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 agentmods add instructions/plenoai/metsuke/claude-mdgit clone --depth 1 https://github.com/plenoai/metsukeWrote 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/instructions/plenoai/metsuke/claude-md)<a href="https://agentmods.dev/instructions/plenoai/metsuke/claude-md"><img src="https://agentmods.dev/badge/instructions/plenoai/metsuke/claude-md.svg" alt="Measured on agentmods" 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 | $0.01228 | $0.01228 |
| Opus 5 | $0.00614 | $0.00614 |
| Sonnet 5 | $0.00246 | $0.00246 |
| Haiku 4.5 | $0.00123 | $0.00123 |
Grade A, and why
metsuke CLAUDE.md 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 3d 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 — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Metsuke — Project Conventions
CSS Naming: Strict BEM
All CSS classes follow BEM (Block Element Modifier).
Syntax
.block
.block__element
.block--modifier
.block__element--modifier
- Block: standalone component (
header,repo-card,dash-card,pagination) - Element: part of a block, separated by
__(header__left,repo-card__name) - Modifier: variant/state of a block or element, separated by
--(badge--pass,dash-card--empty) - Multi-word blocks/elements use single hyphens:
page-header,repo-card,inline-row
State classes
JS-toggled states use is-* prefix as adjacent classes:
<a class="nav__item is-active">
<button class="btn--verify is-running">
<div class="pr-item is-selected">
Utility classes
Spacing and text utilities use u- prefix:
.u-mt-0 /* margin-top: 0 */
.u-mb-0 /* margin-bottom: 0 */
.u-text-muted
Rules
- Never style IDs (
#foo). IDs are for JS targeting only. - Never reuse a component class (e.g.
btn--verify) on unrelated elements. Use the component's own scoped styles. - Tag selectors inside blocks are acceptable for leaf nodes (
.findings__table th). - Keep nesting flat:
block__elementonly. Neverblock__element__subelement. - New CSS classes MUST follow this BEM convention.
Block inventory
| Block | Purpose |
|---|---|
header |
Site header (brand, nav, user badge) |
nav |
Navigation links |
user-badge |
Logged-in user info + logout |
section |
Content section with title |
card |
Generic card container |
badge |
Status badge (--pass, --fail, --review, --na, --private) |
btn |
Button (--link, --verify) |
page-header |
Page title + toolbar row |
toolbar |
Horizontal control group |
inline-row |
Inline flex row for badges (--tight) |
pagination |
Pagination controls |
findings |
Findings table section |
summary |
Summary bar |
form-group |
Form field group |
kbd-help |
Keyboard shortcut overlay |
empty-state |
No-results placeholder |
skeleton |
Loading skeleton |
dashboard |
Dashboard grid |
dash-card |
Dashboard metric card |
activity |
Recent activity list |
repo-card |
Repository list card |
repo-grid |
Repository list container |
pr-item |
Pull request list item |
pr-detail |
PR expanded detail |
pr-state |
PR state badge (--open, --closed, --merged, --draft) |
release-item |
Release list item |
release-form |
Release verification form |
audit |
Audit log section |
type-badge |
Audit type indicator (--release, --pr, --repo) |
trigger-badge |
Audit trigger indicator (--manual, --webhook) |
install |
Settings installation item |
tag |
Label tag (--org) |
code-wrap |
Code block with copy button |
error-page |
Standalone error page |
error-inline |
JS-rendered inline error |
readme |
README display area |
search-bar |
Search input wrapper |
detail-heading |
Repo detail page heading |
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.
- 3d ago First seen · 129 lines · 1,228 tokens per session scan A 2ef3870da7db
metsuke CLAUDE.md is an instructions file published in the GitHub repository plenoai/metsuke (0 stars, last pushed 6d ago), licensed MIT. It adds 1,228 tokens to every session, about $0.0061 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 2026-08-31.
Other instructions, from other repositories
weapp-tailwindcss CLAUDE.md
Claude Code instructions for sonofmagic/weapp-tailwindcss, covering claude.md, project overview, monorepo structure, common development commands and build.
gsap-skills scrolltrigger.instructions.md
Instructions for greensock/gsap-skills: When writing or suggesting scroll-linked GSAP code (ScrollTrigger).
nextarter-tailwind AGENTS.md
AGENTS.md instructions for agustinusnathaniel/nextarter-tailwind, covering next.js: always read docs before coding, ai agent guidance: nextarter-tailwind, 1. mental model for agents, implementation delegation and tooling strategy.
lovable-boilerplate global.instructions.md
Instructions for chihebnabil/lovable-boilerplate, covering react + vite + shadcn/ui template - premium design system, critical rules - read first, never create, always create and essential commands.
screenshot-to-code CLAUDE.md
Instructions for abi/screenshot-to-code, a project described as: Drop in a screenshot and convert it to clean code (HTML/Tailwind/React/Vue).
shadcn-htmx AGENTS.md
Instructions for productdevbook/shadcn-htmx, covering agent guide, hard rules, what's vendored, and when to read each, repos/htmx/ — htmx v4 source (branch: four-dev) and repos/tailwindcss/ — tailwind css v4 source (branch: main).