code-review-criteria

Generate task-specific code review criteria from ADRs and code conventions.

Updated Jul 1, 2026
One-click install
npx skills add https://github.com/Jsplix/ACODIAN-ORIGIN --skill code-review-criteria-jsplix
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review-criteria
Source: https://github.com/Jsplix/ACODIAN-ORIGIN/tree/main/.codex/skills/code-review-criteria
Command: npx skills add https://github.com/Jsplix/ACODIAN-ORIGIN --skill code-review-criteria-jsplix

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Manually compiling scattered architecture decision records (ADRs) and code convention rules into consistent, task-specific code review criteria is time-consuming and error-prone, especially for large cross-module pull requests.

Core Features & Use Cases

  • Relevant ADR Extraction: Filters module-specific ADR documents by the current task's tech stack to pull only applicable architectural decisions and their implementation requirements.
  • Convention Merging: Combines extracted ADR requirements with matching code convention rules to generate a unified, tailored evaluation guide.
  • Use Case: For a large API module PR, this skill automatically pulls only the ADRs related to the API's tech stack, merges them with the API's code conventions, and produces a ready-to-use review criteria document, eliminating the need to manually sift through dozens of unrelated ADRs.

Quick Start

Use the code-review-criteria skill to generate a tailored code review evaluation guide for your current module's pull request.

Frequently Asked Questions about code-review-criteria

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

FAQPage Schema
How do I generate task-specific code review criteria for a pull request?

To generate task-specific code review criteria, this skill consolidates relevant architecture decision records and code convention rules based on your current task's tech stack, producing a tailored evaluation guide for your pull request.

How does ADR extraction work for cross-module code reviews?

ADR extraction filters module-specific architecture decision records by the current task's tech stack, pulling only applicable architectural decisions and their implementation requirements while ignoring unrelated documents.

Do I need YAML files to generate code review evaluation guides?

Yes, you need access to module-specific ADR and code convention YAML files, which the skill merges to produce a unified, ready-to-use review criteria document for your pull request.

What's the best way to align code quality across development teams before a pull request?

The best way to align code quality is to auto-generate review criteria by merging extracted ADR requirements with matching code convention rules, ensuring consistent enforcement of architectural and coding standards.

Why use a two-stage isolated subagent workflow for code review preparation?

A two-stage isolated subagent workflow is used during code review preparation to avoid context pollution when extracting ADRs and merging code convention rules into a unified criteria document.

Can I use this for large cross-module pull requests with many unrelated ADRs?

Yes, for large cross-module pull requests, this skill filters out unrelated ADRs by matching the current task's tech stack, eliminating the need to manually sift through dozens of documents.