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.
npx skills add tmchow/agent-skills --skill clawpatchgit clone --depth 1 https://github.com/tmchow/agent-skillsWrote 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/tmchow/agent-skills/clawpatch)<a href="https://agentmods.dev/skills/tmchow/agent-skills/clawpatch"><img src="https://agentmods.dev/badge/skills/tmchow/agent-skills/clawpatch/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/tmchow/agent-skills/clawpatch"><img src="https://agentmods.dev/badge/skills/tmchow/agent-skills/clawpatch.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00158 | $0.03515 |
| Opus 5.5 | $0.00063 | $0.01406 |
| Sonnet 5.5 | $0.00032 | $0.00703 |
| Haiku 4.5 | $0.00016 | $0.00351 |
Grade C, and why
clawpatch scanned grade C 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
(`[ -d .clawpatch ] && rm -r .clawpatch` — guarded, never `rm -rf`) is a last How it starts
The opening of the file, as written. The whole thing — 258 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Clawpatch — agent workflow
Clawpatch (
openclaw/clawpatch, https://clawpatch.ai) is an agent-driven CLI: it reviews a repo into structured findings and can apply validated per-finding fixes. It needs Node.js 22+ and a provider CLI (codex by default);fixrefuses dirty source worktrees by default, whilereview/ci/revalidatecan include dirty code when explicitly scoped.This skill is judgment + orchestration, not a CLI reference. For exact flags, JSON fields, and exit codes, trust
clawpatch <cmd> --helpand the live--jsonoutput over anything written here — Clawpatch is pre-1.0 and its surface drifts, so a written copy will mislead. What this skill carries is what--helpcan't tell you: when to reach for it, how to fix findings in parallel, and the traps.
When to use
Use Clawpatch when the user names it, or wants whole-repo automated review that yields locatable findings plus optional validated fixes.
Don't use it for: a small diff already under review (a whole-repo review cycle
is overkill — use a diff-review tool); a bug the user already understands and
just wants written; or when no provider CLI (codex/claude/cursor/grok/
opencode/pi/acpx, or mock for tests) is installed — clawpatch doctor
fails fast, so say so and stop.
Setup
clawpatch doctor verifies the install and the provider. The published CLI
requires Node.js 22+. If clawpatch is missing, install the current published
package (npm install -g [email protected]). The provider (codex by default) is
the user's to install and authenticate — don't run login flows on their behalf.
Gitignore .clawpatch/ early to keep status, commits, and PR diffs clean.
Clawpatch allows its own state directory to change during runs; fix's dirty
guard is for source worktree changes, not .clawpatch/ churn. Add the single
line .clawpatch/ before review/fix when the repo doesn't already ignore it,
but remember .gitignore is tracked — confirm with the user before editing it.
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.
- 1mo ago First seen · 258 lines · 158 tokens per session scan C a61c8b3bc0b5
clawpatch is a skill published in the GitHub repository tmchow/agent-skills (65 stars, last pushed 2mo ago), licensed MIT. It adds 158 tokens to every session and 3,515 once invoked, about $0.0006 per session on Opus 5.5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
autoreview
Structured code review through the shared OpenClaw agent-skills installation.
omh-code-review
This is an OMH code-review workflow skill, projected for Agent Skills hosts (Claude Code, Codex, Cursor, opencode, OpenClaw, pi).
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
re0-merge
Review and land an external contribution the way this suite does: gate it against the thesis, land it with the author's credit intact, complete a new skill rather than merging it raw, then approve, credit, and explain before closing. Use when reviewing a pull request, as any collaborator or maintainer, not only the…
explore-unknowns
Guide the user through a quadrant walk that maps the unknowns of a task — open by listing the known knowns, then work through known unknowns, unknown knowns, and unknown unknowns one stage at a time, ending with a complete four-quadrant map in the user's hands. Use when a request is ambiguous or underspecified, the…
asplos-writing-style
Use when drafting or revising ASPLOS prose — engineering the first two pages that rapid review actually reads, stating the cross-layer insight as one quotable sentence, structuring per-layer mechanism sections, writing bounded claims, and fitting everything into 11 self-contained pages that include figures and…