code-review

Enforces evidence-based code reviews before merges and completion claims.

8|2|Updated Jul 13, 2015

Search Tags:#verification #subagent #quality-assurance #process #code-review #rigor #pull-request

Ready to run

npx skills add https://github.com/tstapler/dotfiles --skill code-review-tstapler

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/tstapler/dotfiles/tree/main/.claude/skills/code-review
Command: npx skills add https://github.com/tstapler/dotfiles --skill code-review-tstapler

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Code reviews are often rushed or vague; this Skill enforces a disciplined approach that centers on technical rigor, verified evidence, and clear justification before any completion claims.

Core Features & Use Cases

Receiving feedback protocol: READ → UNDERSTAND → VERIFY → EVALUATE → RESPOND → IMPLEMENT to ensure feedback is technically sound and actionable.
Requesting reviews: Dispatch the code-reviewer subagent after each task or major feature, with BASE_SHA, HEAD_SHA, and plan details to surface issues early.
Verification gates: Run verification commands and provide fresh evidence before claiming success, preventing false completion.

Quick Start

Apply the three-step process: receive feedback with technical rigor, request a code-review, and verify evidence before claiming completion.

Frequently Asked Questions about code-review

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

FAQPage Schema

How do I enforce evidence-based code reviews before merging a pull request?▼

A formal feedback reception protocol ensures code review rigor by requiring reviewers to READ, UNDERSTAND, VERIFY, EVALUATE, RESPOND, and IMPLEMENT feedback. This structured process guarantees feedback is technically sound and actionable before any development continues.

How do I trigger a code review after completing a major feature?▼

A verification gate prevents false completion claims by requiring developers to run verification commands and provide fresh evidence before claiming success. This enforces evidence-based status claims and stops rushed or vague code reviews from approving incomplete work.

What is the best way to structure feedback reception during a code review?▼

The best way to structure feedback reception is using a formal protocol: READ, UNDERSTAND, VERIFY, EVALUATE, RESPOND, and IMPLEMENT. This ensures all code review feedback is technically sound, evaluated properly, and actionable before implementation.

Does subagent-driven development work with automated pull request reviews?▼

Yes, subagent-driven development integrates with pull request reviews by dispatching a code-reviewer subagent after each task. The subagent uses BASE_SHA and HEAD_SHA references to evaluate changes and surface technical issues before merges.