Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/Edupets-Studio/Edu-pets/llms.txt

Use this file to discover all available pages before exploring further.

EduPets ships with a vercel.json configuration file that tells Vercel exactly how to build and route the application. Because the app has no server-side database and no secret environment variables, a first deployment is literally a single command after you have the Vercel CLI installed. The sections below walk through every aspect of the deployment, from how the config works to ongoing GitHub-based continuous deployment.

​
How it works

Vercel uses the vercel.json file at the project root to determine how to build each part of the application and how to route incoming requests. EduPets defines two builds and two routes:
  • @vercel/python build for main.py — Vercel’s Python runtime wraps the FastAPI ASGI app in a serverless function. Every non-static request is handled by this function.
  • @vercel/static build for static/** — all files inside the static/ directory are uploaded to Vercel’s CDN and served directly without hitting the Python function.
  • Static route /static/(.*) — requests that match this pattern are forwarded straight to the CDN-hosted static asset.
  • Catch-all route /(.*) — every other request is forwarded to main.py, where FastAPI’s PAGES and LEGACY_PAGES dictionaries map URL paths to the correct Jinja2 template.
{
  "version": 2,
  "builds": [
    {
      "src": "main.py",
      "use": "@vercel/python"
    },
    {
      "src": "static/**",
      "use": "@vercel/static"
    }
  ],
  "routes": [
    {
      "src": "/static/(.*)",
      "dest": "/static/$1"
    },
    {
      "src": "/(.*)",
      "dest": "main.py"
    }
  ]
}

​
Deploy with the Vercel CLI

1

Install the Vercel CLI

Install the Vercel CLI globally via npm. You need Node.js 18 or later.
npm install -g vercel
2

Log in to Vercel

Authenticate the CLI with your Vercel account. This opens a browser window for OAuth.
vercel login
3

Deploy to a preview URL

Run vercel from the project root. The CLI will ask you to confirm the project name, the linked Vercel team (if any), and whether to override any settings. Accept the defaults to deploy immediately.
vercel
Once the build completes, the CLI prints a unique preview URL such as https://edu-pets-abc123.vercel.app where you can test the deployment before promoting it to production.
4

Promote to production

When you are happy with the preview, deploy to your production domain with the --prod flag.
vercel --prod
This assigns the deployment to your project’s production URL (e.g. https://edu-pets.vercel.app or any custom domain you have configured).

​
Deploy via GitHub

For teams or ongoing projects, the easiest workflow is to connect the repository to Vercel through the dashboard:
  1. Go to vercel.com/new and click Import Git Repository.
  2. Select the Edupets-Studio/Edu-pets repository (grant Vercel access to it if prompted).
  3. Vercel detects vercel.json automatically — no framework preset needs to be selected.
  4. Click Deploy.
From that point on, every push to the default branch triggers a new production deployment, and every pull request gets its own preview URL. No manual CLI commands are needed.

​
Static assets

All files under the static/ directory — stylesheets, JavaScript files, images, audio, and fonts — are handled by the @vercel/static builder and served from Vercel’s global CDN edge network. Requests matching /static/** are routed directly to the CDN by the first route in vercel.json and never reach the Python serverless function, keeping response times fast and function invocation costs low. If you add new assets to the static/ directory, they are picked up automatically on the next deployment — no configuration changes are required.

​
Health check

EduPets exposes a lightweight health endpoint that returns a JSON response: 
GET /health

{"status": "ok"}
This route is defined directly in main.py as a dedicated FastAPI endpoint (separate from the template-rendering routes) and is useful for uptime monitors, load-balancer health checks, or CI smoke tests after a deployment. 
@app.get("/health")
async def health() -> dict[str, str]:
    return {"status": "ok"}
EduPets uses no server-side database — all game state is stored in each user’s browser localStorage. This means no database credentials, no connection strings, and no secret environment variables are needed for a basic deployment. You can deploy the project to Vercel without setting a single environment variable and the app will work in full.

Build docs developers (and LLMs) love

Get started for freeTalk to us