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 instructions/dreamrec/tdpilot/agents-mdgit clone --depth 1 https://github.com/dreamrec/TDPilotWhat 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 | $0.00374 | $0.00374 |
| Opus 5 | $0.00187 | $0.00187 |
| Sonnet 5 | $0.00075 | $0.00075 |
| Haiku 4.5 | $0.00037 | $0.00037 |
Grade A, and why
TDPilot AGENTS.md 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
TDPilot Agent Guide
Default Workflow
For TouchDesigner work, use the TDPilot brain skills:
tdpilot-brain-explorerfor read-only project discovery.tdpilot-brain-builderfor non-trivial network construction.tdpilot-brain-validatorfor concept, graph, runtime, and visual checks.tdpilot-brain-recoveryafter failed transactions or unstable TD state.tdpilot-brain-releasebefore publishing MCP surface or plugin changes.
TouchDesigner Safety
- Inspect the relevant live scope before mutating; batch independent reads when useful.
- Route exact validated patterns through
td_brain_plan. - Route artistic, multi-domain, spatial, camera/depth/fog, or implicit architecture through
td_brain_ground→ author →td_brain_propose. - Execute only valid plans: call
td_brain_executeortd_transaction_apply, not raw free text. - Treat incomplete intent coverage and unresolved semantic edges as execution hard stops. A blocked pattern plan may continue through the concept-authoring route.
- Check errors at logical checkpoints and completion, not after every primitive operation.
- Learn only validated outcomes.
Verification
Run focused tests for changed areas, then:
uv run pytest -q
uv run python scripts/eval_brain_golden.py
uv run python scripts/brain_live_smoke.py --dry-run
uv run python scripts/audit_brain_skills.py
uv run python scripts/audit_plugin_surface.py
uv run python scripts/smoke_mcp_registry.py
uv run python scripts/check_versions.py
If TouchDesigner is available, add uv run python scripts/brain_live_smoke.py --live.
If TouchDesigner is not running, state that live smoke validation was not run.
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 · 39 lines · 374 tokens per session scan A c80ed69cdd5f
TDPilot AGENTS.md is an instructions file published in the GitHub repository dreamrec/TDPilot (11 stars, last pushed 1mo ago), licensed MIT. It adds 374 tokens to every session, about $0.0019 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 instructions, from other repositories
genart-skill AGENTS.md
Instructions for camilleroux/genart-skill, covering working on this repo, architecture invariants and testing.
liminal AGENTS.md
AGENTS.md instructions for KyaniteLabs/liminal, covering agents.md — liminal agent instructions, mandatory coding skill, agent skills, active matt pocock skill surface and issue tracker.
liminal CLAUDE.md
Claude Code instructions for KyaniteLabs/liminal, covering liminal — project instructions, mandatory coding skill, git hygiene (mandatory for all agents), the 6 rules and parallel agent isolation (mandatory).
liminal copilot-instructions.md
Copilot instructions for KyaniteLabs/liminal, covering kyanitelabs — ai agent instructions, organization, code quality, security and testing.
genart-skill CLAUDE.md
Instructions for camilleroux/genart-skill, a project described as: Generative art skill for Claude Code — deterministic, hash-seeded, onchain-ready.
Embody CLAUDE.md
Instructions for dylanroscover/Embody, covering embody + envoy, project overview, critical rules, approach guidelines and project structure.