Build your first app

From a plain-language description to a deployed full-stack app, in the IDE or from the command line.

In the IDE

  1. Sign in at molecule.dev and start a new project (the free tier includes real sandbox time).
  2. Describe the app in plain language: "a booking site for a surf school with classes, instructors, and online payment".
  3. Synthase, the built-in AI developer, asks clarifying questions, proposes a plan, then scaffolds the project from real @molecule/* packages.
  4. A live preview opens as soon as the app boots, and every subsequent change hot-reloads it.
  5. Iterate by chatting: add features, restyle screens, change data models. The project stays type-checked and linted as it grows.
  6. Press Deploy. The same pipeline the CLI uses builds, ships, and hands you a production URL.
Tip

Prefer starting from something real? Pick one of the 100+ templates: a complete app to shape with AI.

From the command line

The same result from the terminal: mlcl scaffolds without an account until you deploy. Install once with npm install -g mlcl, or prefix commands with npx for one-offs:

mlcl create my-app --template blog --no-interactive
cd my-app
# put credentials in api/.env and app/.env (local defaults work for Postgres/SQLite)
npm run dev

mlcl create can also compose your own stack: pick a project type (api, app, full-stack, static, status-page, status-page-edge) and name the packages; the scaffold resolves default providers, transitive companions, and locale bonds for you.

mlcl create my-app --type full-stack --framework react \
  --packages @molecule/api-database-postgresql,@molecule/app-auth \
  --no-interactive

What you get

  • An Express API with routes, handlers, and database migrations.
  • A React app with real screens, auth, and i18n.
  • Tests and e2e specs that already run, plus CI security workflows.
  • .env templates for every secret the app expects.
  • An AGENTS.md that teaches any AI coding agent the project's conventions.

The full anatomy is on Project structure.

Deploy it

mlcl login                      # device-code: browser → Approve
mlcl deploy --project <id> --follow

In the IDE this is the Deploy button; it is the same pipeline. Custom domains, logs, and health checks are covered in Deploying your app.

Note

The code is yours from the first scaffold: the packages are Apache-2.0 and your app is under your own license. Licensing has the details.