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 tony/skills --skill extractgit clone --depth 1 https://github.com/tony/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/tony/skills/extract)<a href="https://agentmods.dev/skills/tony/skills/extract"><img src="https://agentmods.dev/badge/skills/tony/skills/extract/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/tony/skills/extract"><img src="https://agentmods.dev/badge/skills/tony/skills/extract.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 | $0.00023 | $0.02012 |
| Opus 5.5 | $0.00009 | $0.00805 |
| Sonnet 5.5 | $0.00005 | $0.00402 |
| Haiku 4.5 | $0.00002 | $0.00201 |
Grade A, and why
extract 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 5d 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.
How it starts
The opening of the file, as written. The whole thing — 193 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/scholar:extract
Stage 2. Harvest the terms the corpus uses for its own things, each with a citation that shows the term in use.
Read ../../references/citation.md for what a citation must carry, including
why a quotation containing its own term is circular, and
../../references/source-pins.md for repository pins. Read
../../references/stage-gates.md for this stage's exit condition. The
analyzer is ../../scripts/term_metrics.py; <analyzer> below is its absolute
path resolved from this skill's directory.
User arguments: $ARGUMENTS
Read ../../references/output-contract.md before writing study files.
What counts as a term
A term is a word the corpus uses for one of its own concepts. In code that is
the names it gives things: type and class names, module and package names,
recurring verbs in function names and commit subjects, the roles in its
configuration, the nouns in its error messages. In prose it is the author's
own vocabulary, taken from quotations.jsonl, the quotation set annotate
writes in the study directory.
A term is not a word that merely appears. handler is a term where the
project distinguishes handlers from services; it is noise where it is just
what someone called a function once.
Procedure
1. Read the source list
Take sources.jsonl from gather. Harvest only from sources marked
"read": true.
2. Put the corpus on disk, read-only
Read each code source at its recorded commit, never at a moving branch.
<pins> is ../../scripts/scholar_pins.py, resolved against this skill's directory to
an absolute path.
A checkout already on disk is read in place when this exits zero; <study> is
the study directory, which the dirty-tree check leaves out when it sits inside
the checkout:
$ python3 <pins> local <corpus> \
--study <study> \
--expect <commit>
A refusal (uncommitted edits, or HEAD at another commit) is reported, not harvested.
A remote source gets a throwaway checkout outside the study and outside the
current repository, fetched with plain git from the source URL, so any host
that serves git over HTTPS works. Take URL and COMMIT from sources.jsonl.
The command refuses a URL or commit it does not accept (mark that source
unharvested) and prints the checkout path on success; use that literal path
afterwards:
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.
- 5d ago Changed · -6 lines e51cb005b3b1
- 8d ago Changed · +101 lines d3e67ec10e1b
- 1mo ago First seen · 98 lines · 23 tokens per session scan A 1b02f6dbffbe
extract is a skill published in the GitHub repository tony/skills (0 stars, last pushed 3d ago), licensed MIT. It adds 23 tokens to every session and 2,012 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-09-03.
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"…
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…
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…
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…