What is Lamina?
Lamina is a headless product-design skill for AI coding agents. It designs how your app works — states, roles, flows, invariants — produces a contract your agent implements, then verifies the live build with parallel persona walks. It never writes app source.
Not uselamina.ai. Lamina (lamina.dev ) is an open-source product-design skill for AI coding agents. It is not affiliated with uselamina.ai , a separate creative API for media generation.
Quick path
/lamina-init <domain>→ project context- Lamina designs product behavior →
implement.md - Your coding agent implements → app source
- Lamina verifies the live app →
fix.md - Your agent fixes → re-verify
Behavior over pixels
| UI skills | Lamina | |
|---|---|---|
| Focus | How it looks | How it works |
| Output | Design tokens, components, layouts | States, roles, flows, invariants |
| When | During or after UI work | Before build, verified after |
| Writes app code? | Sometimes | Never |
UI skills like Impeccable, UI UX Pro Max, and frontend-design polish interfaces. Lamina designs states, roles, recovery paths, and permission matrices — then verifies them on your live app.
How Lamina compares
UI skills
They polish how it looks. Lamina designs how it works — actors, flows, empty/error/loading states, invariants. Pair any UI skill; Lamina stays out of pixels. See Pair with UI skills.
Design craft skills
BMAD, ai-ux-skills, and design-skills teach design judgment — heuristics, critique, accessibility. Lamina runs a workflow: slash commands → run.json / implement.md → live-app verify.
Generic agent prompting
Fine for happy paths. Weak on permission matrices, stale states, and mid-flow failures. Lamina structures behavior before build and walks the live app after — but only when you invoke it with slash commands such as /lamina-design or route through /lamina <request>. It does not activate passively after install.
Spec tools
Spec Kit, Kiro, and spec-driven dev structure engineering work. Lamina structures product behavior. Run /lamina-init, then /lamina-design, then feed implement.md into Spec Kit. See Pair with Spec Kit.
App builders
v0, Lovable, and Bolt generate apps — often stack-locked. Lamina does not generate code or pick your framework; it fits your repo, agent, and UI stack.
Core principles
- Design is how it works — states, edges, and flows before your agent writes app code
- Your agent writes app source — Lamina never does
- Verify on the live app — persona walks read real UI, not mocks
- Any stack — web, mobile, desktop, CLI — pair any UI skill
Any stack
Lamina slots into whatever you already use. It is unopinionated on tech stack and AI skills.
| Any AI coding tool | Cursor, Claude Code, Codex, Gemini, Pi, etc. |
| Any framework | Next.js, Angular, Astro, Svelte, React Native, Flutter, FastAPI, Gin, Express, etc. |
| Any database | Postgres, MySQL, MongoDB, Cassandra, Redis, Neo4j, etc. |
| Any language | JavaScript, Python, Go, Rust, Elixir, PHP, C#, etc. |
| Any UI library | TailwindCSS, Chakra UI, shadcn, MUI, etc. |
| Any UI design skill | Impeccable, UI UX Pro Max, frontend-design, etc. |
| Any workflow skill | obra/superpowers, mattpocock/skills, everything-claude-code, etc. |
| Any interface | Websites, mobile apps, desktop, PWAs, chat bots, CLIs, etc. |
Direct mode
/lamina can answer focused product design questions without a full workflow — routing to a single capability skill for invariants, error handling, dependencies, and more. See /lamina command and Skill map.
Advanced: Lamina includes ~60 capability skills organized into systems thinking, UX/product expression, and agent simulation. The orchestrator routes design and verify sections to the right skills via audit profiles.
Next steps
- The loop — design → implement → verify cycle
- Quickstart — run your first cycle
- Demo: HavenStay — see Lamina in action