# Lamina

> Design how it works. Transactional product-knowledge graph for AI coding agents. Invoke /lamina-init, /lamina-design, and /lamina-verify to publish intended behavior, source observations, and isolated Persona Mission evidence.

Lamina is an Apache-2.0 transactional product graph. Ladybug stores canonical Resources, Statements, versions, sessions, and Runs; CocoIndex incrementally supplies explicit source Observations through graphd.

Positioning: Product-design skill for developers who build with AI — know what to build, iterate faster.

Who it's for: Developers using AI coding agents (Cursor, Claude Code, Codex, Gemini, Pi) — not designers or PMs.

The loop: /lamina-init → /lamina-design → implement → /lamina-verify → fix. Run slash commands explicitly in agent chat — Lamina does not activate passively after install. Slash commands produce `.lamina/` artifacts; your coding agent implements app source; Lamina verifies the live build.

Lamina (lamina.dev) is an open-source product-design skill for AI coding agents — not uselamina.ai (creative API for media generation).

Key facts:

- Not a UI library — never writes app source or picks your component library
- Not an app builder — does not generate stack-locked apps like v0 or Lovable
- Not a pixel skill — Impeccable and frontend-design polish how it looks; Lamina designs how it works
- Open-source Apache-2.0 — https://github.com/aryaniyaps/lamina
- Install: `curl -fsSL https://github.com/aryaniyaps/lamina/releases/latest/download/install.sh | sh && npx skills add aryaniyaps/lamina --skill '*' -a <active-agent> -y`
- Commands: /lamina-init, /lamina-design, /lamina-verify, /lamina (optional router)
- Atomic, versioned product behavior with semantic branch diffs
- Independent Missions for every active Persona
- Epistemically separated intent, observations, simulation, and runtime evidence
- Fits any stack — framework, database, language, UI library, and AI coding tool
- Creator: Aryan Iyappan (https://aryaniyappan.com)
- Contact: https://github.com/aryaniyaps/lamina/issues

## Product

- [Homepage](https://lamina.dev): Product overview, positioning, and install command
- [What is Lamina?](https://lamina.dev/docs/concepts/what-is-lamina): Behavior vs pixels, comparisons, core principles
- [HavenStay demo](https://lamina.dev/docs/guides/demo-havenstay): Side-by-side build with and without Lamina
- [Quickstart](https://lamina.dev/docs/getting-started/quickstart): Run the full init → design → implement → verify loop
- [Installation](https://lamina.dev/docs/getting-started/installation): Install in Cursor, Claude Code, Codex, Gemini, or Pi

## Learn

- [Learn hub](https://lamina.dev/learn): Product behavior, transactional graph lifecycle, independent Persona Missions, FAQs, and guide index
- [What is headless product design?](https://lamina.dev/learn/what-is-headless-product-design): Category definition: behavior contracts for AI coding agents, vs UI skills, headless CMS, and app builders
- [Compare Lamina to your workflow](https://lamina.dev/learn/compare): Lamina's transactional product graph plus agent workflow vs UI skills, app builders, plan mode, spec tools, and more
- [Lamina vs Impeccable](https://lamina.dev/learn/compare/lamina-vs-impeccable): Impeccable polishes how your app looks. Lamina specs how it works: states, flows, permissions. Compare layers for AI agent workflows.
- [Lamina vs Google Stitch](https://lamina.dev/learn/compare/lamina-vs-stitch): Google Stitch generates high-fidelity UI from prompts. Lamina specs product behavior and verifies the live build.
- [Lamina vs shadcn/ui skill](https://lamina.dev/learn/compare/lamina-vs-shadcn-skill): The shadcn/ui agent skill makes component code correct. Lamina specs product behavior. Compare UI layer vs behavior layer.
- [Lamina vs UI design skills](https://lamina.dev/learn/compare/lamina-vs-ui-skills): UI design skills polish pixels. Lamina specs product behavior design behavior. Compare the UI skill category for AI agents.
- [Lamina vs Figma to code](https://lamina.dev/learn/compare/lamina-vs-figma-to-code): Figma to code translates screens into UI. Lamina specs product behavior before implementation.
- [Lamina vs v0](https://lamina.dev/learn/compare/lamina-vs-v0): v0 generates React UI from prompts. Lamina specs product behavior design behavior in the local product graph and verifies live builds. Compare generation vs contracts.
- [Lamina vs Magic Patterns](https://lamina.dev/learn/compare/lamina-vs-magic-patterns): Magic Patterns generates design-system UI. Lamina specs product behavior design behavior and verifies live builds. Compare UI gen vs contracts.
- [Lamina vs app builders](https://lamina.dev/learn/compare/lamina-vs-app-builders): v0, Lovable, and Bolt generate apps. Lamina specs product behavior design behavior in your repo with validated graph knowledge and verification.
- [Lamina vs plan mode](https://lamina.dev/learn/compare/lamina-vs-plan-mode): Plan mode helps you think. Lamina produces validated graph knowledge and verifies the live build. Compare ephemeral planning vs durable specs.
- [Lamina vs spec tools](https://lamina.dev/learn/compare/lamina-vs-spec-tools): Spec tools structure engineering work. Lamina structures product behavior before the engineering spec.
- [Lamina vs product-design skill](https://lamina.dev/learn/compare/lamina-vs-product-design-skill): The mblode product-design skill handles interaction judgment. Lamina produces validated graph knowledge and verifies live builds.
- [Lamina vs E2E tests](https://lamina.dev/learn/compare/lamina-vs-e2e-tests): /lamina-verify walks validated graph knowledge. E2E tests assert scenarios. Compare independent Persona Missions vs Playwright.
- [Lamina vs headless CMS](https://lamina.dev/learn/compare/lamina-vs-headless-cms): Headless CMS separates content from presentation. product behavior design separates behavior from app source. Not the same thing.
- [Lamina vs uselamina.ai](https://lamina.dev/learn/compare/lamina-vs-uselamina-ai): lamina.dev is a product-design skill for AI agents. uselamina.ai is a separate media API. Not affiliated.
- [Lamina vs roll your own](https://lamina.dev/learn/compare/lamina-vs-diy): You can spec behavior with plan mode, XState, or custom SKILL.md. Lamina packages the design → verify loop.
- [What is Lamina?](https://lamina.dev/docs/concepts/what-is-lamina): Product overview: CLI, portable skills, transactional graph, GraphVersion projections, and independent Persona Missions.
- [HavenStay demo walkthrough](https://lamina.dev/docs/guides/demo-havenstay): Legacy pre-graph proof: the same hotel booking app built twice, with Lamina's earlier product-design workflow and without.
- [Pair Lamina with your stack](https://lamina.dev/docs/guides/pair-with): Compose with UI skills, plan mode, and engineering spec tools without overlap.

