[Bug] False positive model-switch trigger in 3D graphics discussion

Resolved 💬 2 comments Opened Jun 12, 2026 by TsChain Closed Jun 15, 2026

Bug Description
I believe this was a false positive model-switch trigger.

The conversation was a normal software/product development discussion about a 3D body-map feature for a wellness/bodywork record app. The topic involved MakeHuman/MPFB2, Blender, glTF/GLB export, skinned meshes, skeleton rigs, bone names, vertex weights, normals, and three.js loading behavior.

The terms “body,” “skin,” “skeleton,” “bones,” and “muscles” were used strictly in the 3D graphics / rigging sense, not in any unsafe biological, medical, or harmful context.

The switch interrupted a safe technical debugging workflow. In particular, we were diagnosing why a rigged 3D human model’s limb bones were not responding in three.js. The root cause was that three.js sanitized bone names by stripping dots, so names like upperarm01.L became upperarm01L, causing the bone map to fail. This is ordinary 3D asset / WebGL debugging.

Please review this as a likely false positive and consider refining the classifier for safe 3D modeling, anatomy visualization, and body-map application development conversations.

Environment Info

  • Platform: darwin
  • Terminal: Apple_Terminal
  • Version: 2.1.173
  • Feedback ID: 2acacb06-ffe6-4c24-ae71-6eb801afa1f7

Errors

[]

View original on GitHub ↗

This issue has 2 comments on GitHub. Read the full discussion on GitHub ↗