git-pr-codex-loop

A workflow for opening a GitHub pull request and repeatedly sending it to Codex Code Review, an automated review bot. It handles review findings and asks for another review until the pass is clean, but leaves merging to a person.

In plain words
What is it for?
Taking a ready feature branch through pull-request creation, automated review, fixing reported issues, replying to findings, and requesting another review.
Why use it?
It keeps review fixes and follow-up requests in one process, reducing the chance that findings are missed before a change is merged.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/charlesjones-dev/claude-code-plugins-dev/git-pr-codex-loop
Any agent
npx skills add charlesjones-dev/claude-code-plugins-dev --skill git-pr-codex-loop
Clone the repo
git clone --depth 1 https://github.com/charlesjones-dev/claude-code-plugins-dev

Made for: Claude Code, Codex.

Per session 75 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,815 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00075 $0.01815
Opus 5 $0.00037 $0.00907
Sonnet 5 $0.00015 $0.00363
Haiku 4.5 $0.00007 $0.00181

Measured 2d ago against content hash 0e2271fc1244, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

git-pr-codex-loop scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/ai-git/skills/git-pr-codex-loop/SKILL.md · 161 lines

How it starts

The opening of the file, as written. The whole thing — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/git-pr-codex-loop - Codex Code Review loop

Positioning: Claude Code's native /code-review command (with --fix, --comment, and the cloud "ultra" mode) now covers automated review-and-fix loops natively, with no external dependencies. This skill is specifically for repositories standardized on OpenAI's Codex Code Review bot, and remains the right tool for that cross-tool workflow.

You drive the current branch's pull request through Codex Code Review until the review is clean. The human makes the final merge decision; you never merge.

Prerequisites

  • The GitHub CLI (gh) is installed and authenticated (gh auth login).
  • Codex Code Review is enabled for the repository in Codex settings. With Automatic reviews on, Codex reviews every new PR; otherwise you trigger a review manually with @codex review. Reviews post as the Codex bot (chatgpt-codex-connector).
  • Codex applies a ## Review guidelines section from your repo's AGENTS.md when present, so project-specific review rules are honored automatically.

Process

Step 1: Open or find the PR

  1. Get the current branch: git branch --show-current. If it is main or master, stop and tell the user to switch to a feature branch.
  2. Look for an existing PR: gh pr view --json number,url,state 2>/dev/null. If none exists, push the branch (git push -u origin HEAD) and open one with gh pr create --fill (write a real title/description if the change warrants it). Match the PR attribution style used by the git-commit-push-pr skill.
  3. Capture the PR number and URL for the rest of the run.

Step 2: Monitor CI

Watch checks with gh pr checks --watch. If any check fails, fix the cause, commit, push, and watch again. Do not continue while checks are red — Codex reviewing a non-compiling PR is wasted.

Step 3: Monitor for the Codex review

When a review is requested — automatically on PR open, or by your @codex review reply — Codex first acknowledges with a 👀 (eyes) reaction on the triggering post, then works. The 👀 means "received the request," not "done" — do not act on it. Poll for the real outcome with gh pr view --comments, the review-thread query below, and the reactions check below.

Read the full file on GitHub · 161 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 161 lines · 75 tokens per session scan A 0e2271fc1244

Subscribe to this mod's changes

git-pr-codex-loop is a skill published in the GitHub repository charlesjones-dev/claude-code-plugins-dev (34 stars, last pushed 1mo ago), licensed MIT. It adds 75 tokens to every session and 1,815 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

gwm

Manage git worktrees across any repository with the gwm Rust binary (CLI + ratatui TUI). Use when the user asks to create / list / remove / bootstrap / switch / link worktrees, run a command across worktrees (gwm exec) or reclaim build artifacts (gwm clean), materialise a PR into a worktree (gwm review), drive a…

kbrdn1/gwm-cli · 379 tokens

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens