Reviews code changes against a project checklist and generates structured review reports.
npx skills add https://github.com/PiLastDigit/TRIP-workflow --skill trip-review
Please help me install this Agent Skill. Skill: TRIP-review Source: https://github.com/PiLastDigit/TRIP-workflow/tree/main/skills/TRIP-review Command: npx skills add https://github.com/PiLastDigit/TRIP-workflow --skill trip-review
Code reviews often lack consistent criteria and produce unstructured feedback that is hard to audit later. This Skill provides a manual fallback and audit path for reviewing code against a single-source-of-truth checklist, producing standardized review records.
docs/3-code-review/CR_wa_vx.y.z.md with findings, checklist outcomes, and an approval verdict.Ask the AI to review version 1.2.0 using the TRIP-review checklist and save the review report to the docs folder.
High-intent search queries and answers about installing and using this skill.
Invoke the review skill with a version or feature argument, then walk each checklist section against the change. Failing items become findings classified by severity, and results are saved as a structured markdown file in docs/3-code-review/.
A solid checklist covers functional requirements, code quality, architectural compliance, error handling, security, and performance. This Skill's checklist also defines severity levels from Critical to Suggestions and an approval gate requiring no open critical or major issues.
Use this manual path to audit past versions, review unplanned work, or replace the automated Codex review loop when it is unavailable. Normal reviews happen automatically inside the TRIP-2-implement phase.
Findings are classified as Critical (block deployment, e.g., security vulnerabilities), Major (require immediate fix, e.g., incorrect logic), Minor (should fix, e.g., style issues), or Suggestions (nice to have, e.g., optimizations).
Review records are saved to docs/3-code-review/ using the naming pattern CR_wa_vx.y.z.md, where a is the project week and x.y.z is the version. The report follows a standard template with findings, checklist outcomes, and a verdict.
The-Vibe-CompanyImplement work from a spec or tickets using TDD and code review.
t0labAnalyze Conventional Commits to draft changelogs and create release commits and tags.
code-review-benchmarkAutomate local diff reviews with CodeRabbit until zero findings remain.
Dev10x-GuruAudit merged GitHub Pull Requests for unresolved review threads and create follow-up issues.