Lamina task analysis
Headless Product Design skill for AI coding agents | Design how it works, verify what you ship.
npx -y skills add aryaniyaps/lamina --skill lamina-task-analysisAssembled 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
Operations actors perform — map tasks to workflows and navigation. Use when workflows[] or IA don't match actor goals.
SKILL.md
2.0 KB, 344 tokens by cl100k_base, as published. Nobody here has run it
Task Analysis (agent-native)
Decompose actor goals into operations — verbs on domain entities — that become workflows[] steps and navigation structure.
Procedure
- From primary actor goals (
personas.json), list operations: "download ticket", "assign venue". - Group operations into workflows by outcome, not by backend module.
- Rank deal-breaker operations — if blocked, actor abandons product.
- Map working-set (daily) operations to shortest nav path in screen specs.
- Rare operations → progressive disclosure (
platform-posture).
Contract encoding
workflows[].steps= ordered operations; userequires+dependencies[]for prerequisitestask-analysisprose optional inreport.mdfor complex domains- Navigation labels = operation vocabulary, not org chart
Design checklists
- Nav structured by tasks, not implementation modules.
- Index/wayfinding separated from work area on sovereign apps.
- Key paths ≤ few steps for working-set tasks.
- Edge tasks reachable but not prominent.
- Each operation links to permission in
actors.
Verify checks
- Actor walk: deal-breaker operations completable on live product.
- Measure steps vs contract workflow length.
Anti-patterns
- Org-chart or file-system navigation.
- Deep hierarchy for daily tasks.
- Operations in contract that no actor has permission to perform.