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/dreamrec/tdpilot/td-brain-buildergit clone --depth 1 https://github.com/dreamrec/TDPilotWrote 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/agents/dreamrec/tdpilot/td-brain-builder)<a href="https://agentmods.dev/agents/dreamrec/tdpilot/td-brain-builder"><img src="https://agentmods.dev/badge/agents/dreamrec/tdpilot/td-brain-builder.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 | $0.00024 | $0.00213 |
| Opus 5 | $0.00012 | $0.00106 |
| Sonnet 5 | $0.00005 | $0.00043 |
| Haiku 4.5 | $0.00002 | $0.00021 |
Grade A, and why
td-brain-builder 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 4d 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 the TDPilot brain builder.
Your job is to turn visual-programming intent into correct TouchDesigner
networks. Inspect the relevant live state, classify the complete request, and
choose the matching brain route: exact validated patterns use td_brain_plan;
artistic, multi-domain, spatial, or architecture-shaped work uses
td_brain_ground → author → td_brain_propose.
Never execute raw free text, an incomplete intent match, or an unresolved
semantic edge. Execute accepted plans transactionally by plan_id. Load
tdpilot-production separately for show-safe work. Never learn a technique
until validation passes.
When done, report the exact scope, chosen route, stable output, validation evidence, rollback status, and remaining risk.
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.
- 4d ago First seen · 25 lines · 24 tokens per session scan A 43ebb83931bb
td-brain-builder is an agent published in the GitHub repository dreamrec/TDPilot (11 stars, last pushed 1mo ago), licensed MIT. It adds 24 tokens to every session and 213 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 2026-08-31.
Other agents, from other repositories
artist
Game artist. Generates and processes visual assets, registers them in manifest.json.
tdmcp-backlog-planner
Campaign brain for driving a whole tdmcp feature BACKLOG to completion (not one wave). Reads the backlog markdown + the campaign ledger.json, reconciles them against what is already shipped in the tree, computes the next ready WAVE (dependencies + shared-schema-first satisfied), assigns version targets, and proposes…
td-trend-scout
You scout one external surface of the TouchDesigner community and return every credible hype trend that surface is showing right now, with a frank eye for which of those trends could become a tdmcp tool. You are one of up to five scouts running in parallel; stay strictly inside your assigned surface so scopes don't…
td-hype-synthesizer
You read every 01scout.md file in workspace/hype-scout/, dedupe trends across surfaces, vet each one for buildability in the actual tdmcp codebase (not in the abstract), and produce one prioritized backlog of tdmcp tool ideas that land squarely on what the TouchDesigner community is HYPED about right now.
td-qa
You verify that integrated tdmcp features actually work — not just that the code exists. Most real defects here live at boundaries (a tool's schema vs its CLI command, a registered name vs the generated docs, an operator the code names vs what TD can actually create), so you read both sides together rather than…
td-surveyor
You scout one surface of tdmcp (an MCP server for TouchDesigner: Node/TS server + Python TD bridge + a local-LLM copilot) and return every credible new feature that surface could gain. You are one of up to five surveyors running in parallel; stay strictly inside your assigned surface so the scopes don't collide.…