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 asale-ai/anything-to-skill --skill repolishgit clone --depth 1 https://github.com/asale-ai/anything-to-skillWrote 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/asale-ai/anything-to-skill/repolish)<a href="https://agentmods.dev/skills/asale-ai/anything-to-skill/repolish"><img src="https://agentmods.dev/badge/skills/asale-ai/anything-to-skill/repolish/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/asale-ai/anything-to-skill/repolish"><img src="https://agentmods.dev/badge/skills/asale-ai/anything-to-skill/repolish.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 | $0.00098 | $0.03948 |
| Opus 5 | $0.00049 | $0.01974 |
| Sonnet 5 | $0.00020 | $0.00790 |
| Haiku 4.5 | $0.00010 | $0.00395 |
Grade D, and why
repolish scanned grade D with 3 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 10d 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.
Subtle steeringmediumPrompt injection
Instructions that bias recommendations or shape behaviour without the user noticing.
reason. Never drop the prefix partway through, and never tell the user to run Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
curl -fsSL https://raw.githubusercontent.com/asale-ai/repolish/main/install.sh | sh Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -fsSL https://raw.githubusercontent.com/asale-ai/repolish/main/install.sh | sh How it starts
The opening of the file, as written. The whole thing — 332 lines — stays where its author put it; the contents beside it link to each section on GitHub.
repolish
A command-line tool that scores 22 concrete signals about a repository and, for every point it deducts, names the file and line and says what to write instead.
No model is in the scoring path. The same commit always produces the same score. You may suggest wording; you must never claim a number the tool did not produce.
The rule that matters most
Do not rewrite the README. Your instinct will be to replace it with a well-structured one. That destroys the author's voice, their layout, their examples, and usually their accuracy. Instead:
- Measure.
repolish— one command. It scores the repository and prints every file it would touch, and it writes nothing. - Show the user that plan, then apply it.
repolish --apply— badge, table of contents, issue/PR templates,CONTRIBUTING.md, the CI workflow. It only ever inserts; the diff is new lines and nothing else. - Hand back what needs judgement. A missing quickstart, a vague tagline, a README claim whose command no longer exists — those need the author's knowledge, or a targeted edit you can justify from evidence.
- Measure again and report the delta —
repolish --stages check --base <ref>does the arithmetic for you and lists only the checks that moved.
Only edit prose directly when a finding names a specific line and the fix is unambiguous, and say which finding you were acting on.
Install
Check first, then install. Every command below calls repolish by name, so
start by finding out whether it is already there:
repolish --version
If that prints a version, use repolish as written throughout this document. If
it says "command not found", reach for npx — it needs nothing installed and
works wherever Node does:
npx -y @asale/repolish
Then read every repolish … below as npx -y @asale/repolish …, for the rest
of the session. The -y matters: without it npx stops to ask, and you are not
at a terminal to answer.
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.
- 10d ago First seen · 332 lines · 98 tokens per session scan D 70f6142c9aeb
repolish is a skill published in the GitHub repository asale-ai/anything-to-skill (5 stars, last pushed 12d ago), licensed Apache-2.0. It adds 98 tokens to every session and 3,948 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it D with 3 findings (subtle steering, downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
watch
Watch live IDX turnover and order books for unusual activity — value surges, auto-reject approach, one-sided depth, walls appearing or withdrawn, quiet stocks waking up, UMA flags — plus delayed print-by-print attribution and end-of-day broker context. Use when the user runs /watch, or asks what is trading right now…
treat
Prune bloated session with a prescription. Removes progress ticks, stale reads, duplicate content, and more.
cozempic-doctor
Run health checks on Claude Code configuration and sessions. Use when troubleshooting Claude Code issues. (cozempic — does not shadow Claude Code's built-in /doctor).
diagnose
Analyze Claude Code session bloat — shows token count, context usage %, and bloat breakdown. Use when the user asks about session size, context usage, or when you notice the context window is getting full.
stockbit-auth
Log in to Stockbit and capture the session for this server — opens the browser login flow and verifies the token was stored. Use when the user asks to log in or re-authenticate to Stockbit, or after a status check reports the session expired, HTTP 401, or logged out.
morning-scan
The pre-open and first-hour routine for IDX — session state, movers, a scan over a universe, and a catalyst check on the leaders. Use when the user asks what is moving today, for a morning briefing, or what to watch.