mcpField
C
calibrate-model

calibrate-model

github.com/playcanvasDevelopment31309 installsUpdated 1w ago
Repository

About calibrate-model

A calibration skill for placing GLB models consistently in an app. It records measured bounds, intended size, scale, position, and facing per asset, with guidance for animated models and support points.

Key features

  1. 01Measure each GLB once
  2. 02Record per-asset tuning values
  3. 03Calculate scale and vertical offset
  4. 04Set facing after app verification
  5. 05Nest roots, yaw, and visual transforms
  6. 06Check animated model contact

Use cases

  1. 01Calibrating imported models to real or relative dimensions
  2. 02Aligning models consistently on measured supports

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/playcanvas/skills --skill calibrate-model

After installing, type/calibrate-modelin Claude Code to confirm it loaded.

Bundled files

References and scripts installed along with SKILL.md.

references/
direct-engine.md
react.md
web-components.md

Details

From the project README

Model calibration

Measure each unique GLB once with inspect-glb, only trusting bounds when boundsSource is vertices, and keep one tuning record per asset:

const ASSET_TUNING = {
    model: {
        boundsSource: 'vertices', dims: [8, 7, 18], center: [0, 2.5, 0], groundOffset: 1,
        intended: { dimension: 'length', size: 18 }, scale: 1, y: 1, yaw: 180
    }
} as const;
  1. Pick the dimension that defines the model's real size, in metres or relative to an already calibrated model.
  2. Set scale = intendedSize / measuredDimension and, for floor-resting models, y = groundOffset * scale. Record a deliberate offset for models that should not rest on the floor.
  3. Confirm facing once in the running app and record yaw.

Nest three levels so a root position means the same for every asset: the semantic root, a yaw wrapper, then the visual child with the scale and local position (-center.x * scale, y, -center.z * scale). Apply calibration once, never per instance.

Place the root on a measured support point, such as a mount point or raycast, not a global AABB maximum. Skinned bounds are bind-pose estimates; confirm contact in animated poses.

Read only the reference matching the code: direct Engine, React, or Web Components.

Comments

Markdown supported