A practical, end-to-end walkthrough of the project-scoped blog-post-excerpt Claude skill that lives in this repository under .claude/skills/. The guide follows the chapter structure of Anthropic's Complete Guide to Building Skills for Claude and applies each principle to the working skill that automates Jekyll front-matter excerpt generation. Learn how progressive disclosure splits across SKILL.md, its body, and the references folder; how to craft a triggering description; how to inject an inline SVG illustration plus a TLDR into a single line of YAML; and how a final verification step keeps the front matter from breaking on every run.

This deep-dive shows how to run the Google DeepMind Gemma 4 E4B model — a 4.5B-effective dense network with Per-Layer Embeddings — on a Minisforum AI X1 Pro driven by the AMD Ryzen AI 9 HX 470, Radeon 890M iGPU and XDNA 2 NPU. It walks the verified vLLM Docker recipe on ROCm 7.2, decomposes the hybrid sliding-window plus global attention that makes a 128K context fit on a 16GB-class memory budget, and shows where MIGraphX can offload an ONNX sidecar to the NPU. The result is a layered guide from architecture math through tuning, quantisation, and benchmarking.