skills.rest
gener8v
gener8vOFFICIAL
Explore all skills by this author →
WebsiteGitHub

code-review

Reviews delivered tickets against specifications, constraints, and architecture decisions for traceability compliance.

3|Updated Feb 18, 2026
Search Tags:#code review#requirements#verification#compliance#acceptance criteria#traceability#ticket
One-click install
npx skills add https://github.com/gener8v/gener8v.claude-skills --skill code-review-gener8v
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/gener8v/gener8v.claude-skills/tree/main/skills/code-review
Command: npx skills add https://github.com/gener8v/gener8v.claude-skills --skill code-review-gener8v

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve? Verifying that delivered code actually matches what a ticket, specification, and technical design asked for is tedious and error-prone, especially when traceability must be proven requirement by requirement. This Skill automates that compliance review, producing traceability tables, severity-ranked findings, and a verdict. ## Core Features & Use Cases - Traceability Verification: Checks acceptance criteria, requirement coverage, constraint compliance, architecture decision adherence, and @spec annotations against the delivered code. - Two-Phase Review: A findings phase that can run in a fresh-context reviewer agent in parallel with quality and security reviews, followed by an interactive resolution phase where findings are approved, modified, deferred, or dismissed. - Verification Re-run: Re-executes the delivery record's Verification Run commands and flags non-reproducing results as Critical findings. - Use Case: After a ticket is delivered, ask for a review of TICKET-003 and receive a report with requirement-to-code traceability tables, findings with severities, and an Approved / Changes Required verdict written to the change's reviews directory. ## Quick Start Review the delivery of TICKET-003 in the current change against its ticket, specification, constraints, and technical design.

Frequently Asked Questions about code-review

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

FAQPage Schema
How do I review a delivered ticket against its specification?▼

Invoke the review with the ticket ID, such as 'review TICKET-003'. The Skill reads the delivery record, ticket, specification, constraints, and technical design, then produces traceability tables and findings with a verdict.

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

Code review asks whether you built the right thing: ticket compliance, requirement coverage, and constraint adherence. Quality review asks whether you built it well: readability, tests, and maintainability. They run in parallel as separate reviewer agents.

Can the findings phase run in parallel with other reviews?▼

Yes. The findings phase runs as the code-reviewer agent in a fresh context, in parallel with quality-review and security-review. Resolution phases run one at a time in the main session so reviewers never edit the same file concurrently.

What happens if the delivery record is missing?▼

The review stops and recommends running the Delivery skill first, since there is no record of what was built. If constraints or technical design are missing, the review proceeds but marks those sections as not available.

Why is a failed verification run a Critical finding?▼

The review re-executes the delivery record's Verification Run commands and compares exit codes. A record with no verification run, a non-Delivered status, or commands that no longer pass means the code was never proven, which blocks approval.

When should I not use this code review skill?▼

Do not use it for readability or test quality alone (use quality-review), for vulnerabilities (use security-review), or for code with no ticket behind it. Every finding must trace to a specific pipeline artifact.

Related Skills

Explore More Software Engineering Skills→
avibebuilderavibebuilder
113

give-plan

Create a reviewable implementation plan from repository files before coding.

Official
Intermediate
PastaSusPastaSus

bmad-loop-resolve

Resolves CRITICAL escalations in paused bmad-loop runs by updating frozen specs with human decisions.

Community
Advanced
namhoanggnamhoangg

subagent-driven-development

Dispatch fresh subagents per task with two-stage spec and code reviews.

Community
Advanced
windyroadwindyroad
4

wr-itil:work-problem

Select the highest-WSJF ticket and delegate its workflow to manage-problem.

Official
Advanced