agent-review

agent-review is a skill for Codex from ithinkihaveacat/dotfiles. It costs 91 tokens per session (1,139 once invoked), scanned A, original, Apache-2.0.

A read-only workflow for reviewing code changes in branches, commits, working trees, or pull requests, sometimes using a separate command-line AI reviewer.

In plain words
What is it for?
Checking diffs and pull requests, obtaining an independent review, and reporting actionable issues the author would likely fix.
Why use it?
It focuses review on defects supported by evidence, reducing the chance of overlooking problems or filling the review with unrelated suggestions.

Skill for Codex

Written for Codex: agents/openai.yaml present. Also seen: mentions Codex; $skill-name invocation.

Good fit Checking diffs and pull requests, obtaining an independent review, and reporting actionable issues the author would likely fix.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/ithinkihaveacat/dotfiles/agent-review
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.

Any agent
npx skills add ithinkihaveacat/dotfiles --skill agent-review
Clone the repo
git clone --depth 1 https://github.com/ithinkihaveacat/dotfiles

Made for: Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for agent-review

README.md
[![agentmods](https://agentmods.dev/badge/skills/ithinkihaveacat/dotfiles/agent-review.svg)](https://agentmods.dev/skills/ithinkihaveacat/dotfiles/agent-review)
Your own site
<a href="https://agentmods.dev/skills/ithinkihaveacat/dotfiles/agent-review"><img src="https://agentmods.dev/badge/skills/ithinkihaveacat/dotfiles/agent-review.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for agent-review

Your own site · 80×15
<a href="https://agentmods.dev/skills/ithinkihaveacat/dotfiles/agent-review"><img src="https://agentmods.dev/badge/skills/ithinkihaveacat/dotfiles/agent-review.svg?style=web" alt="Reviewed on agentmods" width="80" height="15"></a>
Per session 91 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,139 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 2 findings, up to high

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • high Excessive Agency · line 56
    Skill selects an external model or provider that may use a different account or billing plan than the operator expects. Undisclosed model switches can cause unexpected cost or quota consumption.
    Fix: Remove the model/provider override or disclose it prominently and require explicit operator approval before invoking an external coding CLI or billed model.
  • high Excessive Agency · line 57
    Skill selects an external model or provider that may use a different account or billing plan than the operator expects. Undisclosed model switches can cause unexpected cost or quota consumption.
    Fix: Remove the model/provider override or disclose it prominently and require explicit operator approval before invoking an external coding CLI or billed model.
How audits are shown
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 5.1 $0.00091 $0.01139
Opus 5 $0.00046 $0.00570
Sonnet 5 $0.00018 $0.00228
Haiku 4.5 $0.00009 $0.00114

Measured 9d ago against content hash 1e2592d95c80, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

agent-review 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 9d 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.

skills/agent-review/SKILL.md · 117 lines

How it starts

The opening of the file, as written. The whole thing — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Code Review

Review changes independently and return only defects that the author would likely fix. Treat findings as claims that need evidence, not suggestions to improve code in general.

Choose the execution role

Determine which role this invocation has before doing anything else:

  • Reviewer session: The current agent was asked to review the repository, or the prompt says this is a synthesized reviewer session. Perform the workflow below yourself. Do not launch another agent.
  • Delegating session: The user asked to obtain an independent review from a named CLI agent. Run scripts/review-with-agent from this skill. The launcher uses native review commands where possible and explicitly invokes this skill for synthesized fallbacks.

Native review commands are process entry points, not tools a reviewer session must recursively call. Do not launch a second copy of the current agent merely because its CLI has native review support.

Delegate non-interactively

Run the launcher from the repository being reviewed:

scripts/review-with-agent --agent codex --base main
scripts/review-with-agent --agent claude --base main
scripts/review-with-agent --agent agy --commit HEAD
scripts/review-with-agent --agent agy --uncommitted

Paths beginning with scripts/ are relative to this skill directory. Use --dry-run to inspect the selected command without starting an agent. Use --synthesized to test the portable workflow even when a native route exists.

For a direct CLI invocation without the launcher, use the same distinction:

codex exec review --base main
claude -p "/code-review main...HEAD" --permission-mode dontAsk --no-session-persistence
agy --print "Use the installed agent-review skill to review the current branch against main. Perform the review yourself; do not delegate." --mode plan --sandbox

The launcher currently selects:

Agent Direct route Synthesized fallback
Codex Native base-to-HEAD, commit, and uncommitted review Explicit $agent-review prompt
Claude Local /code-review for a base range or single commit Explicit /agent-review prompt
agy None Explicit natural-language prompt

Read the full file on GitHub · 117 lines

Files

What ships with it

5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 9d ago First seen · 117 lines · 91 tokens per session scan A 1e2592d95c80

Subscribe to this mod's changes

agent-review is a skill published in the GitHub repository ithinkihaveacat/dotfiles (48 stars, last pushed yesterday), licensed Apache-2.0. It adds 91 tokens to every session and 1,139 once invoked, about $0.0005 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

GitHub PR comment fetcher

Easily query for open discussions and reviews on a pull request to help resolve them.

paulirish/dotfiles · 24 tokens

audit-architecture

Run an application-wide, read-only architecture audit that finds materially useful simplifications in a codebase's data structures, state representation, control flow, algorithms, and ownership boundaries. Fans out bounded read-only agents per subsystem, verifies every finding against the repo, and ranks results P0 to…

freekmurze/dotfiles · 119 tokens

explain-changes

Explain how the code on the current branch works, as an HTML page opened in the browser, written so someone can understand the change and review it. Use when asked to explain a branch, explain these changes, walk me through this code, help me review this, or explain what an agent just did. Most useful when someone…

freekmurze/dotfiles · 86 tokens

review-pr

Review and merge GitHub pull requests for Spatie packages. Use when asked to review a PR, review a pull request, merge a PR, or when given a GitHub PR URL to review. Also triggers on 'review this PR,' 'check this pull request,' 'merge this,' or '/review-pr'. Uses gh CLI for all GitHub operations.

freekmurze/dotfiles · 75 tokens

review-code

Review all changed code against project conventions.

freekmurze/dotfiles · 10 tokens

receiving-code-review

Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation.

sodiumjoe/dotfiles · 38 tokens