code-review

A detailed review workflow for finding problems in code changes. It examines local Git changes or a GitHub pull request, which is a proposed set of changes for review.

In plain words
What is it for?
Reviewing local diffs or GitHub pull requests and, for pull requests, posting findings as line-level review comments.
Why use it?
It looks for bugs, security issues, concurrency problems, edge cases, missing tests, and assumptions that may cause trouble after release.

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/suyoumo/clawprobench/code-review
Any agent
npx skills add suyoumo/ClawProBench --skill code-review
Clone the repo
git clone --depth 1 https://github.com/suyoumo/ClawProBench

Made for: Claude Code, Codex.

Per session 57 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,817 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. 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 $0.00057 $0.02817
Opus 5 $0.00028 $0.01409
Sonnet 5 $0.00011 $0.00563
Haiku 4.5 $0.00006 $0.00282

Measured 2d ago against content hash 7741d37ba9c8, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

code-review scanned grade A with 1 finding 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

url=f"https://api.github.com/repos/{owner}/{repo}/contents/{urllib.parse.quote(path, safe='')}?ref={head_sha}",
ironclaw/skills/code-review/SKILL.md · 231 lines

How it starts

The opening of the file, as written. The whole thing — 231 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Paranoid Architect Code Review

You are reviewing this change as a paranoid architect. Your job is to find every bug, vulnerability, race condition, edge case, and undocumented assumption before it ships. Assume adversarial users, concurrent access, and Murphy's law.

You handle two input shapes:

  • Local changes — uncommitted edits or recent commits in the working tree.
  • GitHub pull requestowner/repo N, owner/repo#N, or a github.com/.../pull/N URL. If the message contains anything shaped like owner/repo followed by a number, treat it as a PR request and use the GitHub path, not git. Exception: if the message also contains locally or local, use the local path instead.

Step 1 — Load the changes

GitHub PR path

Wrap the whole flow in async def and return from it, then FINAL(await review()). FINAL() only records the answer, it does not stop execution; without a return, code after FINAL(...) keeps running and crashes when it tries to use a variable that was never set on an error path.

async def review():
    pr_url    = f"https://api.github.com/repos/{owner}/{repo}/pulls/{number}"
    files_url = f"{pr_url}/files?per_page=100"
    # Sequential awaits instead of asyncio.gather — the Monty sandbox
    # does NOT reliably capture `import asyncio` into the function
    # closure, and the LLM has hit `NameError: name 'asyncio' is not
    # defined` when calls cross repl-block boundaries. Three serial
    # GETs against api.github.com are fast enough, and this avoids the
    # whole class of closure-capture bugs.
    meta_r = await http(method="GET", url=pr_url)
    diff_r = await http(
        method="GET", url=pr_url,
        headers=[{"name": "Accept", "value": "application/vnd.github.v3.diff"}],
    )
    files_r = await http(method="GET", url=files_url)
    for r, label in [(meta_r, "metadata"), (diff_r, "diff"), (files_r, "files")]:
        if r["status"] != 200:
            return (f"GitHub {label} fetch for {owner}/{repo}#{number} "
                    f"returned HTTP {r['status']}: {r['body']}")
    pr    = meta_r["body"]       # dict: title, state, head, base, user, head.sha, ...
    diff  = diff_r["body"]       # str: unified diff
    files = files_r["body"]      # list: per-file summaries with patch hunks
    head_sha = pr["head"]["sha"] # needed if you post line-level comments later
    # ... build the review ...
    return body

FINAL(await review())

Read the full file on GitHub · 231 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 · 231 lines · 57 tokens per session scan A 7741d37ba9c8

Subscribe to this mod's changes

code-review is a skill published in the GitHub repository suyoumo/ClawProBench (823 stars, last pushed 7d ago), licensed Apache-2.0. It adds 57 tokens to every session and 2,817 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). 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

lastlight-evals

Scaffold, configure and run a Last Light EVALS workspace — the harness that runs Last Light's real workflows against a mocked GitHub and grades them deterministically. Use when the user wants to "set up / scaffold Last Light Evals", "create an evals workspace or instance", "run evals", "compare models", or author new…

nearform/lastlight · 185 tokens

Evaluation

Frames model, prompt, and system evaluation as a reproducible experiment with baselines, datasets, and explicit metrics.

agentic-in/elephant-agent · 25 tokens

open-code-review

Performs AI-powered code review on Git changes using the ocr CLI from alibaba/open-code-review. Use when the user asks to review code, review a pull request, review staged/unstaged changes, review a commit, or compare branches for code quality issues. Produces line-level review comments and can automatically apply…

alibaba/open-code-review · 98 tokens

harness-creator

Build, audit, and improve harnesses that make AI coding agents reliable: AGENTS.md/CLAUDE.md instruction files, feature/state tracking, verification gates, scope boundaries, session handoff, memory persistence, context budgets, tool-permission safety, and multi-agent coordination. Use this whenever a coding agent is…

walkinglabs/learn-harness-engineering · 142 tokens

gh-pr-description

Drafts and reviews GitHub pull request descriptions for the eve repository. Use when opening, updating, or reviewing a PR, or when summarizing a branch for reviewers.

vercel/eve · 38 tokens

technical-writing

Write, edit, review, or audit user-facing documentation for the eve repository. Use for changes under docs/, documentation tied to eve APIs or CLI behavior, docs work based on Slack or support feedback, and requests to make eve docs clearer, more natural, or less AI-patterned while verifying claims against current…

vercel/eve · 78 tokens