What happens if the reviewer finds a 'Critical' issue?

Critical issues must be fixed immediately. The workflow dictates that you should not proceed to the next task until Critical and Important issues are resolved.

Can I disagree with the AI subagent's feedback?

Yes. If the reviewer is incorrect, you should push back with technical reasoning, show code or tests that prove your implementation works, or request further clarification.

When is it mandatory to use the Requesting Code Review skill?

It is mandatory after completing each task in subagent-driven development, after implementing a major feature, and immediately before merging code into the main branch.

How does the AI reviewer know what implementation to check?

The skill utilizes Git SHAs (BASE and HEAD) to identify the specific code changes and compares them against the provided plan or requirements document.

Requesting Code Review

Name: Requesting Code Review
Author: Kingly-Agency

byKingly-Agency

0•

协作工具

Dispatches a specialized subagent to audit code implementations against requirements and plans before merging or proceeding.

The Requesting Code Review skill integrates an automated, subagent-driven review workflow into the Claude Code environment. By dispatching a dedicated code-reviewer subagent, it allows developers to catch logic errors, architectural deviations, and requirement gaps early in the development cycle. The skill uses Git SHAs to isolate changes and compares them against specific project documentation or plans, providing a structured feedback loop that categorizes issues by severity to ensure high-quality, production-ready code before it ever reaches a human reviewer or the main branch.

主要功能

01Requirement-based validation against design docs and plans

02Seamless integration with subagent-driven development workflows

03Automated subagent dispatching for targeted code audits

04Differential analysis using Git SHAs to isolate new changes

05Severity-based issue categorization (Critical, Important, Minor)

060 GitHub stars

使用场景

01Verifying individual task completion within a complex development plan

02Conducting a final quality assurance check before merging feature branches

03Obtaining an objective technical perspective when stuck on complex logic

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add kingly-agency/kingly-claude-adapter requesting-code-review

For use in Claude.ai and ChatGPT

Download Skill