## Benchmarks

- [Product Bench](https://lamina.dev/benchmarks): Initial public three-arm benchmark release across four greenfield thin vertical slices. Lamina mean Product behavior score 0.689, +0.104 over direct, +0.130 over plan-first, and the highest task-cell median on 4/4 tasks.
- [Benchmark documentation](https://lamina.dev/docs/advanced/lamina-bench): Product Bench protocol, task corpus, matched-arm design, limitations, and release lifecycle


## Licensing

- [Apache-2.0 source license](https://github.com/aryaniyaps/lamina/blob/main/LICENSE): Lamina CLI, skills, and source
- [RSL policy](https://lamina.dev/rsl.txt): Permitted AI use, attribution, and content update cadence for lamina.dev

## Documentation

- [index](https://lamina.dev/docs)
- [getting-started/first-run](https://lamina.dev/docs/getting-started/first-run)
- [Installation](https://lamina.dev/docs/getting-started/installation): Install the Lamina CLI and skills independently
- [getting-started/quickstart](https://lamina.dev/docs/getting-started/quickstart)
- [getting-started/troubleshooting](https://lamina.dev/docs/getting-started/troubleshooting)
- [concepts/brownfield-and-greenfield](https://lamina.dev/docs/concepts/brownfield-and-greenfield)
- [concepts/contracts-and-runs](https://lamina.dev/docs/concepts/contracts-and-runs)
- [concepts/personas](https://lamina.dev/docs/concepts/personas)
- [concepts/the-loop](https://lamina.dev/docs/concepts/the-loop)
- [concepts/what-is-lamina](https://lamina.dev/docs/concepts/what-is-lamina)
- [commands/index](https://lamina.dev/docs/commands/index)
- [commands/lamina](https://lamina.dev/docs/commands/lamina)
- [commands/lamina-design](https://lamina.dev/docs/commands/lamina-design)
- [/lamina-init](https://lamina.dev/docs/commands/lamina-init): Establish or update the domain charter and actor cast
- [commands/lamina-verify](https://lamina.dev/docs/commands/lamina-verify)
- [guides/demo-havenstay](https://lamina.dev/docs/guides/demo-havenstay)
- [guides/pair-with](https://lamina.dev/docs/guides/pair-with)
- [guides/pair-with-spec-kit](https://lamina.dev/docs/guides/pair-with-spec-kit)
- [Pair with UI skills](https://lamina.dev/docs/guides/pair-with-ui-skills): Lamina for behavior, UI skills for pixels — recommended workflow
- [guides/working-with-findings](https://lamina.dev/docs/guides/working-with-findings)
- [reference/artifacts](https://lamina.dev/docs/reference/artifacts)
- [reference/global-artifacts](https://lamina.dev/docs/reference/global-artifacts)
- [reference/glossary](https://lamina.dev/docs/reference/glossary)
- [reference/hybrid-retrieval](https://lamina.dev/docs/reference/hybrid-retrieval)
- [Skill map](https://lamina.dev/docs/reference/skill-map): The 10 public Lamina workflow and product capability skills
- [reference/transactional-graph](https://lamina.dev/docs/reference/transactional-graph)
- [reference/transactional-plan-acceptance](https://lamina.dev/docs/reference/transactional-plan-acceptance)
- [reference/validation](https://lamina.dev/docs/reference/validation)
- [LaminaBench](https://lamina.dev/docs/advanced/lamina-bench): Harbor v4 behavior benchmark that proves Lamina works under matched arms
- [advanced/orchestrator](https://lamina.dev/docs/advanced/orchestrator)
- [Crash-safe runner](https://lamina.dev/docs/advanced/safe-runner): Mandatory process-tree safety boundary for resource-intensive Lamina development and qualification

## Optional

- [Full docs dump](https://lamina.dev/llms-full.txt): Complete markdown export of all docs — skip if per-page URLs suffice
- [GitHub](https://github.com/aryaniyaps/lamina): Source, issues, and contributions
- [Creator](https://aryaniyappan.com): Aryan Iyappan