Skip to main content
skill CX Codex

peer-review-initiator — Claude Code skill

peer-review-initiator builds a review request for your agent: commit SHAs, what changed, what it should do. Then hands it to a reviewer subagent.

LicenseRef-Proprietary 133 installs ★ 50 updated 2026-01-05

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add qodex-ai/ai-agent-skills --skill peer-review-initiator -a claude-code
Open on GitHub

What peer-review-initiator does

The skill fixes the order of a review handoff. The agent resolves a base and a head commit with git rev-parse, then fills the code-reviewer.md template: what was built, the requirements behind it, the diff range, a short summary. That request goes to the superpowers:code-reviewer subagent.

Triggers are spelled out. Every task in subagent-driven work, a finished feature, a merge to main. Two optional cases too: you are stuck, or you are about to refactor. Feedback gets sorted — critical fixed right away, important before the next step, minor noted for later. If the reviewer is wrong, the skill says push back with technical reasoning.

It expects the superpowers:code-reviewer subagent and the template at requesting-code-review/code-reviewer.md to be in place. Without them there is nothing to dispatch, and placeholders like {BASE_SHA} or {HEAD_SHA} have nowhere to go. The license is proprietary, terms in LICENSE.txt.

What it helps with

  • assemble a review request with SHAs and a change summary
  • hand the diff to a reviewer subagent instead of retelling it
  • sort reviewer feedback into fix-now and fix-later
  • keep review in the loop after every task in subagent work
  • push back on a wrong review with reasoning and tests

Example prompts for the agent

  • Request review for this branch: base origin/main, head is HEAD.
  • Run a review before I merge to main and triage the findings.
  • I'm stuck on this bug — get a fresh review pass on the current diff.

How to install peer-review-initiator

1Skills CLI (Vercel)

npx skills add qodex-ai/ai-agent-skills --skill peer-review-initiator -a claude-code

2Manually: put the files in place

~/.claude/skills/peer-review-initiator/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/peer-review-initiator/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about peer-review-initiator

How to install peer-review-initiator in Claude Code?

Run npx skills add qodex-ai/ai-agent-skills --skill peer-review-initiator -a claude-code. The skill lands in ~/.claude/skills/peer-review-initiator/SKILL.md.

What does peer-review-initiator do?

It assembles a code review request with commit SHAs and a description of the changes and requirements, sends it to the superpowers:code-reviewer subagent, then helps you work through the answer.

What does it need to run?

The superpowers:code-reviewer subagent and the requesting-code-review/code-reviewer.md template, whose placeholders the skill fills in.

Similar skills

All skills for Code review and refactoring →

From the same set42

The whole set →
animated-message-composer animated-message-composer builds animated GIFs that fit Slack's limits: frames drawn with Python and Pillow, then checked for size and palette. CX Codex +1 135installs distributed-task-execution The distributed-task-execution skill splits unrelated problems across agents, runs investigations in parallel, and gathers their results. It fits debugging work with separate failures. CX Codex +1 134installs cloud-expense-management cloud-expense-management tracks AWS expenses, calculates resource costs before deployment, and inspects CloudWatch telemetry. Useful for budget tracking and detecting billing spikes. CX Codex +1 134installs capability-activation The capability-activation skill makes an agent check for applicable skills before responding and sets an order for using them. It fits workflows that depend on explicit process rules. CX Codex +1 134installs selection-randomizer selection-randomizer is a Claude Code skill that randomly picks winners and participants from lists. It can apply weights and exclude past winners. CX Codex +1 133installs branch-finalization branch-finalization verifies tests, presents four clear next steps, and handles merging, pull requests, or branch cleanup. CX Codex +1 131installs autonomous-cloud-orchestration autonomous-cloud-orchestration walks your agent through AWS Bedrock AgentCore: picking the right service and wiring the deployment with the AWS CLI. CX Codex +1 131installs capability-documentation The capability-documentation skill helps document other skills with clear structure, examples, and behavior checks. It fits reusable agent guidance that needs verification. CX Codex +1 130installs

Project sponsors

Companies that keep this analytics open