deploy

Create a computer vision endpoint in seconds with a beautiful CLI experience.

Usage

Run with npx (no installation needed):

npx deploy

The CLI will guide you through:

1. Defining what objects you want to detect
2. Configuring active learning
3. Deploying your endpoint
4. (Optional) Scaffolding a sample application

What it does

deploy creates and deploys a serverless computer vision endpoint powered by Roboflow. You'll get:

• A live endpoint URL ready to process images
• Active learning to improve your model over time (optional)
• Sample code in Python, Node.js, or as a web app (optional)

Example Output

$ npx deploy

 ╦  ╦┬┌─┐┬┌─┐┌┐┌  ╔═╗┌─┐┌─┐
 ╚╗╔╝│└─┐││ ││││  ╠═╣├─┘├─┘
  ╚╝ ┴└─┘┴└─┘┘└┘  ╩ ╩┴  ┴

  Create a computer vision endpoint in seconds

┌  deploy
│
◇  What objects are you looking for?
│  people, cars
│
◇  Enable active learning so your model improves as it sees more?
│  Yes
│
◆  Deployed successfully! 🎉
│
│  ✓ Your app is live at: https://serverless.roboflow.com/brad-dwyer/workflows/people-vision-app
│
│  Try it like this:
│    curl -X POST https://serverless.roboflow.com/... \\
│      -H "Content-Type: application/json" \\
│      -d '{"image_url": "https://example.com/image.jpg"}'
│
└  All done! Happy building! 🚀

Development

# Install dependencies
npm install

# Run locally
npm run dev

# Build for distribution
npm run build

Tech Stack

• @clack/prompts - Beautiful CLI prompts
• chalk - Terminal colors
• figlet - ASCII art
• gradient-string - Gradient text
• nanospinner - Loading animations

License

MIT