to-html

name

to-html

description

Render a plan, review, design, or any structured topic as a single skimmable HTML file in the current directory, built for visual review (diagrams, charts, stats, side-by-side, tables, callouts) instead of prose. Use whenever the user says "make an html for this", "show me this as a page", "render this as html", "visualize this", "build a review page", "build a dashboard", "/skill:to-html", or when the topic has flows, before/after, trade-offs, file-by-file changes, metrics, quantitative data, or risks that would be faster to grasp visually than to read in chat. Bias toward triggering whenever a wall of text in chat would be harder to skim than a page.

to-html

Produce a single skimmable HTML file at ./<slug>.html so the user can review a topic visually instead of scrolling chat prose.

The artifact is for the user, on their laptop, regenerated cheaply and often. Optimize for first-paint scannability, not document depth.

When to use / when NOT to use

Use when	Skip when
topic has flows, before/after, trade-offs, file-by-file changes, options, risks, metrics	answer fits in a few chat lines
user explicitly asks for an html / page / visual	topic is one paragraph of prose explanation
there are ≥3 distinct things to compare, list, or sequence	reply is a single bullet list
user wants to sit with a plan before approving	user wants conversation, not artifact

If you would otherwise produce a wall of prose with two headings, write the prose in chat instead and don't trigger this skill.

Output contract

• One file: ./<slug>.html in the current working directory. Overwrite if it exists — regen is cheap.