dgleich

This skill should be used when adding a new slide element type (text/image/arrow/demo/demo-piece/notebook/video/cover) OR adding/changing a property on an existing type (e.g. arrow `heads`/`opacity`, image `borderRadius`). Eigendeck has SEVEN independent render/output paths plus data, store, inspector, and doc concerns — a change applied to only the editor will silently render wrong (or vanish) in present mode, HTML export (app GUI AND headless CLI — wired separately), PDF/print, thumbnails, or the link overlay. This is the checklist of everything to update and how to verify WYSIWYG across all paths.

Cut a new Eigendeck release (CalVer YY.M.D). Use when the user says to cut/tag/ship a release, bump the version, or "do a release". Covers refreshing mathjax-fonts, updating the open-source/font credits, bumping the version everywhere, building/testing, tagging to trigger the GitHub build, and verifying the draft.

Update Eigendeck's bundled MathJax math fonts to the latest builds from the dgleich/mathjax-fonts repo. Use when the user says to update the fonts / font packs, points at a new mathjax-fonts commit, or wants to add a new font. mathjax-fonts now ships PREBUILT -nosre bundles, so the eigendeck side is just pull (or clone at the pinned commit) + npm run setup — no local rebuild.

This skill should be used when the user asks to commit modified or new `.eigendeck` files (examples/ or test-presentations/) to git. It enforces two prerequisites before committing - history-state is stripped (no temporal rows or asset_cache pollution from a test session) and the file loads cleanly under the current schema.

Set up and run headless end-to-end (e2e) tests for the Eigendeck Tauri desktop app on Linux — driving the REAL built app via tauri-driver + WebKitWebDriver + xvfb. Use when running the e2e/ scenarios, writing a new e2e test for a feature, debugging a frontend↔Rust boundary bug, or re-provisioning the e2e toolchain after a container reset.

Build, inspect, and edit .eigendeck decks from the command line — especially BUILDING a deck programmatically from JSON, including decks with embedded demo/image assets (the store-asset → export --with-assets → expand JSON → import json round-trip). Use when you need to generate a test deck, a per-font/per-theme matrix, a fixture, or bulk-edit slides/elements without the GUI.

Tag LLM-generated vs. human-authored text using the tilde-fence provenance syntax (tilde "llm" fences, :::llm colon fences, llm/human HTML elements, inline tilde spans). Use this skill whenever the user asks to mark, tag, label, annotate, or attribute AI/LLM/machine-generated text; asks for "provenance" markup; asks Claude to identify which parts of a document it wrote; mentions the llm-provenance spec, llm tags, or tilde fences; or asks Claude to append/insert generated content into a document while keeping authorship distinguishable. Also use when editing a document that already contains llm/human provenance markup, so the markup is preserved and updated correctly.