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 agents/bottobot/touchdesigner-mcp-server/implement-experimentalgit clone --depth 1 https://github.com/bottobot/touchdesigner-mcp-serverWhat 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 | $0.00015 | $0.00576 |
| Opus 5 | $0.00008 | $0.00288 |
| Sonnet 5 | $0.00003 | $0.00115 |
| Haiku 4.5 | $0.00002 | $0.00058 |
Grade A, and why
implement-experimental 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 2d 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.
What it actually says
You are implementing MCP tools for experimental TouchDesigner techniques.
Project location: /home/robert/Documents/TD-MCP/touchdesigner-mcp-server
Prerequisite: /wiki/data/experimental/ should have been populated by the previous research agent.
Tasks:
-
Create /tools/get_experimental_techniques.js: InputSchema: { category: { type: 'string', enum: ['glsl','gpu-compute','machine-learning','generative-systems','audio-visual','networking','python-advanced'], required: false }, name: { type: 'string', description: 'Technique name to look up', required: false } } Logic: Load /wiki/data/experimental/{category}.json or search all categories if no category given Return: Full technique details including codeExample
-
Create /tools/search_experimental.js: InputSchema: { query: { type: 'string', required: true, description: 'Search term for experimental techniques' }, version: { type: 'string', required: false, description: 'Filter by minimum TD version' } } Logic: Search across all experimental categories, return matching techniques
-
Create /tools/get_glsl_pattern.js: InputSchema: { pattern: { type: 'string', required: true, description: 'GLSL pattern type (raymarching/reaction-diffusion/feedback/procedural)' }, includeCode: { type: 'boolean', required: false, default: true } } Logic: Return GLSL code examples and setup instructions from the glsl experimental data
-
Add experimental operator data files to /wiki/data/processed/:
- feedback_top.json (Feedback TOP — requires version 099+, key for generative systems)
- noise_top.json (Noise TOP — procedural generation workhorse)
- render_top.json (Render TOP — 3D scene rendering)
- geo_comp.json (Geometry COMP — 3D geometry container)
- script_top.json (ensure it exists with GPU compute info, don't overwrite if already comprehensive)
- null_top.json (Null TOP — pass-through, debugging)
- level_top.json (Level TOP — color grading and correction)
- blur_top.json (Blur TOP — multiple blur algorithms)
- transform_top.json (Transform TOP — 2D transformations)
- composite_top.json (Composite TOP — blending modes)
For each operator file, include versionSupport field.
-
Update /index.js to register: get_experimental_techniques, search_experimental, get_glsl_pattern
-
Update /wiki/data/search-index/search-index.json to include all new experimental operators and techniques
Follow exact code patterns from existing tools for consistency.
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.
- 2d ago First seen · 55 lines · 15 tokens per session scan A fa1cd74e08ec
implement-experimental is an agent published in the GitHub repository bottobot/touchdesigner-mcp-server (76 stars, last pushed 14d ago), licensed MIT. It adds 15 tokens to every session and 576 once invoked, about $0.0001 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 agents, from other repositories
discovery
How an agent finds these surfaces without being told: the API catalog, DNS-based discovery, an agent-skills index, and the places people actually look.
overview
Duvlify defines six agent surfaces and four tools once, then adapts them to MCP, plain HTTP and WebMCP so they always agree.
mcp
The page outline agents read for line offsets, the in-browser WebMCP bridge and when to run it yourself, the rate limits, and how the surfaces are tested.
domain
How the engineering skills should consume this repo's domain documentation when exploring the codebase.
omp
Install Remarc's public OMP plugins, verify the MCP connection, and pair a session for instant comment delivery.
evaluator-spawn
Generic TaskGraph evaluator shell — executes Transition and Hypothesis nodes the orchestrator dispatched via bobpreparenode. Carries the union of evaluator-family tools; the dispatched brief's allowedtoolsfornode[] is the per-spawn constraint enforced by the X.6 mechanical verifier on bobfinalizenode.