create-vision-app

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

Usage

Run with npx (no installation needed):

``bash
npx create-vision-app
`

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

create-vision-app 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

`bash
$ npx create-vision-app

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

Create a computer vision endpoint in seconds

┌ create-vision-app
│
◇ 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

`bash


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