corgispec-review

Review completed Task Groups and post evidence-based reports to GitLab issues.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It prevents low-quality or incorrect AI-generated changes from moving forward by running evidence-based quality checks and requiring a deliberate human decision at the review gate.

Core Features & Use Cases

  • Evidence-first GitLab review: Selects the completed Task Group, runs quality checks, and posts a structured Review Report as GitLab issue notes.
  • Human-controlled approve/reject: Forces an interactive decision (Approve, Reject, or Discuss) and only applies workflow mutations after the user chooses.
  • Worktree-aware change discovery: Supports isolation modes, including resolving and reviewing changes inside git worktrees to ensure the review targets the actual artifacts.

Quick Start

Use the corgispec-review skill to review a completed Task Group for change testing and quality evidence before advancing it.

Frequently Asked Questions about corgispec-review

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

FAQPage Schema
How do I enforce human approval for AI coding changes in GitLab?

GitLab code review automation for AI changes works by gathering evidence across code quality, spec coverage, security, and performance, then posting a structured Review Report to the child issue and requiring a deliberate human decision before advancing labels.

What is a human approval gate in workflow automation?

A human approval gate in workflow automation is a mandatory checkpoint that blocks low-quality AI-generated changes from moving forward until a user makes a deliberate decision to Approve, Reject, or Discuss the evidence-based review findings.

How do I review code changes isolated in a git worktree?

To review code changes isolated in a git worktree, the review process reads OpenSpec configuration, resolves the worktree isolation mode, and targets the actual artifacts inside the worktree to ensure quality checks evaluate the correct changes.

Does code review automation auto-decide workflow outcomes for task groups?

Code review automation for task groups does not auto-decide workflow outcomes; it executes quality-check procedures and only applies workflow label mutations after the user explicitly chooses to Approve, Reject, or Discuss the review report.

What checks are required before advancing GitLab issue labels for completed tasks?

Before advancing GitLab issue labels for completed tasks, the system requires evidence-based checks across code quality, spec coverage, functional verification, architecture, security, and performance to validate the AI coding changes.