Skip to main content
Glama

Set up MediaPipe plugin (multi-modal)

setup_mediapipe_plugin

Load the torinmb MediaPipe engine once and enable face, hand, body, and segmentation pipelines in a single shot, avoiding multiple competing MediaPipe COMPs for the webcam.

Instructions

Drop the torinmb mediapipe-touchdesigner ENGINE in one shot and enable any combination of face, hand, body, and segmentation pipelines. Use this instead of running setup_face_tracking + setup_hand_tracking + setup_body_tracking + setup_segmentation separately — those tools each re-load the engine, resulting in multiple competing MediaPipe COMPs fighting for the webcam. This tool loads the engine ONCE and toggles its Face/Hand/Body/Segmentation pars. IMPORTANT: there is NO stock TouchDesigner MediaPipe; all five mediapipe tools (this one + the four setup_*_tracking tools) rely on the free torinmb plugin — install it first with tdmcp install mediapipe-touchdesigner. Output paths for face/hand/body are DATs (JSON landmark streams from the plugin), not CHOPs — use a Script CHOP adapter to convert to numeric channels. The engine requires the TD timeline to be PLAYING (uses an embedded browser for webcam capture).

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
tox_pathNoOverride path to the torinmb mediapipe-touchdesigner ENGINE .tox (MediaPipe.tox — the full tracker with webcam capture, NOT the bare pose_tracking.tox or hand_tracking.tox processors). Defaults to the package staged by `tdmcp install mediapipe-touchdesigner`.
parent_pathNoExisting COMP to load the engine into./project1
enable_faceNoEnable the Face detection pipeline inside the engine.
enable_handNoEnable the Hand tracking pipeline inside the engine.
enable_bodyNoEnable the Body/Pose tracking pipeline inside the engine.
enable_segmentationNoEnable the Segmentation pipeline (outputs a matte TOP; heavier GPU cost than the landmark pipelines).
source_video_pathNoOptional path to a video file to use as input instead of the live webcam. The engine's Camera/Source/Videofile/File par is probed in that order and the first match is set.
container_nameNoInner baseCOMP name. Matches the default used by setup_body_tracking / setup_hand_tracking so re-running is idempotent (the engine is reused, not duplicated).MediaPipe
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Beyond annotations (readOnlyHint=false, destructiveHint=false), the description reveals key behaviors: the engine loads once, toggles pars, outputs DATs not CHOPs, requires playing timeline, and has a probe order for source video. No contradiction with annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is fairly long but well-structured and front-loaded with the core purpose. Every sentence provides value, though minor redundancy exists (e.g., 'loads the engine ONCE' and 'Drop... ENGINE in one shot' overlap slightly). Still highly informative without being wasteful.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's complexity (8 params, no output schema, many sibling tools), the description covers prerequisites, behavioral constraints (DAT vs CHOP, timeline playing), source detection, idempotency, and differences from related tools. It provides all necessary context for correct use.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Despite 100% schema description coverage, the description adds significant meaning: clarifies tox_path is the full tracker, not bare processors; explains source_video_path probe order; emphasizes container_name for idempotency; and notes GPU cost for segmentation. These details aid correct invocation.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool drops the torinmb mediapipe-touchdesigner ENGINE and enables any combination of face, hand, body, and segmentation pipelines. It distinguishes itself from siblings like setup_face_tracking and others by explaining the advantage of a single engine load and avoiding competing COMPs.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly states when to use this tool instead of running individual setup tools, and warns against using them due to re-loading issues. Also includes prerequisite installation instructions (tdmcp install mediapipe-touchdesigner) and important notes about output types and timeline playback.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Pantani/tdmcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server