Draw Sebastian-Raschka-gallery-style TikZ architecture diagrams for any HuggingFace decoder-only LLM, with per-block parameter formulas and concrete numbers. Supports MHA, GQA, MLA, DeepSeek-V4-Flash (Hyper-Connections + Sparse Attention with learned indexer), dense and MoE FFNs (incl. hash routing), and MTP heads. Use when the user asks to visualize / diagram / illustrate a transformer or LLM architecture (DeepSeek, Qwen, Llama, Mistral, gpt-oss, etc.), wants a Raschka-style figure, or wants a TikZ/LaTeX rendering of an HF model.
$npx -y skills add yzlnew/infra-skills --skill hf-architecture-tikzInstalls into the current project.
Run `npx skills use "https://github.com/yzlnew/infra-skills" --skill "yzlnew/infra-skills/hf-architecture-tikz"` 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/yzlnew/infra-skills" that are relevant to the current task. Run `npx skills add "https://github.com/yzlnew/infra-skills"` and select the relevant skills, then follow their instructions.