Education & Research

Anatomy

by Kushal Sokke Malleshappa

Overview

Anatomy opens an interactive 3D human body inside ChatGPT. Ask about a muscle, bone, nerve, joint or organ and ChatGPT frames it, fades whatever covers it, and labels what matters. Peel away muscle layers, cut through the body, isolate a structure, colour-code what belongs together, and play movements like bending, squatting or lifting a knee. Anatomy can also draw a muscle's line of pull across the hip or knee, so you can watch a muscle change its job as the joint moves. Tap anything to ask about it, mark where something hurts, or ask for a quick find-it quiz. Anatomy shows reference anatomy for learning and explanation; it does not diagnose or give medical advice.

Tools

find_anatomy

ChatGPT

Search the 3D model’s exact structure names across the whole body (bones, muscles, nerves, joints and ligaments, vessels, organs). Plain words and groups work, several at once: "hamstrings, rotator cuff, sciatic nerve". Paired structures end in .l / .r. Use before naming structures in a scene.

open_anatomy

ChatGPT

Open Anatomy, an interactive whole-body 3D viewer of the human body, on a scene. A scene is a set of verbs on one layered body. FRAME (focus, view, zoom) · REVEAL (systems, side, isolate, hide, ghost, xray, peel, section, reveal) · MARK (marks, labels, pins) · MOVE (movement or keyframes, pull) · GUIDE (steps) · TEST (quiz). The person can do the same things by hand: tap to select, hide/fade/isolate, peel, cut, search, scrub, mark where it hurts, and tap Ask. Their actions come back through read_anatomy_view. Put everything the person should see in this one call (use steps for a multi-stage explanation): some hosts mount the widget after the response. Returns a sessionId for update_anatomy and read_anatomy_view.

read_anatomy_view

ChatGPT

Before answering a follow-up, read the viewer: whether it shows the latest scene, which step, what the person tapped (and which bones that structure touches), what they hid, faded, isolated, peeled or cut, where they marked pain, the paused movement frame with joint angles, measured pull-line moment arms, muscle length change, and quiz answers.

update_anatomy

ChatGPT

Replace the scene in the viewer that is already open (pass its sessionId). Send the whole scene you want; omitted fields reset to defaults except systems and side, which carry over. A scene is a set of verbs on one layered body. FRAME (focus, view, zoom) · REVEAL (systems, side, isolate, hide, ghost, xray, peel, section, reveal) · MARK (marks, labels, pins) · MOVE (movement or keyframes, pull) · GUIDE (steps) · TEST (quiz). The person can do the same things by hand: tap to select, hide/fade/isolate, peel, cut, search, scrub, mark where it hurts, and tap Ask. Their actions come back through read_anatomy_view. viewerReceipt "confirmed" means the mounted viewer applied it; "queued" means it appears when the viewer mounts.

Capabilities

WritesInteractive

Example Prompts

Click any prompt to copy it.

Links

Documentation ↗Privacy Policy ↗

App Stats

Tools

Prompts

ChatGPT

Platforms