Skip to content
Floci Studio

Quick Start

Floci Studio is a local AWS emulator cockpit. You get a full-featured GUI, an AWS-compatible API endpoint on :4566, and an MCP server your AI agent can drive — all running on your machine, at zero cost.

Prerequisites

Section titled “Prerequisites”

That’s it. No AWS account. No Python. No Node.js required just to run.

1. Clone the repo

Section titled “1. Clone the repo”
Terminal window
git clone https://github.com/pmaojo/floci-studio
cd floci-studio

2. Start the stack

Section titled “2. Start the stack”
Terminal window
docker compose up

This starts three containers:

  • floci-engine — the AWS emulator, exposed on :4566
  • floci-sidecar — the FastAPI backend, exposed on :8000
  • floci-studio — the React UI, exposed on :3000

First run pulls images and takes ~60 seconds. Subsequent starts are under 5 seconds.

3. Open the console

Section titled “3. Open the console”

Navigate to http://localhost:3000 in your browser.

You’ll see the floci cockpit. The event stream at the bottom turns green when the emulator is reachable.

4. Create your first resource

Section titled “4. Create your first resource”

Click SQS in the sidebar, then Create Queue. Name it my-queue and click Create. Your queue is now live on the local AWS endpoint.

What’s next

Section titled “What’s next”