executing-plans
executing-plans
Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available
hunyuan3d-2Route Hunyuan3D-2 repo-specific shape, texture, service, and backend workflows.
Independent GitHub snapshot · Automated classification · Original documentation governs
Use this repo skill when a task names Hunyuan3D-2, hy3dgen, Hunyuan3D-DiT, Hunyuan3D-Paint, Tencent Hunyuan image-to-3D/texture generation, the repository's FastAPI/Gradio/Blender integrations, or its CUDA texture extensions.
| User task | Read next |
|---|---|
| Generate a white/geometry mesh from one image, multiview images, mini/fast/turbo variants, FlashVDM, VAE, or shape export parameters | sub-skills/shape-generation/ |
| Texture a generated or existing mesh, use Hunyuan3D-Paint, clean/reduce meshes, diagnose rasterizer/renderer/UV failures | sub-skills/texture-and-mesh/ |
| Launch or call the FastAPI server, Gradio UI, Blender add-on, REST payloads, asynchronous status polling, service ports | sub-skills/services-and-integrations/ |
Install hy3dgen, choose PyTorch/CUDA, build custom_rasterizer/mesh_processor, plan model caches, verify backend readiness | sub-skills/environment-and-model-setup/ |
From a copied skill directory, these checks avoid model downloads:
python sub-skills/environment-and-model-setup/scripts/check_install.py --json
python sub-skills/shape-generation/scripts/generate_shape.py --preset base --image input.png --dry-run
python sub-skills/texture-and-mesh/scripts/texture_mesh.py --mesh mesh.glb --image input.png --dry-run
python sub-skills/services-and-integrations/scripts/request_api.py --image input.png --dry-run
Use real generation only after the environment and model-cache constraints are acceptable.