Deploying to Cloudflare Workers

Today you'll be building and deploying a Cloudflare Worker that uses Prisma to save every request to a MongoDB database for inspection later.

This guide will cover Prisma, TypeScript, MongoDB, Prisma Data Proxy and Cloudflare Workers from the ground up. Let's get started!

Early Access Feature
Please note that the Prisma Data Proxy is an early-stage product and should not be used in production environments.


  • Free MongoDB Atlas database with an accessible URL
    • e.g. mongodb+srv://<user>:<password>
  • Free Cloudflare Worker account.
  • Free GitHub account.
  • Node.js & NPM installed.
  • Git installed.

1. Set up your Application

Open your terminal and navigate to a location of your choice. Run the following commands to set up your application.

$mkdir prisma-mongodb-cloudflare
$cd prisma-mongodb-cloudflare
$npm init -y
$npm install -D prisma typescript @cloudflare/wrangler webpack

2. Set up Wrangler

Wrangler is the official Cloudflare Worker CLI. You'll use it to develop and deploy Cloudflare Workers.

$npx wrangler init

This will create a wrangler.toml file with some initial configuration.

Next, you'll want to authenticate the Wrangler CLI with your Cloudflare Worker account. To do this, run the following command in your terminal:

$npx wrangler login
$Allow Wrangler to open a page in your browser? [y/n] y

You can verify that you're logged in by running npx wrangler whoami.

3. Set up TypeScript

The Cloudflare Worker environment natively supports Javascript, Rust, C, and C++. To get TypeScript working on Cloudflare Workers, you need to compile TypeScript to JavaScript before deploying to a Worker.

To setup this up, create a tsconfig.json in your project root with the following content:

"compilerOptions": {
"outDir": "./dist",
"module": "commonjs",
"target": "esnext",
"lib": ["esnext"],
"alwaysStrict": true,
"strict": true,
"preserveConstEnums": true,
"moduleResolution": "node",
"sourceMap": true,
"esModuleInterop": true,
"types": ["@cloudflare/workers-types"]
"include": ["src"],
"exclude": ["node_modules", "dist"]

And then install @cloudflare/workers-types with npm:

$npm install -D @cloudflare/workers-types ts-loader@8.3.0

You need to pin ts-loader for the Wrangler CLI to work properly. See this issue for more details.

4. Set up Webpack

Wrangler has built-in webpack support that can be used to compile your code in development and before publishing on Cloudflare.

To configure webpack support, first create a webpack.config.js in the project root with the following code:

const path = require('path')
module.exports = {
entry: './src/index.ts',
output: {
filename: 'worker.js',
path: path.join(__dirname, 'dist'),
// Cloudflare Worker environment is similar to a webworker
target: 'webworker',
resolve: {
extensions: ['.ts', '.tsx', '.js'],
// Alias for resolving the Prisma Client properly
alias: {
'@prisma/client$': require.resolve('@prisma/client'),
mode: 'development',
// Wrangler doesn't like eval which devtools use in development.
devtool: 'none',
module: {
rules: [
// Compile Typescript code
test: /\.tsx?$/,
loader: 'ts-loader',
options: {
transpileOnly: true,

Next, add the following line to your wrangler.toml file to point Wrangler to use this webpack configuration:

name = "prisma-mongodb-cloudflare"
type = "webpack"
webpack_config = "webpack.config.js"

5. Set up Prisma

Now you're ready to add Prisma to the project.

$npx prisma init

This will create a Prisma Schema in prisma/schema.prisma and an .env file.

Inside prisma/schema.prisma, add the following schema:

generator client {
provider = "prisma-client-js"
previewFeatures = ["dataProxy"]
datasource db {
provider = "mongodb"
url = env("DATABASE_URL")
model Log {
id String @id @default(auto()) @map("_id") @db.ObjectId
level Level
message String
meta Json
enum Level {

This data model will be used to store incoming requests from your Worker.

6. Create Repository and Push to GitHub

To prepare for the steps ahead, let's create a private repository on GitHub.

Create a private repository

Next, initialize your repository, then push your changes up to GitHub.

$git init -b main
$git remote add origin<username>/prisma-mongodb-cloudflare
$git add .
$git commit -m "initial commit"
$git push -u origin main

You're ready to import your project into the Prisma Data Platform.

7. Importing your Project into the Prisma Data Platform

With Cloudflare Workers, you can't directly access your database because there is no TCP support. Fortunately, Prisma has your back with the Prisma Data Proxy.

To get started, sign up for a free Prisma Data Platform account.

Signup for Prisma Data Platform

Once you're in, click New Project, then Import a Project. Fill in the repository and project details, and then click Create Project.

Import a Project

Next, you'll connect the Prisma Data Platform to MongoDB Atlas database and set up the Prisma Data Proxy:

Connect your Database

Click Create Project to test the connection and set up the Data Proxy.

If you see, "The database needs to be empty to proceed", you can simply use a different database name. Using the screenshot above, that would be renaming logs to something else.

If all goes well, you'll be greeted with a new connection string that starts with prisma://. Copy this connection string to your clipboard.

Data Proxy Page

Then you can hop back into your code editor and paste the connection string into your .env file:


You're all setup and ready to generate a Prisma Client!

Please be aware that the Data Proxy is in Early Access and should not be used in production.

9. Generate a Prisma Client

Next you'll generate a Prisma Client that connects through the Data Proxy over HTTP.

$PRISMA_CLIENT_ENGINE_TYPE=dataproxy npx prisma generate

This client is optimized for edge environments like Cloudflare Workers.

Please be aware that the proxy-enabled Prisma Client is in Early Access and subject to change.

10. Develop the Cloudflare Worker function

You're now ready to create a Cloudflare Worker. Create a src/index.ts file with the following code:

import { PrismaClient } from '@prisma/client'
const prisma = new PrismaClient()
addEventListener('fetch', (event) => {
async function handleEvent(event: FetchEvent): Promise<Response> {
const { request } = event
// waitUntil method is used for sending logs, after response is sent
data: {
level: 'Info',
message: `${request.method} ${request.url}`,
meta: {
headers: JSON.stringify(request.headers),
return new Response(`request method: ${request.method}!`)

Run npx wrangler dev to see your worker in development:

👂 Listening on

Go ahead and open If all goes well, you should see:

request method: GET!

Refresh the page a couple times to verify that it's working. Now if you click on the Data Browser tab on your Prisma Cloud project, you should see Info logs written to your database.

Prisma Data Browser

It's working locally!

11. Publishing to Cloudflare Workers

You're now ready to deploy to Cloudflare Workers. Run the following command:

$npx wrangler publish

This will pack your application with webpack and upload to Cloudflare. With a bit of luck, you'll see the following:

✨ Built successfully, built project size is 94 KiB.
✨ Successfully published your script to

Visit your deployment URL and you'll again see:

request method: GET!

You're all set! You've successfully deployed a Cloudflare Worker written in TypeScript that uses Prisma to talk to your MongoDB database.

Give yourself a pat on the back, you deserve it!

Edit this page on GitHub