LaunchKit · 2026
Back to Skills

Other tools need perfect images.

3
820 downloads
by @nitishgargiitd

Setup & Installation

openclaw skills install @nitishgargiitd/3d-cog

Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:

npx clawhub install 3d-cog

What This Skill Does

CellCog converts text descriptions, rough sketches, or product photos into production-ready GLB 3D models. It handles the full pipeline internally: reasoning about the object, generating an optimized reference image, and performing the 3D conversion. Batch generation lets you produce multiple models from a single prompt.

Unlike image-to-3D tools that require a clean, well-composed reference photo, this skill accepts unpolished or purely textual input and generates the reference itself before converting to 3D.

When to use it

  • Generating a set of low-poly weapons for a mobile RPG
  • Converting product photos to rotatable 3D models for an e-commerce store
  • Turning a rough character sketch into a game-ready GLB asset
  • Creating a matched furniture collection for a VR room configurator
  • Producing tabletop miniature files for 3D printing from text descriptions

Example Workflow

Here's how your AI assistant might use this skill in practice.

INPUT

User asks: Create 3D models for 5 fantasy weapons — enchanted longsword, dwarven war hammer, elven bow, shadow dagger, holy mace. Low poly, game-ready, PBR materials, GLB format.

AGENT
  1. 1Parse the list and identify each weapon's visual characteristics
  2. 2Generate an optimized reference image for each weapon with correct composition and angle
  3. 3Convert each reference image to a textured 3D model
  4. 4Package each model as a GLB file with PBR materials embedded
  5. 5Return all 5 GLB files via daemon notification
OUTPUT

5 GLB files, each a low-poly game-ready weapon with PBR materials, delivered when processing completes