quality-review

Review delivered code for engineering quality, readability, error handling, and test coverage.

3|Updated Feb 18, 2026
One-click install
npx skills add https://github.com/gener8v/gener8v.claude-skills --skill quality-review-gener8v
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: quality-review
Source: https://github.com/gener8v/gener8v.claude-skills/tree/main/skills/quality-review
Command: npx skills add https://github.com/gener8v/gener8v.claude-skills --skill quality-review-gener8v

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and assets (resource) components.

What problem does it solve? Delivered code often passes its specification while still being hard to maintain, poorly tested, or inconsistent with the codebase's established patterns. This Skill evaluates engineering quality independently of the spec, producing a structured review report with severity-rated findings and an interactive resolution workflow. ## Core Features & Use Cases - Five-category quality assessment: Rates code organization, readability, error handling, test coverage, and observability/operability, calibrated to the project's maturity and existing conventions. - Structured findings report: Writes a report to .gener8v/changes/<change-slug>/reviews/ with Critical/Concern/Suggestion findings, a resolution log, and a verdict (Approved / Approved with Notes / Changes Required) that pipeline tooling can parse. - Two-phase workflow: Findings run in a fresh reviewer context in parallel with code-review and security-review; resolution runs interactively with the user, applying approved fixes and re-verifying them. - Use Case: After a ticket is delivered, run the review to catch a missing test for a safety bound, flag swallowed errors, and block the ticket with Changes Required until tests exist. ## Quick Start Review the delivery of TICKET-001 in the active change for engineering quality and test coverage.

Frequently Asked Questions about quality-review

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

FAQPage Schema
How do I review code quality after a ticket is delivered?▼

Invoke the review with the ticket ID, optionally naming the change slug. It reads the delivery record for the file list, assesses five quality categories, writes a findings report, and walks you through resolving each finding interactively.

What is the difference between quality review and code review?▼

Quality review evaluates engineering quality — readability, error handling, tests, patterns — independent of the specification. Code review checks the delivery against its ticket and spec traceability. They run in parallel on the same code.

Can I review files that have no ticket or delivery record?▼

Yes. Point the review directly at code files and it degrades gracefully, inferring conventions from the surrounding codebase. You must specify where the report goes, and applied changes are recorded in the resolution log instead of a delivery record.

What happens when delivered code has no tests?▼

A ticket with testable acceptance criteria and no tests receives a Critical finding and a Changes Required verdict, since nothing about the ticket is proven. The verdict line is parsed by pipeline state tooling to hold the ticket.

Why is my quality review report not detected by the pipeline?▼

The state script looks for an exact path: `.gener8v/changes/<change-slug>/reviews/<area-slug>-ticket-nnn-quality-review.md` with lowercase ticket-nnn. Any other name or directory is invisible, so rename or move the file rather than duplicating it.