dotgui guides
Guides
The spec answers "what does this attribute do." These answer "how is .gui different from what I already know" and "how do I actually build a good file" — for people evaluating the format and for agents mid-task with a .gui file already open.
Comparisons
CSS styles a document that already exists somewhere else. .gui is the document — structure, layout, and appearance in one file, with no cascade to fight.
Read guide →HTML is a programming surface that happens to describe UI. .gui is a static, closed vocabulary that always terminates in the same picture, everywhere.
Read guide →A Figma file only exists inside Figma. A .gui file is plain text in a zip — readable, diffable, and versionable without opening any design tool.
Read guide →A picture is a dead end — an agent has to guess spacing, hierarchy, and color from pixels. A .gui file hands those values over directly, already structured.
Read guide →SVG made graphics text. .gui borrows that idea but is purpose-built for interface layout and semantics — things SVG was never designed to express.
Read guide →Best practices
gui setup teaches your own agent the format. Here’s the mental model and what to actually say.
Read guide →An agent that hasn’t been taught the format brings habits over from HTML and CSS that don’t belong in .gui. Here are the five that show up most.
Read guide →The single most common habit agents bring over from HTML/CSS — and the one-attribute fix.
Read guide →The four levels explained on /spec/quality, turned into concrete things to actually go fix.
Read guide →Three layout elements cover almost every screen. Here’s how to pick between them without guessing.
Read guide →The tells of an unedited AI-generated screen, and the structural (not aesthetic) fixes .gui makes available.
Read guide →