SuperAPI AI Directory

requesting-code-review

Coding & development

View repository

Updated 10-16 09:55

About this skill

Request Code Review

Dispatch a code reviewer subagent to catch issues before they compound. The reviewer gets precisely crafted context for evaluation - never your session history. This keeps the reviewer focused on the work product, not your thought process, while preserving your own context for continued work.

Core Principle: Review early, review often.

When to Request Review

Mandatory:

  • After each task in subagent-driven development
  • After completing major features
  • Before merging to main

Optional but valuable:

  • When stuck (to get a fresh perspective)
  • Before refactoring (baseline check)
  • After fixing complex bugs

How to Request

1. Get git SHAs:

BASE_SHA=$(git rev-parse HEAD~1)  # or origin/main
HEAD_SHA=$(git rev-parse HEAD)

2. Dispatch a code reviewer subagent:

Use the general-purpose type's Task tool, fill in the template in code-reviewer.md template in

Placeholders:

  • {DESCRIPTION} - A brief summary of what you built
  • {PLAN_OR_REQUIREMENTS} - What it should do
  • {BASE_SHA} - Starting commit
  • {HEAD_SHA} - Ending commit

3. Handle feedback:

  • Fix Critical issues immediately
  • Fix Important issues before continuing
  • Record Minor issues for later handling
  • If the reviewer is wrong, push back (with reasoning)

Example

[刚完成 Task 2:添加验证函数]

你:继续之前,我先请求代码审查。

BASE_SHA=$(git log --oneline | grep "Task 1" | head -1 | awk '{print $1}')
HEAD_SHA=$(git rev-parse HEAD)

[分派代码审查者子代理]
  DESCRIPTION: 添加了 verifyIndex() 和 repairIndex(),包含 4 种问题类型
  PLAN_OR_REQUIREMENTS: docs/superpowers/plans/deployment-plan.md 中的 Task 2
  BASE_SHA: a7981ec
  HEAD_SHA: 3df7661

[子代理返回]:
  Strengths: 架构干净,测试真实
  Issues:
    Important: 缺少进度指示器
    Minor: 报告间隔使用魔法数字 (100)
  Assessment: 可以继续

你:[修复进度指示器]
[继续 Task 3]

Integration with Workflow

Subagent-Driven Development:

  • Review after every task
  • Catch issues before they accumulate
  • Fix before moving to the next task

Executing Plans:

  • Review after every task or at natural checkpoints
  • Get feedback, apply, continue

Ad-Hoc Development:

  • Review before merging
  • Review when stuck

Red Flags

Never:

  • Skip review because "it's simple"
  • Ignore Critical issues
  • Continue with unfixed Important issues
  • Argue with valid technical feedback

If the reviewer is wrong:

  • Push back with technical reasoning
  • Show code/tests that prove it works
  • Request clarification

See template: requesting-code-review/code-reviewer.md