local-review

Run automated local review loops until WORKFLOW.md standards pass.

Updated May 14, 2026
One-click install
npx skills add https://github.com/libz-renlab-ai/Matrix --skill local-review-libz-renlab-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: local-review
Source: https://github.com/libz-renlab-ai/Matrix/tree/main/.claude/skills/local-review
Command: npx skills add https://github.com/libz-renlab-ai/Matrix --skill local-review-libz-renlab-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

本地 review 自动循环,按 WORKFLOW.md 的两项标准对变更进行本地评审,未通过时自动派打回 subagent 修正,循环至通过或达到最大重试上限。

Core Features & Use Cases

  • 自动触发本地 review 循环,通过 /local-review 或手动触发。
  • 依据 WORKFLOW.md 的两项标准:verdict=pass 与全量测试通过且无合并冲突,确保可回归的变更。

Quick Start

直接触发本地 review 循环,让系统自动重跑评审,直到通过或达到最大重试次数。

Frequently Asked Questions about local-review

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

FAQPage Schema
How do I automate local code review loops until all tests pass?

Automated local code review loops run by dispatching subagents to fix changes until WORKFLOW.md standards are met, enforcing a passing verdict, completed tests, and no uncommitted changes.

How does a subagent fix code changes during an automated local review?

During an automated local review, a subagent fixes code changes by addressing failures identified against WORKFLOW.md criteria, automatically re-running the review loop until tests pass or the retry limit is reached.

What criteria must be satisfied for a local code review to pass?

For a local code review to pass, the review artifact must record a verdict of pass, and all tests must pass with no uncommitted changes or branch conflicts.

Can I trigger the automated review loop manually after feature completion?

Yes, you can trigger the automated review loop manually after feature completion by invoking the /local-review command to enforce WORKFLOW.md standards.

What happens when code changes fail to meet review criteria repeatedly?

When code changes fail to meet review criteria repeatedly, the system loops the subagent correction process until a passing verdict is achieved or the maximum retry limit is reached.

Does the local review process check for uncommitted changes and branch conflicts?

Yes, the local review process checks for uncommitted changes and branch conflicts, ensuring the codebase is clean and all tests pass before recording a passing review verdict.