corgispec-gh-review

Review completed GitHub Task Groups and generate evidence-based Review Reports.

113|9|Updated Apr 30, 2026
One-click install
npx skills add https://github.com/ricoyudog/Coding_Corgi_flow --skill corgispec-gh-review
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: corgispec-gh-review
Source: https://github.com/ricoyudog/Coding_Corgi_flow/tree/main/.opencode/skills/molecules/corgispec-gh-review
Command: npx skills add https://github.com/ricoyudog/Coding_Corgi_flow --skill corgispec-gh-review

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It resolves the uncertainty of whether an AI-completed Task Group is correct and review-ready by turning the review step into a structured, evidence-backed GitHub issues workflow with a mandatory human decision gate.

Core Features & Use Cases

  • Evidence-first review workflow: Selects the right change, verifies the target Task Group is completed, and presents existing GitHub feedback before proceeding.
  • Quality-check driven Review Report: Runs automated quality checks across code quality, spec coverage, functional verification, architecture, and performance risk detection, and compiles results into a Review Report for the human gate.
  • Strict review guardrails: Prevents auto-approve/auto-reject, avoids label changes and issue closure until the user explicitly approves, and never fabricates test/screenshot results.

Quick Start

Have your AI run the review for the current completed Task Group on GitHub and then ask you to approve, reject, or discuss based on the generated Review Report.

Frequently Asked Questions about corgispec-gh-review

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

FAQPage Schema
How do I automate GitHub code review with a mandatory human approval gate?

You can gate merges with a human approval gate by running an evidence-backed GitHub review that compiles code quality, spec coverage, and performance checks into a Review Report, requiring explicit interactive approve or reject input before proceeding.

What is evidence-based code review for GitHub issues?

Evidence-based code review is a workflow that discovers the correct change and Task Group, runs automated quality checks, and compiles functional verification evidence into a Review Report for human decision-making without auto-approving or auto-rejecting changes.

How to verify spec coverage and code quality before merging a Task Group?

To verify spec coverage and code quality before merging, run automated quality checks across architecture, performance risks, and functional evidence, then review the compiled results and enforce a mandatory interactive approve or reject gate.

Do I need GitHub CLI access to run quality checks on a completed Task Group?

Yes, you need GitHub CLI access to run quality checks, as the workflow reads .github.yaml and tasks.md for context, discovers GitHub Issues feedback, and evaluates spec coverage before enforcing the human approval gate.

Can I auto-approve a code review after quality checks pass on GitHub?

No, you cannot auto-approve a code review, as the workflow prevents auto-approve and auto-reject actions, avoids label changes and issue closure, and enforces a mandatory interactive human decision based on the generated Review Report.

Why does my GitHub review workflow prevent issue closure until manual approval?

Your GitHub review workflow prevents issue closure because it enforces strict review guardrails that avoid label changes and issue closure until you explicitly approve, ensuring a human decision gate based on evidence-backed quality checks.