Lamina content design
Headless Product Design skill for AI coding agents | Design how it works, verify what you ship.
npx -y skills add aryaniyaps/lamina --skill lamina-content-designAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
What its author says it does
Copied from the file, not written here
Copy and labels in contracts — scan-first labels, error text, and empty-state messaging. Use when screen copy blocks actor walks.
SKILL.md
2.2 KB, as published. Nobody here has run it
Content Design (agent-native)
Specify user-visible text in surfaces[] — labels, headings, errors, empty states — so implementers and verify subagents share one source of truth.
Contract encoding
Per screen:
title,primary_actionlabelempty_statecopy whenscenariosincludecollection_emptyerror_messageskeyed by scenario id- Navigation labels consistent with
workflowsoperation names
Concise copy: every word earns its place; no happy talk on task screens.
Frameworks
- Scan-first: headings, bullets, one primary CTA visible without reading paragraphs.
- Satisficing: first visible option on primary path should be correct.
- Mindless choices: mutually exclusive nav labels — depth beats ambiguous breadth.
- Eliminate instructions: if copy explains the UI, fix structure (
discoverability).
Design checklists
- Page/screen title matches link that leads there.
- Error copy: what happened → why → what to do next.
- Labels use actor vocabulary from
entities[], not internal jargon. - Home/landing answers: what site, what can I do, why here, why trust — if applicable.
- Halve marketing copy on task screens.
Verify checks
- Actor walk: simulated actor finds primary action without re-reading labels.
- Walkthrough: link text ≠ page title mismatches → finding.
- a11y: link purpose clear out of context.
Anti-patterns
- Welcome paragraphs users skip.
- "Click here" or jargon labels.
- Instructions compensating for bad layout.
- Clever headings that fail actor-walk goal completion.