Quickstart

This walkthrough completes one coherent cycle: init → design → implement → verify → fix. The example uses a single bounded flow — guest checkout with payment hold — not an entire marketplace.

Canonical sequence

SHELL
npx skills install aryaniyaps/lamina

AGENT CHAT — start a fresh session in your project
/lamina-init <your product domain and primary users>
/lamina-design <one feature or flow>

ORDINARY CODING MODE
Implement from .lamina/runs/<run_id>/implement.md
Start the product

AGENT CHAT
/lamina-verify <one feature or flow> at <base_url>

ORDINARY CODING MODE
Apply fix.md when findings remain

AGENT CHAT
/lamina-verify <one feature or flow> at <base_url> again

Prerequisites

• Lamina installed in your AI coding agent (Installation)
• A project directory open in your agent

1. Install (shell)

npx skills install aryaniyaps/lamina

Open your project root and start a fresh agent session so the skill loads.

2. Initialize (agent chat)

Run init directly — do not treat /lamina as a required setup step.

/lamina-init Boutique hotel booking marketplace in the United States

If Lamina asks consequential business questions, answer them. Reversible unknowns may remain labeled as assumptions.

Success state: init complete when Lamina produces and validates both project-level artifacts:

.lamina/
├── business-context.md
└── personas.json

Run init once per project or domain. Use /lamina-init update only when the business use case, market, scope, or actor cast materially changes.

Next action: run one bounded design command.

3. Design (agent chat)

Design one feature or flow, not an entire platform:

/lamina-design Guest checkout with inventory hold, payment, and cancellation

If the request is too vague, Lamina asks for clarification and does not present the flow as build-ready. If init is missing or invalid, Lamina stops and routes you back to /lamina-init — it never silently initializes on your behalf.

Success state: ready_to_build with three run artifacts:

.lamina/runs/<run_id>/
├── run.json          # status: ready_to_build
├── run.md            # readable contract rendering
└── implement.md      # implementation handoff

report.md is produced by verify, not design.

Next action: switch to ordinary coding mode and implement from implement.md. A fresh implementation session is optional; the mode change is mandatory — application edits never happen inside /lamina-design.

Implement from .lamina/runs/<run_id>/implement.md

Use any framework, database, and UI library. Lamina is stack-agnostic.

4. Run the product (ordinary coding mode)

Start your dev server and confirm the URL works in a browser before verify:

npm run dev

You need a running, browser-reachable product at a known base_url (for example http://localhost:3000).

5. Verify (agent chat)

Run verify with the bounded flow and live URL — verify requires both:

/lamina-verify guest checkout at http://localhost:3000

Lamina compares the live product against the contract and produces:

.lamina/runs/<run_id>/
├── run.json          # findings[] populated, status updated
├── report.md         # verification narrative
├── fix.md            # fix brief (written even when no product fixes remain)
└── walkthrough/      # optional live-capture evidence

6. Fix and re-verify

Each open finding routes to one branch:

Ordinary coding mode — product fixes:

Apply the product fixes in .lamina/runs/<run_id>/fix.md

Restart or refresh the product as needed, then re-verify in agent chat:

/lamina-verify guest checkout at http://localhost:3000

Exit condition: the cycle is complete when there are no open product or contract findings — not merely when artifacts were created. fix_target: ops observations remain in report.md and do not block this exit.

What you should have

After one full cycle:

Not sure which command to run? /lamina <what you are trying to do> is an optional router — not a required setup step.

Next steps

• First run — completion signals, pauses, and retries
• Demo: HavenStay — explore a full reference implementation
• Contracts and runs — artifact model deep dive