requesting-code-review_skill

This skill helps you conduct rigorous code reviews by dispatching the code-reviewer subagent after tasks to ensure alignment with plans and quality.
  • Rust

13

GitHub Stars

2

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill mosif16/codex-skills --skill requesting-code-review

  • code-reviewer.md3.3 KB
  • SKILL.md2.8 KB

Overview

This skill dispatches a code-reviewer subagent to validate recent changes against the plan or requirements before you proceed. It’s designed to catch issues early, guide remediation, and ensure readiness to merge. Use it as a routine gate after tasks, major features, or before merges to main.

How this skill works

You provide commit SHAs (BASE_SHA and HEAD_SHA), a concise description of what was implemented, and the plan or requirements the change should meet. The skill launches the superpowers:code-reviewer subagent, which inspects the diff, tests, and architecture against the supplied plan and returns strengths, categorized issues (Critical/Important/Minor), and an overall assessment. Use the feedback to prioritize fixes: address Critical immediately, resolve Important before proceeding, and track Minor for later.

When to use it

  • After each completed task in subagent-driven development
  • After finishing a major feature or batch of tasks
  • Before merging to main or an integration branch
  • When you’re stuck and need a fresh perspective
  • Before a large refactor to establish a baseline

Best practices

  • Always include BASE_SHA and HEAD_SHA so the reviewer inspects the exact diff
  • Supply a clear WHAT_WAS_IMPLEMENTED and PLAN_OR_REQUIREMENTS to focus the review
  • Act on Critical issues immediately and Important issues before moving on
  • Document or push back with tests and reasoning if you disagree with feedback
  • Review early and often to prevent cascading problems

Example use cases

  • Just finished Task 2: dispatch review to verify new verification and repair functions
  • Before merging a feature branch that implements a new API endpoint
  • After completing a three-task batch from an execution plan to confirm readiness
  • When a complex bug fix touches core invariants and needs an extra check
  • Prior to a large refactor to ensure regressions are caught early

FAQ

Provide BASE_SHA for the starting commit (or origin/main) and HEAD_SHA for the current tip so the subagent reviews the exact changes.

How should I handle reviewer disagreements?

Push back with technical reasoning, reproducible tests, or additional context. If reviewer is mistaken, show evidence; if not, accept and fix.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational