$npx -y skills add rawwerks/VibeCAD --skill render-glbRender GLB 3D models to PNG images for visual verification. Use when you need to SEE your generated 3D model, verify geometry, check visual correctness, or show results to users. Triggers on: render GLB, preview model, visualize 3D, see my model, check geometry, visual verificati
| 1 | # Render GLB to Image |
| 2 | |
| 3 | Render 3D GLB files to PNG images so agents can **visually verify their work**. |
| 4 | |
| 5 | ## Zero-Setup with bunx |
| 6 | |
| 7 | ```bash |
| 8 | bunx render-glb model.glb output.png |
| 9 | ``` |
| 10 | |
| 11 | No installation required. First run downloads the tool, subsequent runs are instant. |
| 12 | |
| 13 | ## Why This Matters |
| 14 | |
| 15 | Agents can generate 3D models but typically can't see them. This creates a **visual verification loop**: |
| 16 | |
| 17 | 1. Generate model → export GLB |
| 18 | 2. Render GLB → PNG |
| 19 | 3. Read PNG → see what you built |
| 20 | 4. Iterate if needed |
| 21 | |
| 22 | ## Basic Usage |
| 23 | |
| 24 | ```bash |
| 25 | # Render with defaults (good for most cases) |
| 26 | bunx render-glb model.glb preview.png |
| 27 | |
| 28 | # Then read the image to see your work |
| 29 | # (use Read tool on preview.png) |
| 30 | ``` |
| 31 | |
| 32 | ## Options |
| 33 | |
| 34 | ```bash |
| 35 | bunx render-glb <input.glb> <output.png> [options] |
| 36 | |
| 37 | Options: |
| 38 | --width <n> Image width in pixels (default: 800) |
| 39 | --height <n> Image height in pixels (default: 600) |
| 40 | --background <hex> Background color (default: #808080) |
| 41 | ``` |
| 42 | |
| 43 | ## Visual Verification Workflow |
| 44 | |
| 45 | When building 3D models, use this pattern: |
| 46 | |
| 47 | ```python |
| 48 | # 1. Build your model |
| 49 | from build123d import Box, Sphere, export_gltf |
| 50 | model = Box(20, 20, 20) + Sphere(radius=15) |
| 51 | export_gltf(model, "./model.glb", binary=True) |
| 52 | ``` |
| 53 | |
| 54 | ```bash |
| 55 | # 2. Render to image |
| 56 | bunx render-glb model.glb preview.png |
| 57 | ``` |
| 58 | |
| 59 | ``` |
| 60 | # 3. Read the image (using Read tool) |
| 61 | # Now you can SEE what you built and verify it's correct |
| 62 | ``` |
| 63 | |
| 64 | ## Integration with CAD Skills |
| 65 | |
| 66 | This skill complements: |
| 67 | - **build123d** - Python CAD modeling |
| 68 | - Any tool that exports GLB format |
| 69 | |
| 70 | Typical workflow: |
| 71 | 1. Use CAD skill to generate geometry |
| 72 | 2. Export to GLB |
| 73 | 3. Use render-glb to verify visually |
| 74 | 4. Iterate if model isn't correct |
| 75 | |
| 76 | ## Links |
| 77 | |
| 78 | - npm: https://www.npmjs.com/package/render-glb |
| 79 | - GitHub: https://github.com/rawwerks/render-glb |