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/varkart/visualcave/claude-mdgit clone --depth 1 https://github.com/varkart/visualcaveWrote 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/varkart/visualcave/claude-md)<a href="https://agentmods.dev/instructions/varkart/visualcave/claude-md"><img src="https://agentmods.dev/badge/instructions/varkart/visualcave/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 | $0.00619 | $0.00619 |
| Opus 5 | $0.00309 | $0.00309 |
| Sonnet 5 | $0.00124 | $0.00124 |
| Haiku 4.5 | $0.00062 | $0.00062 |
Grade A, and why
visualcave 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 — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Contributing to visualcave
Before You Open a PR
-
Test your change renders. Run
npm test— all Playwright tests must pass. Each test opens an HTML file in a real browser and asserts the Mermaid SVG renders without error. -
Keep SKILL.md under 700 words. Every word loads on every invocation. Move reference material to
references/.skills/visualcave/SKILL.mdis a byte-identical mirror required by the Antigravity CLI plugin loader — every SKILL.md edit must update both copies (enforced bytests/skill-md-guard.spec.js). -
One thing per PR. A fix, a new example, or a SKILL.md change — not all three.
-
New diagram type? Add an example file in
examples/and a row inreferences/diagram-types.md. Add the example to the Playwright test list intests/validate-examples.spec.js. -
New feature in capture.js / step-through.js / hover.js? Include a before/after test showing it works. These files run in the browser or via Node — prose alone isn't enough.
What Won't Be Accepted
- Third-party npm dependencies in production code.
capture.jsuses puppeteer — that's already a debt. Don't add more. - Hardcoded SVG coordinates. Everything must use Mermaid — no
<rect x="..." y="...">. - Breaking the template. The HTML template in SKILL.md is the core output format. Changes must keep:
data-sourcesnapshot beforemermaid.initialize(),prefers-reduced-motion, dark mode CSS vars, and thecopyMermaid/toggleThemepattern. - Renaming classDef colors. Other files depend on
:::yellow,:::blue, etc.
Running Tests
npm install
npx playwright install chromium
npm test # headless
npm run test:ui # Playwright UI mode
npm run test:headed # visible browser
Screenshots are saved to tests/screenshots/ (gitignored) after each run.
File Map
SKILL.md ← core skill (loaded on every invocation)
references/ ← loaded on demand
diagram-types.md ← type → keyword → example lookup
dark-mode.md ← dark mode token reference
theming.md ← custom brand theming guide
animated.md ← SVG animation classes (non-Mermaid only)
astro-component.md ← varkart.dev Astro tokens, component template, React/TSX output
examples/ ← live rendered output files
capture.js ← PNG/SVG/PDF/GIF/OG export (Node + puppeteer)
step-through.js ← sequential reveal UI
hover.js ← hover dim/highlight
tests/ ← Playwright validation suite
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 · 50 lines · 619 tokens per session scan A ac9dd40dac4e
visualcave CLAUDE.md is an instructions file published in the GitHub repository varkart/visualcave (2 stars, last pushed 1mo ago), licensed MIT. It adds 619 tokens to every session, about $0.0031 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
meta-agent-skills AGENTS.md
AGENTS.md instructions for yu-iskw/meta-agent-skills, covering agents.md, available skills, available agents, agent skills and what are agent skills?.
meta-agent-skills CLAUDE.md
Claude Code instructions for yu-iskw/meta-agent-skills, covering claude.md, available skills, available agents, agent skills and what are agent skills?.
skills CLAUDE.md
Claude Code instructions for lovstudio/skills, covering claude.md, what this is, repo layout, how users install and skills.yaml schema.
nopua AGENTS.md
Instructions for wuji-labs/nopua, covering nopua · codex 入仓规则, 一 · 项目定位, 二 · 起手必读, 三 · 工程与研究纪律 and 四 · 对外署名.
Dragon-Brain CLAUDE.md
Instructions for iikarus/Dragon-Brain, covering dragon brain — claude.md, the harness, audit remediation (april–may 2026, complete 2026-05-09), the lie this audit closed and the contract that matters now.
cli-continues parsers.instructions.md
Instructions for yigitkonur/cli-continues, covering parser review guidelines, crash safety (critical), required exports, jsonl streaming and tool summarizer.