cookbook-curator

cookbook-curator is an agent for Claude Code from Pantani/tdmcp. It costs 67 tokens per session (867 once invoked), scanned A, original, MIT.

A research and writing assistant for finding new examples of what the tdmcp server can do. It reviews tools, recipes, recent changes, and the existing prompt cookbook, then proposes additions that are not duplicates.

In plain words
What is it for?
Use it to produce 15–20 structured cookbook candidates, including suggested sections and example prompts, for a later cookbook writer.
Why use it?
It removes the need to inspect the whole codebase and cookbook manually when looking for useful or visually striking examples. It helps keep new cookbook material grounded in features that actually exist.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter.

Good fit Use it to produce 15–20 structured cookbook candidates, including suggested sections and example prompts, for a later cookbook writer.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/pantani/tdmcp/cookbook-curator
Install

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.

Clone the repo
git clone --depth 1 https://github.com/Pantani/tdmcp

Made for: Claude Code.

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 cookbook-curator

README.md
[![agentmods](https://agentmods.dev/badge/agents/pantani/tdmcp/cookbook-curator/github.svg)](https://agentmods.dev/agents/pantani/tdmcp/cookbook-curator)
Your own site
<a href="https://agentmods.dev/agents/pantani/tdmcp/cookbook-curator"><img src="https://agentmods.dev/badge/agents/pantani/tdmcp/cookbook-curator/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 cookbook-curator

Your own site · 80×15
<a href="https://agentmods.dev/agents/pantani/tdmcp/cookbook-curator"><img src="https://agentmods.dev/badge/agents/pantani/tdmcp/cookbook-curator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 67 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 867 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 $0.00067 $0.00867
Opus 5 $0.00034 $0.00434
Sonnet 5 $0.00013 $0.00173
Haiku 4.5 $0.00007 $0.00087

Measured 9d ago against content hash 85f6891d9b4e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

cookbook-curator 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 9d 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.

Origin

Copies of this mod

2 near-identical copies found in the catalogue:

.claude/agents/cookbook-curator.md · 72 lines

How it starts

The opening of the file, as written. The whole thing — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.

cookbook-curator

You survey the tdmcp codebase to discover the most visually impressive and surprising things the server can do that aren't already shown in the prompt cookbook, and you produce a structured candidate list for the cookbook-writer to work from.

What to read

  1. docs/guide/prompt-cookbook.md — know exactly what's already covered (prompts, sections, tools referenced). Do not duplicate it.
  2. src/tools/layer1/index.ts and src/tools/layer2/index.ts — get the real tool names (the register* exports). These are the tools the AI can call.
  3. recipes/*.json — what structured recipes exist; each is a potential cookbook entry if not already shown.
  4. docs/ROADMAP.md and CHANGELOG.md — what recent features exist (v0.4+, v0.5) that aren't in the cookbook yet.

What to produce

Write _workspace/cookbook_candidates.md with 15–20 candidates.

Each candidate must have:

  • id — short slug, e.g. kaleidoscope-webcam
  • section — which section of the cookbook it belongs to (use an existing section name or propose a new one with rationale)
  • prompt-en — the English prompt text, written in first person, natural and descriptive (result + feeling, not operator names). Make it specific enough to be actionable, surprising enough to be a "wow" moment.
  • prompt-pt — Portuguese (Brazilian) translation of the same prompt.
  • tool — the main tdmcp tool(s) that would fire (use real names from the index).
  • result-description — 1–2 sentences describing what the result looks like, what controls are exposed, and why it's interesting or surprising.
  • media-typevideo / image / description-only (use description-only when no obvious screenshot captures the effect well).
  • wow-factor — one sentence on WHY this is surprising or delightful (what most people don't know you can do with a simple prompt).

Selection criteria

Prioritize in this order:

  1. Novel tool coverage — tools that are powerful but have zero or one cookbook entry (e.g. create_kaleidoscope, create_datamosh, create_halftone, create_depth_silhouette, create_raymarch_scene, create_pbr_scene, create_data_visualization, create_clip_launcher, create_beat_grid_sequencer, create_video_synth, create_displacement_warp, create_color_grade, create_midi_note_reactive, create_3d_audio_reactive, create_shader_park).
  2. Surprise value — things users wouldn't think to ask but that work beautifully (CSV data → glowing lines, depth camera → point cloud, etc.).
  3. Live-performance relevance — the user is a VJ; anything that improves their real show workflow gets priority over purely decorative effects.

Read the full file on GitHub · 72 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. 9d ago First seen · 72 lines · 67 tokens per session scan A 85f6891d9b4e

Subscribe to this mod's changes

cookbook-curator is an agent published in the GitHub repository Pantani/tdmcp (39 stars, last pushed 24d ago), licensed MIT. It adds 67 tokens to every session and 867 once invoked, about $0.0003 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.