detect conditional side effects and structural risks. This surfaces high-risk patterns across modified files, providing actionable remediation suggestions for merge readiness.

Does AI code review detect SQL injection risks in feature branch diffs?

AI code review detects SQL safety issues by validating queries and data modifications against best practices within the feature branch diff. It proactively surfaces high-risk SQL patterns and provides concrete remediation suggestions before merging.

What is a two-pass workflow for diff analysis and remediation?

A two-pass workflow for diff analysis first surfaces actionable findings, then generates fixed-first remediation suggestions. This approach ensures critical issues are identified before proposing concrete fixes, streamlining merge readiness decisions.

When do I need pre-landing code review for merge readiness?

Pre-landing code review is needed when your feature branch contains SQL modifications, LLM integrations, or logic with potential side effects. It proactively surfaces structural risks and provides escalation prompts to ensure safe merge decisions.