Detects SQL injection, LLM trust boundary violations, race conditions, and enum gaps in pull request diffs.
npx skills add https://github.com/a29paul/hexlens --skill review-a29paul
Please help me install this Agent Skill. Skill: review Source: https://github.com/a29paul/hexlens/tree/main/.claude/skills/gstack/review Command: npx skills add https://github.com/a29paul/hexlens --skill review-a29paul
This Skill solves the problem of shipping code with critical safety issues, SQL injection vulnerabilities, LLM trust boundary violations, and race conditions that are easy to miss in manual review. It automates the detection of structural issues in pull request diffs before code lands on the main branch.
Use the review skill to analyze the current git diff against the base branch and report any critical SQL safety, LLM trust boundary, or race condition issues found in the changed files.
High-intent search queries and answers about installing and using this skill.
To detect SQL injection vulnerabilities in a pull request diff, analyze the changed files for string interpolation in SQL queries and N+1 query patterns before merge. This prevents unsafe data access code from landing on the main branch.
To check for missing enum handlers during code review, trace any new enum values through all consumers across the codebase. This cross-file tracing ensures every branch handles the new status correctly, preventing runtime crashes.
An LLM trust boundary validation checks unsanitized LLM outputs written to databases or passed to mailers. Validating it before merge identifies SSRF risks from LLM-generated URLs and prevents malicious content from executing.
You can automate race condition detection for pre-landing review by analyzing pull request diffs against the base branch. This process specifically targets TOCTOU race conditions to catch critical safety issues before code lands.
Pre-landing code review works without external dependencies by parsing git diffs directly and applying pattern matching for SQL, LLM, and shell injection vectors. It requires only the pull request diff to function.
The best way to prevent unsafe code from shipping is automating structural issue detection in pull request diffs. This catches critical safety issues, injection vectors, and race conditions that are easy to miss in manual review.
huichenAnalyze branch diffs for structural issues and classify plan items.
akshatdaltonReview GitHub Pull Requests for architecture, code quality, and safety concerns.
yanochka11Analyze repository diffs to identify defects, security risks, and structural problems.
robertzengcnReview pull request diffs for critical safety, security, and correctness issues.
