Getting Started

Quick Start

This guide takes you from a fresh install to a running, exportable backend — one API endpoint, one database query, in about five minutes. It's meant to give you the full loop once, end to end; for the deeper version of each step, see Your First Backend.

1. Sign in and create a workspace

Fimaflow runs in the browser — no desktop app to install (that's planned, not shipped yet). Sign in at fimaflow.com and a default workspace is created for you automatically.

Prefer a terminal? The CLI covers everything past the visual builder itself — triggering runs, tailing logs, and exporting a backend without a browser open:

bash
❯ curl -fsSL https://raw.githubusercontent.com/Nathan-Josue/Fimaflow-cli/main/scripts/install.sh | sh
❯ fimaflow login

2. Build your first workflow

Open the Visual Builder and create a new workflow. Drag an API Endpoint node onto the canvas and set:

API Endpoint
endpointMethod: GET
endpointPath:   /hello

Connect it to a Response node, and set its body:

JSON
{
  "message": "Hello from Fimaflow!"
}

Click Run. Fimaflow executes the workflow immediately and shows each step live in the built-in debugger:

Execution
1. API Endpoint   ✓  received GET /hello
2. Response       ✓  200 { "message": "Hello from Fimaflow!" }

Run finished in 6ms — 2/2 nodes succeeded

Two nodes, no server process to start by hand — the workflow is already listening. See Visual Builder Overview for how the canvas works in general.

3. Add your first database query

Real endpoints usually touch a database. Insert a Database node between the API Endpoint and the Response, set dbMode to local for an instant, zero-config database — no connection string to set up for this first run — then add a SQL node right after it:

SQL node
sqlMethod: SELECT
table:     users
where:     { id: 1 }

Run the workflow again. The execution panel now shows a third step, with the exact rows the query returned as that node's output — click it to inspect the input, output and timing of every node in the chain, the same way you inspected the first two. See Database for connecting a real remote database once you're past prototyping.

4. Export your backend

Once the workflow behaves the way you expect, turn it into real source code. Pick a framework and export from the builder, or from a terminal:

bash
❯ fimaflow export <workflow-id> --target hono --out ./backend

Supported targets are Hono, Fastify, Express, Elysia and FastAPI — the generated project is a normal, readable codebase with no runtime dependency on Fimaflow. See Framework Export for exactly what gets generated, or Deployment Overview if you'd rather skip exporting and deploy directly from Fimaflow.

5. Continue building

The same loop — add a node, run, inspect, export — is how every larger backend gets built here, not just this two-node example. From here, the next things worth adding are usually authentication on your endpoints, environment variables for a real database connection, and AI nodes if your backend needs to call a model.

Next steps