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-address-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-address-review)<a href="https://agentmods.dev/skills/hardwood-hq/hardwood/hardwood-address-review"><img src="https://agentmods.dev/badge/skills/hardwood-hq/hardwood/hardwood-address-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-address-review"><img src="https://agentmods.dev/badge/skills/hardwood-hq/hardwood/hardwood-address-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 3 findings, up to medium
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 →
- medium Excessive Agency · line 22 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 48 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- low Excessive Agency · line 89 Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.Fix: Limit the skill's scope to its documented purpose. Remove instructions that enable the agent to perform actions outside its stated functionality.
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 | $0.00101 | $0.02271 |
| Opus 5.5 | $0.00040 | $0.00908 |
| Sonnet 5.5 | $0.00020 | $0.00454 |
| Haiku 4.5 | $0.00010 | $0.00227 |
Grade B, and why
hardwood-address-review scanned grade B 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 1mo 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
git fetch https://github.com/hardwood-hq/hardwood.git pull/<N>/head git worktree add -b <branch> .claude/worktrees/<branch> FETCH_HEAD How it starts
The opening of the file, as written. The whole thing — 137 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hardwood address-review
Companion to the hardwood-review skill. The review skill produces _reviews/pr-<N>-review.md and then invokes this skill, which consumes the file — implements the fixes, ticks the checkboxes, and commits them as one commit separate from the reviewed work.
Workflow
1. Identify the target
When invoked by hardwood-review, the target is handed over: a PR number, a branch name, or working-tree. The review file is _reviews/pr-<N>-review.md for a PR and _reviews/branch-<name>-review.md otherwise.
When the user invokes this skill directly, parse the request for a PR number (e.g. "449", #449, a PR URL). If they didn't supply one, look for the most recently modified file matching _reviews/*-review.md and confirm with the user.
2. Sanity checks
- Review file exists. It lives in the main checkout's
_reviews/(gitignored), not in the worktree created below; keep editing it there. If missing, runhardwood-reviewfirst. - Find the issue number. For a PR:
gh pr view <N> --json title -q .title— the title should start with#<issue> …. For a branch: the branch name's leading number, or the subject prefix of its commits (git log main..<branch> --format=%s). If neither yields one, ask the user; don't invent one.
3. Get a worktree on the target
Per CLAUDE.md, commits go into a worktree of their own under .claude/worktrees/, never into /workspace, which stays on main and may carry other sessions' uncommitted files. The state of /workspace's working tree is therefore no reason to stop.
First run git worktree list. If a worktree already has the branch checked out, work there, provided its git status --porcelain lists nothing outside .claude/ and _reviews/; if it does, stop and tell the user rather than mixing your fixes into someone's in-progress work.
Otherwise create one. For a PR:
gh pr view <N> --json headRefName,isCrossRepository,headRepositoryOwner,headRepository
git fetch https://github.com/hardwood-hq/hardwood.git pull/<N>/head
git worktree add -b <branch> .claude/worktrees/<branch> FETCH_HEAD
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.
- 1mo ago Changed · -5 lines · +26 tokens per session scan A → B 9a4b47f6dfdc
- 1mo ago Changed 2fed42c5ad2f
- 2mo ago First seen · 142 lines · 75 tokens per session scan A e38c2fc97241
hardwood-address-review is a skill published in the GitHub repository hardwood-hq/hardwood (417 stars, last pushed 4d ago), licensed Apache-2.0. It adds 101 tokens to every session and 2,271 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
pull-request-authoring
Workflow for creating WP Rig pull requests that match the repository template, including release-specific guidance for develop-to-master merges.
pr
Draft a PR body from OMC's paper trail — the smallest visual that proves the change, evidence consumed from the verify protocol (never re-collected), reversibility from the ADR test on the diff, glossary language throughout.
merge-batch
Prepare and merge an owner-selected PR batch with contributor credit and exact-candidate admission. Use for a pasted PR list; combined focused validation is diagnostic, and every actual landing still requires its applicable checks and authorization.
otto-pr-review-loop
Run Polly and OCR on a PR until all findings settle.
engram-branch-pr
PR creation workflow for Engram following the issue-first enforcement system. Trigger: When creating a pull request, opening a PR, or preparing changes for review.
github-contributor
A step-by-step guide for contributing code, tests, documentation, or build changes to open-source projects you do not maintain. It explains how to prepare a pull request that follows the project’s rules and is easier for maintainers to review.