Jcputney Agent Peer Review
jcputney-agent-peer-review is a Claude Code plugin for the Ship phase that validates Claude-generated work using OpenAI Codex CLI so two AI perspectives review the same changes.
by jcputney · github.com/jcputney/agent-peer-review
Cross-check Claude Code’s patches by running a second AI reviewer via OpenAI Codex CLI before you trust a merge.
Add it to Claude Code
Install the plugin in Claude Code. One command, paste-ready.
/plugin install jcputney-agent-peer-review@jcputney/agent-peer-reviewBuilt to be called by your agent
Skillselion is itself an MCP server. Your agent can pull this entry and a paste-ready install config straight from the API - no copy-paste.
Retrieve this entry with skillselion.get_details("plugin:jcputney/agent-peer-review") and the paste-ready config with skillselion.get_install_config("plugin:jcputney/agent-peer-review").
What it does
jcputney-agent-peer-review is a Claude Code plugin for builders who already ship with Claude but want a disciplined second pair of eyes—implemented as OpenAI Codex CLI review of Claude’s work. The value proposition is simple: one model generating patches and another model critiquing them catches classes of mistakes that self-review glosses over, especially when you are moving fast as a solo founder. You install the single plugin, ensure Codex CLI is available in your environment, and invoke peer review when a task feels high stakes—auth changes, billing logic, or refactors you cannot manually re-read line by line. It is not a human-in-the-loop product like pair-review; it is automated cross-agent validation with setup overhead traded for coverage. Best used as a gate before merge on agent-heavy branches, not as your only security or compliance audit.
Highlights
- Single-plugin bundle (pluginCount: 1) focused on peer validation, not primary codegen.
- Sends Claude’s output through OpenAI Codex CLI for a second opinion on bugs and design issues.
- Explicit cross-vendor review: two AI perspectives on the same change set.
- Aims to catch issues a single coding agent habitually misses on self-review.
- Lightweight validation layer you add on top of normal Claude Code sessions.
Why builders use it
When Claude codes alone, solo builders still merge subtle bugs because the same agent that wrote the diff rarely challenges its own assumptions.
After registration, Codex CLI peer review flags additional issues on Claude’s output so you fix them before merge with less single-model blind spot risk.
At a glance
- Type - Plugin in Development Tools.
- Adoption - 0 installs, 27 stars, 0 votes.
FAQ
Who is jcputney-agent-peer-review for?
It is for solo builders on Claude Code who install OpenAI Codex CLI and want a systematic second AI to review Claude’s implementations.
When should I use jcputney-agent-peer-review?
Use it on Ship-phase review for risky diffs, large agent-generated refactors, or any merge where you want Codex to challenge Claude’s work before you ship.
How do I add jcputney-agent-peer-review to my agent?
Install the plugin from jcputney/agent-peer-review in Claude Code, configure and authenticate OpenAI Codex CLI on your machine, then trigger peer review from the plugin workflow after Claude completes a task.
Comments
Share how you use jcputney-agent-peer-review, gotchas, or tips for other indie builders.
No comments yet - be the first to share how you use it.