requesting-code-review

Request evidence-first code reviews before merging high-risk work.

968|43|Updated Apr 30, 2026

Search Tags:#architecture #verification #evidence #subagent #baseline #code-review #merge-prep

Ready to run

npx skills add https://github.com/GanyuanRan/Aegis --skill requesting-code-review-ganyuanran

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: requesting-code-review
Source: https://github.com/GanyuanRan/Aegis/tree/main/skills/requesting-code-review
Command: npx skills add https://github.com/GanyuanRan/Aegis --skill requesting-code-review-ganyuanran

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a structured, evidence-first process to request code reviews, ensuring reviews focus on the work product, preserve the reviewer’s context, and prevent cascading issues.

Core Features & Use Cases

Canonical review-workflow for method-pack implementation work.
Dispatches an expert reviewer with precise, relevant context and required evidence.
Ensures review is advisory and evidence-oriented, not a completion gate.

Quick Start

Gather the implementation details, plan, evidence, and baseline references, then dispatch aegis:code-reviewer with a complete review template.

Frequently Asked Questions about requesting-code-review

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

FAQPage Schema

How do I request an evidence-driven code review before merging high-risk work?▼

To request an evidence-driven code review, you dispatch a formal reviewer with a structured template covering scope, success criteria, architecture boundaries, and fresh evidence. This verifies implementation quality before merging high-risk work to main.

What is an evidence-first code review process?▼

An evidence-first code review is an advisory process that focuses on the work product by requiring structured outputs, baseline alignment, and concrete evidence rather than acting as a simple completion gate.

When do I need to run a formal code review for subagent-driven tasks?▼

You need to run a formal code review after each subagent-driven task or major feature completion, and specifically before merging to main, to surface findings and verify product baseline alignment.

What should be included in a code review checklist for architecture verification?▼

A code review checklist for architecture verification should include scope, success criteria, product baseline, architecture boundary, fresh evidence, compatibility boundaries, retirement notes, reviewer responsibilities, and advisory status.

Does this code review workflow act as a completion gate for merging?▼

No, this code review workflow does not act as a completion gate. It is designed to be advisory and evidence-oriented, preserving the reviewer's context while focusing strictly on the work product.

Can I use this structured review process without subagent task implementation?▼

Yes, while it applies after subagent-driven tasks, the structured review process can be used to verify any major feature completion by gathering implementation details, plans, and baseline references to dispatch a reviewer.