Review 3DGS implementation code for correctness, performance bugs, and best practices. Covers CUDA kernels, rendering pipeline, training loop, loss functions. Detects 108+ known bug patterns including MoE-GS, Bayesian complexity control, deformable aggregation, and PBR material disentanglement patterns. Use when: reviewing 3DGS/Gaussian Splatting CUDA code, debugging rendering artifacts, optimizing 3DGS training pipelines, checking loss function implementations, 代码审查/3DGS调试/性能优化.
$npx -y skills add jaccen/awesome-gaussian-skills --skill 3dgs-code-reviewerInstalls into the current project.
Run `npx skills use "https://github.com/jaccen/awesome-gaussian-skills" --skill "jaccen/awesome-gaussian-skills/3dgs-code-reviewer"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.
Use the skills in "https://github.com/jaccen/awesome-gaussian-skills" that are relevant to the current task. Run `npx skills add "https://github.com/jaccen/awesome-gaussian-skills"` and select the relevant skills, then follow their instructions.