Borrowing it
Nothing to install: this file belongs to dylanroscover/Embody. 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/dylanroscover/Embody/main/.claude/skills/specimen-authoring/SKILL.mdgit clone --depth 1 https://github.com/dylanroscover/EmbodyWrote 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/dylanroscover/embody/specimen-authoring)<a href="https://agentmods.dev/skills/dylanroscover/embody/specimen-authoring"><img src="https://agentmods.dev/badge/skills/dylanroscover/embody/specimen-authoring/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/dylanroscover/embody/specimen-authoring"><img src="https://agentmods.dev/badge/skills/dylanroscover/embody/specimen-authoring.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00038 | $0.05049 |
| Opus 5 | $0.00019 | $0.02524 |
| Sonnet 5 | $0.00008 | $0.01010 |
| Haiku 4.5 | $0.00004 | $0.00505 |
Grade A, and why
specimen-authoring 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 9d 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 — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Specimen Authoring
How to build a Specimen for the Embody Collection -- a transparent, reusable TDXN network that demonstrates a TouchDesigner technique. Two specimens set the bar: reaction-diffusion (generative, a GPU feedback simulation) and kaleidoscope (compositing, a reusable polar-mirror component).
The bar -- every Specimen must clear it
- Clear use -- a user can finish "I'd use this for ___" (a VJ loop, a drop-in component, a texture/displacement source, a real learning reference).
- Non-obvious technique -- a "how'd they do that?" moment, not a one-TOP drag.
- Striking -- worth opening and exploring.
- Drop-in -- clean input/output, exposed parameters, self-contained, bounded for performance.
Generic noise plus a colorize is NOT a specimen. If a beginner makes it by accident, cut it.
Workflow (one at a time)
- Build in the sandbox COMP
/specimen_lab/<name>(NOT inside the Embody COMP -- the toe file, somewhere neutral). Iterate freely there. - Gate performance (see
performance.md): baselineget_project_performancebefore, re-check after each heavy step. Feedback loops and GLSL are the usual cost. fps below ~90% of target is a stop condition -- optimize before continuing. Beware: a concurrentcapture_topornumpyArray()stalls the GPU and makes the fps reading dip; take a clean reading without them. - Verify it actually animates and cooks before believing it -- see "Cook demand" below. Never claim animation from a single forced capture.
- Judge the look by actually looking --
capture_topthen read the frame; loadvisual-aesthetics.capture_topforce-cooks, so it can show a frame your live (undemanded) viewer is NOT showing -- a stale viewer and a fresh capture are different frames; that mismatch is real, not a hallucination. - Ask the user to review before persisting. Persist only when it is genuinely good, animating, and clean.
Cook demand -- the trap that bit both specimens
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.
- 9d ago First seen · 131 lines · 38 tokens per session scan A 4bc790438f27
specimen-authoring is a skill published in the GitHub repository dylanroscover/Embody (170 stars, last pushed today), licensed MIT. It adds 38 tokens to every session and 5,049 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
issue_approve
Approve issue to transition to next workflow status.
metagit-cli
CLI-only shortcuts for metagit agents — workspace catalog, discovery, prompts, sync, layout, and config. Use instead of MCP or HTTP API when operating from a shell or agentmode session.
agentcore-investigation
Investigate Bedrock AgentCore runtime sessions via CloudWatch Logs Insights — resolve session/trace IDs, query OTEL spans, filter noise, build timelines. Use when debugging AgentCore agent sessions, tracing tool calls, or analyzing latency.
amazon aurora dsql
Deprecated compatibility redirect for Aurora DSQL guidance. Use when a request concerns DSQL, Aurora DSQL, distributed SQL, DSQL schemas, migrations, queries, authentication, performance, or application development.
daemon-loop
Autonomous recurring agent tasks — converts workflows into persistent background daemons that run on intervals. Stolen from Boris Cherny's Claude Code /loop pattern (2026-03-31).
adhoc-flow
Workflow for the rest of tasks: lightweight documentation, build, track, synchronize, etc.