Borrowing it
Nothing to install: this file belongs to nirholas/three.ws. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/nirholas/three.ws/main/.agents/skills/anatomy/SKILL.mdgit clone --depth 1 https://github.com/nirholas/three.wsWrote 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/nirholas/three.ws/anatomy)<a href="https://agentmods.dev/skills/nirholas/three.ws/anatomy"><img src="https://agentmods.dev/badge/skills/nirholas/three.ws/anatomy/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/nirholas/three.ws/anatomy"><img src="https://agentmods.dev/badge/skills/nirholas/three.ws/anatomy.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 | $0.00126 | $0.01211 |
| Opus 5.5 | $0.00050 | $0.00484 |
| Sonnet 5.5 | $0.00025 | $0.00242 |
| Haiku 4.5 | $0.00013 | $0.00121 |
Grade A, and why
anatomy scanned grade A 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 14d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s -X POST https://three.ws/api/anatomy \ How it starts
The opening of the file, as written. The whole thing — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Anatomy: machines as interactive cutaways
Describe a machine in one line and get back an interactive isometric wireframe cutaway on three.ws. Visitors can rotate it, step through a guided tour, explode it, cut it open, switch to x-ray and click any part to read what it does. Moving parts really move, and the physics is drawn with WebGL shaders.
The result is a permalink such as https://three.ws/anatomy/<id>. Share it as
the answer. It works on desktop and phones, with no account needed to view it.
Two ways to build one
1. You write the spec (best quality, recommended)
You know the machine. Write the spec yourself and publish it.
- Read reference.md. It is the full spec format: parts and shapes, motion, effects, flows, the tour and the quality bar. The server validates against exactly these rules.
- Write one JSON object for the machine the user asked about. If they named a
real machine (for example the Rolls-Royce Trent 900 on the A380), model that
machine and its real proportions, stage counts and numbers. If they were
vague, choose the most iconic real example and say which in
subtitle. - Publish it:
curl -s -X POST https://three.ws/api/anatomy \
-H 'content-type: application/json' \
--data-binary @- <<'JSON'
{ "action": "publish", "prompt": "<the user's request>", "spec": { ...your spec... } }
JSON
The response is { "design": { "id", "url", "title", "stats" }, "warnings": [] }.
Give the user design.url. stats counts parts, moving instances, effects,
flows and tour steps.
warnings lists anything the normalizer repaired, such as an unknown parent, an
unknown focus id or a clamped value. The machine still renders. If a warning
points at something that matters (a missing part, a broken hierarchy), fix the
spec and publish again. Each publish gets a new permalink.
2. Ask three.ws to write it
When you cannot write the spec yourself, let the server's writer do it. This takes one to several minutes.
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.
- 14d ago Changed cbce16e48252
- 27d ago First seen · 97 lines · 126 tokens per session scan A ea5ce6ea7d3e
anatomy is a skill published in the GitHub repository nirholas/three.ws (269 stars, last pushed yesterday), licensed Apache-2.0. It adds 126 tokens to every session and 1,211 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-11.
anatomy compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| anatomynirholas/three.ws | Skill | 269 | yesterday | Apache-2.0 | A | 126 | Claude Code |
| interview-prepRightNow-AI/openfang | Skill | 18k | 4mo ago | Apache-2.0 | A | 18 | Claude Code, Codex |
| transformer-attentionVectifyAI/OpenKB | Skill | 4.9k | 3mo ago | Apache-2.0 | A | 46 | Claude Code, Codex |
| notebooklmjmagar/claude-homelab | Skill | 87 | 1mo ago | MIT | A | 67 | Claude Code |
| offensive-fuzzing-courseMiosa-osa/OSA | Skill | 78 | 9d ago | Apache-2.0 | A | 70 | Claude Code, Codex |
| omd-videoAbyssCN/oh-my-dag | Skill | 44 | 26d ago | MIT | A | 158 | Claude Code, Codex |
| breaking-into-productRefoundAI/lenny-skills | Skill | 1.5k | 3mo ago | MIT | A | 38 | Claude Code, Codex |