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 NairoDorian/TD_MCP --skill td-buildinggit clone --depth 1 https://github.com/NairoDorian/TD_MCPWrote 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/nairodorian/td_mcp/td-building)<a href="https://agentmods.dev/skills/nairodorian/td_mcp/td-building"><img src="https://agentmods.dev/badge/skills/nairodorian/td_mcp/td-building/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/nairodorian/td_mcp/td-building"><img src="https://agentmods.dev/badge/skills/nairodorian/td_mcp/td-building.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 | $0.00000 | $0.00982 |
| Opus 5.5 | $0.00000 | $0.00393 |
| Sonnet 5.5 | $0.00000 | $0.00196 |
| Haiku 4.5 | $0.00000 | $0.00098 |
Grade A, and why
td-building 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 2mo 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 — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
TouchDesigner building — AI assistant skill
You are helping someone build a TouchDesigner (TD) network. TD is node-graph creative-coding software by Derivative. Operator families: TOP (images), CHOP (channels/animation), SOP (geometry), POP (particles, 2023+), DAT (data/text), COMP (containers/3D).
Philosophy
- TD is pull-based: nodes cook only when their output is requested. Never force-cook unless you need a snapshot.
- Prefer CHOPs + expressions over Python for time-varying values.
- Always look up operator/parameter names before coding (use
td_docs_search/td_docs_parameter). Never guess a parameter code. - Wrap mutations in
ui.undofor a single Ctrl+Z revert.
Operator families
- TOP — texture/image operators (Blur, Noise, Level, Feedback, GLSL).
- CHOP — channel/animation operators (LFO, Audio Device In, Math, Filter).
- SOP — CPU geometry (Box, Sphere, Transform, Noise).
- POP — GPU particles, build 2023.10000+.
- DAT — data/text (Table, Text, Script, Execute, Web Client).
- COMP — networks/UI (Container, Geometry, Camera, Light, Render).
Critical conventions
- Parameter identifiers are lowercase, no spaces:
par.file, notfilename. op('x').par.file = 'path'— the classic Movie File In TOP example.- A CHOP→TOP takes its source via the
chopparameter, not an input wire. - Geometry COMP's create type is
geometryCOMP, notgeoCOMP. - Feedback TOP needs an input wired to initialize its buffer.
- GLSL TOP uniforms go on the Vectors page (runtime), not Constants page (compile-time only, fails silently).
Python essentials
op('noise1') # relative path
op('/project1/noise1') # absolute path
op('x').par.roughness.eval() # current value
op('x').par.roughness = 0.5 # direct set
op('x').par.roughness.expr = "me.time.seconds"
with ui.undo: op('x').par.value0 = 1 # undo-safe mutation
op('x').store('key', value) # persistent per-node state
op('x').fetch('key')
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.
- 2mo ago First seen · 93 lines · 0 tokens per session scan A 75a21424e926
td-building is a skill published in the GitHub repository NairoDorian/TD_MCP (2 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 982 tokens. 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.
td-building compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| td-buildingNairoDorian/TD_MCP | Skill | 2 | 2mo ago | MIT | A | 0 | Claude Code, Codex |
| imagegenopenai/codex | Skill | 134k | today | Apache-2.0 | A | 113 | Codex |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | today | MIT | A | 200 | Claude Code, Codex |
| playwright-climicrosoft/playwright | Skill | 99k | today | Apache-2.0 | A | 19 | Claude Code |
| browser-automationopenclaw/openclaw | Skill | 394k | today | MIT | A | 36 | Claude Code |
| systematic-debuggingobra/superpowers | Skill | 314k | 4d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 95 | Claude Code |