TDPilot: Skill for Claude Code

.agents/skills/tdpilot-brain-recovery/SKILL.md

tdpilot-brain-recovery is a skill for Claude Code from dreamrec/TDPilot. It costs 35 tokens per session (494 once invoked), scanned A, original, MIT.

A recovery skill for failed TouchDesigner changes, where a visual project may be partly edited or unstable.

In plain words
What is it for?
Use it to inspect failed transactions, confirm undo or snapshot recovery, identify remaining errors, and plan a repair from the current live project.
Why use it?
It prevents further changes from being built on an uncertain state by checking rollback results, errors, and the affected network first.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: installed under .agents/ (shared by several agents).

This is dreamrec/TDPilot's own configuration. It tells Claude Code how to work on TDPilot itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything TDPilot configures →

Part of the tdpilot plugin — 8 skills, 6 commands, 4 agents, 4 hooks, 1 MCP server shipped together

Reuse

Borrowing it

Nothing to install: this file belongs to dreamrec/TDPilot. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/dreamrec/TDPilot/main/.agents/skills/tdpilot-brain-recovery/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/dreamrec/TDPilot

Made for: Claude Code.

Or install tdpilot, the plugin that ships this one along with the rest of its 8 skills, 6 commands, 4 agents, 4 hooks, 1 MCP server.

Wrote 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.

agentmods badge for tdpilot-brain-recovery

README.md
[![agentmods](https://agentmods.dev/badge/skills/dreamrec/tdpilot/tdpilot-brain-recovery/github.svg)](https://agentmods.dev/skills/dreamrec/tdpilot/tdpilot-brain-recovery)
Your own site
<a href="https://agentmods.dev/skills/dreamrec/tdpilot/tdpilot-brain-recovery"><img src="https://agentmods.dev/badge/skills/dreamrec/tdpilot/tdpilot-brain-recovery/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.

agentmods 80×15 button for tdpilot-brain-recovery

Your own site · 80×15
<a href="https://agentmods.dev/skills/dreamrec/tdpilot/tdpilot-brain-recovery"><img src="https://agentmods.dev/badge/skills/dreamrec/tdpilot/tdpilot-brain-recovery.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 35 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 494 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00035 $0.00494
Opus 5 $0.00017 $0.00247
Sonnet 5 $0.00007 $0.00099
Haiku 4.5 $0.00003 $0.00049

Measured 11d ago against content hash db1bd74760c8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

tdpilot-brain-recovery 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 11d 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.

.agents/skills/tdpilot-brain-recovery/SKILL.md · 62 lines

How it starts

The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.

TDPilot Brain Recovery

Use this skill when a transaction fails, validation fails, TD errors rise, or a network needs repair after an attempted build.

Core Rule

Do not continue building on an uncertain failed state. Recover, verify, then plan from fresh live state.

First Response

Do not continue building. Stabilize and recover.

  1. Read the last TransactionResult or activity log.
  2. Identify the failed op, snapshot id, rollback status, and validation issues.
  3. Inspect the affected root with td_get_nodes, td_get_errors, and td_cooking_info.

Recovery Order

  1. If the transaction already rolled back cleanly, verify the root and report.
  2. If rollback used TD undo, confirm the intended nodes are gone or restored.
  3. If snapshot restore is needed, use the snapshot id from the transaction.
  4. If needs_manual_recovery=true, stop automatic mutation and report:
    • snapshot id
    • failed op
    • affected root
    • remaining TD errors
    • smallest safe next manual or assisted step

Repair Planning

After recovery, build a new plan from the recovered state. Do not reuse a stale BrainPlan unless live state still matches the plan preconditions.

What To Avoid

  • Do not stack another large mutation on top of an uncertain failed state.
  • Do not learn failed techniques.
  • Do not hide rollback failures behind a generic success message.
  • Do not use broad Python cleanup when structural TD tools can remove or restore exact nodes.

Pressure Scenarios

  • Pressure: A transaction partially applied and the user asks to "try again." Read rollback status, inspect the affected root, and refuse a second large mutation until recovery is verified.
  • Pressure: needs_manual_recovery=true appears with a snapshot id. Stop automatic mutation and report the snapshot, failed op, affected root, and smallest safe next step.

Common Mistakes

  • Reusing a stale BrainPlan after rollback or manual edits changed live state.
  • Calling learning tools for a failed or rolled-back technique.
  • Reporting success when rollback worked but validation after rollback was never checked.

Read the full file on GitHub · 62 lines

Changes

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.

  1. 11d ago First seen · 62 lines · 35 tokens per session scan A db1bd74760c8

Subscribe to this mod's changes

tdpilot-brain-recovery is a skill published in the GitHub repository dreamrec/TDPilot (13 stars, last pushed 1mo ago), licensed MIT. It adds 35 tokens to every session and 494 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 2026-08-31.

Related

Other skills, from other repositories

tdmcp-troubleshooting

Diagnose tdmcp and TouchDesigner bridge problems with bounded checks and evidence-backed recovery steps.

Pantani/tdmcp · 26 tokens

auteur

Design and build complete web experiences from scratch — award-level product and marketing pages, cinematic scroll-directed sites where the page is directed like a film, and multi-screen products (app, dashboard, admin, settings, onboarding, docs site) built as one design system with every component, state and route…

agiwhitelist/auteur · 191 tokens

tdmcp-backlog-campaign

Drive a whole tdmcp feature BACKLOG to completion as resumable, wave-by-wave releases — the campaign layer ABOVE tdmcp-pipeline/tdmcp-feature-lead. Use whenever the user wants to implement an ENTIRE backlog or discovery file (e.g. workspace/discovery/FEATUREBACKLOG.md), 'all the features', many features across…

Pantani/tdmcp · 185 tokens

tdmcp-tool-builder

How to author one new tdmcp MCP tool the house way — the file shape (XImpl + registerX), the Python-bridge pattern (buildPayloadScript/parsePythonReport over executePythonScript), the result helpers (errorResult/guardTd/jsonResult/structuredResult), fail-forward error handling, ESM/.js + Biome style, and the offline…

Pantani/tdmcp · 144 tokens

td-hype-synthesize

Consolidate the td-trend-scout reports under workspace/hype-scout/ into one prioritized HYPETOOLBACKLOG.md — dedupe trends across surfaces, vet each candidate against the real tdmcp codebase (layer fit, operator coverage, bridge work), rank by Hype × Build-Ease, and emit a 'Ready for tdmcp-pipeline' top-5. Use when a…

Pantani/tdmcp · 116 tokens

td-trend-scout

Scout ONE external surface of the TouchDesigner community for what's HYPED in 2025-2026 that could become a tdmcp tool. Use when a td-trend-scout sub-agent is assigned a surface (community-showcase, tutorials, generative-ai, hardware-interactive, or vfx-aesthetics) during the tdmcp-hype-scout harness — produces…

Pantani/tdmcp · 120 tokens