Skip to main content

​
What floo does

A cloud platform where everything runs from the CLI. Deploy apps, set environment variables, view logs, and roll back — all from your terminal. Every command supports --json so AI agents can use floo programmatically.

​
How it works

# 1. Install
curl -fsSL https://getfloo.com/install.sh | bash

# 2. Log in (opens browser)
floo auth login

# 3. Initialize your project
cd my-project
floo init my-app

# 4. Connect GitHub — installs the GitHub App, links repo, triggers first deploy
floo apps github connect owner/repo --app my-app
That’s it. Your app is live at https://my-app-abc123.us-central1.run.app. The CLI auto-detects your runtime (Node.js, Python, Go, static), creates floo.app.toml and floo.service.toml config files, and deploys from your connected GitHub repo.

​
After your app is live

Every push to your default branch auto-deploys. floo deploy is for manual redeploys when you want one. 
floo env set DATABASE_URL=postgres://... --app my-app  # env vars (encrypted at rest)
floo domains add myapp.com --app my-app                # custom domain with auto SSL
floo logs --app my-app                                 # tail logs
floo deploy rollback my-app DEPLOY_ID                  # roll back
See the guides for details on each feature.

​
Built for agents

Every command has a --json flag. JSON goes to stdout, human output to stderr. Errors include machine-readable codes and suggestions. 
floo deploy --json 2>/dev/null | jq '.data.deploy.url'

{
  "success": false,
  "error": {
    "code": "DEPLOY_FAILED",
    "message": "Build failed: missing package.json",
    "suggestion": "Add a package.json to your project root"
  }
}
Drop a skill file into your project and your AI agent can deploy, manage env vars, view logs, and roll back — all autonomously.

Claude Code Setup

Add the floo skill to your project

Cursor Setup

Add floo rules to .cursorrules