Borrowing it
Nothing to install: this file belongs to hardwood-hq/hardwood. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/hardwood-hq/hardwood/main/.claude/skills/hardwood-review/SKILL.mdgit clone --depth 1 https://github.com/hardwood-hq/hardwoodWrote 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.
[](https://agentmods.dev/skills/hardwood-hq/hardwood/hardwood-review)<a href="https://agentmods.dev/skills/hardwood-hq/hardwood/hardwood-review"><img src="https://agentmods.dev/badge/skills/hardwood-hq/hardwood/hardwood-review/github.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.
<a href="https://agentmods.dev/skills/hardwood-hq/hardwood/hardwood-review"><img src="https://agentmods.dev/badge/skills/hardwood-hq/hardwood/hardwood-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00102 | $0.03150 |
| Opus 5 | $0.00051 | $0.01575 |
| Sonnet 5 | $0.00020 | $0.00630 |
| Haiku 4.5 | $0.00010 | $0.00315 |
Grade A, and why
hardwood-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 yesterday.
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.
How it starts
The opening of the file, as written. The whole thing — 196 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hardwood code review
Project-specific code review. Encodes the conventions in /workspace/CLAUDE.md and the recurring failure modes the maintainer has surfaced in past reviews, so the review covers them by construction instead of by memory.
Priority frame: the Code Review Pyramid
Review effort is weighted by how expensive the issue is to fix after merge. Width = weight: the wide base is the load-bearing tier, the narrow apex is the least important.
┌────────────────────┐
│ Code style │ ← apex. Automate it; don't burn review cycles here.
┌┴────────────────────┴┐
│ Tests │
┌┴──────────────────────┴┐
│ Documentation │ ← focus your review effort
┌┴────────────────────────┴┐
│ Implementation │ ← focus your review effort
┌┴──────────────────────────┴┐
│ API semantics │ ← base. Hardest to change later — spend the most thought here.
└────────────────────────────┘
When writing the findings file, sort by pyramid tier, not by checklist section. A small API-shape concern outranks a big style nit. Style items (no var, JavaDoc syntax, formatting) belong in a "Nits" footer — flag them so the author can fix in passing, but don't elevate them to "Blockers".
Anchor reference: https://www.morling.dev/images/code_review_pyramid.svg
Workflow
1. Identify the target
Parse the user's request for one of:
- PR number (e.g. "418") →
gh pr view <n>andgh pr diff <n> - PR URL → extract number, same as above
- Branch name →
git diff main...<branch>(or whatever the main branch is) - No argument →
git diff(working tree) plusgit diff --staged; if both empty,gh pr listand ask which one
For PR diffs, persist large output to a file and read it in chunks rather than letting it land in the conversation as one blob. The diff for a non-trivial PR can run 5k+ lines.
Don't build the branch to find out whether it compiles and passes. CI already answers that. Check it with gh pr checks <n>; a green run means no worktree, no ./mvnw, no test run. Reading the diff is the review. Build locally only when CI is red or absent and you need to know whether a failure is real, or when a specific finding turns on behaviour you cannot settle by reading — and then run the narrowest thing that settles it, not verify.
What ships with it
1 file 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.
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.
- yesterday Changed · +2 lines c7755935a024
- 13d ago First seen · 194 lines · 102 tokens per session scan A d9a23b075087
hardwood-review is a skill published in the GitHub repository hardwood-hq/hardwood (372 stars, last pushed today), licensed Apache-2.0. It adds 102 tokens to every session and 3,150 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.
Other skills, from other repositories
code-review
Performs thorough code reviews with focus on best practices, security, performance, and maintainability. Use this skill when reviewing pull requests, auditing code quality, or getting feedback on implementations.
pull-request-authoring
Workflow for creating WP Rig pull requests that match the repository template, including release-specific guidance for develop-to-master merges.
agent-code-review
Mandatory self-review protocol for AI agents to follow before concluding any coding task or pull request phase.
code-quality-standards
Guide for ensuring code quality and adherence to WordPress and WP Rig standards using linting and analysis tools.
code-auditor
Independent pre-merge review of a git diff, PR, or named files. Use when the user asks to review a PR, inspect current git changes, or hunt functional regressions, missed scenarios, wrong assumptions, concurrency bugs, and test gaps as an independent reviewer who does not defend the author's approach /…
river-review-performance
A performance-review guide for changes involving database queries, loops, batch jobs, caches, or large amounts of data.