← Cloudflare For Platforms / cloudflare-for-platforms / workers-for-platforms / configuration

Bindings

When you deploy User Workers through Workers for Platforms, you can attach bindings to give them access to resources like KV namespaces, D1 databases, R2 buckets, and more. This enables your end customers to build more powerful applications without you having to build the infrastructure components yourself.

With bindings, each User Worker can extend functionality to:

Store data with KV, R2, D1, or Durable Objects
Process work asynchronously with Queues and Workflows
Run containers with Containers (bound as a Durable Object)
Connect to private networks with VPC Services, VPC Networks, and Hyperdrive
Collect metrics with Analytics Engine

Resource isolation

Each User Worker can only access the bindings that are explicitly attached to it. For complete isolation, you can create and attach a unique resource (like a D1 database or KV namespace) to every User Worker.

Adding a KV Namespace to a User Worker

This example walks through how to create a KV namespace and attach it to a User Worker. The same process can be used to attach to other bindings.

1. Create a KV namespace

Create a KV namespace using the Cloudflare API.

2. Attach the KV namespace to the User Worker

Use the Upload User Worker API to attach the KV namespace binding to the Worker. You can do this when you're first uploading the Worker script or when updating an existing Worker.

Example API request

curl -X PUT \
  "https://api.cloudflare.com/client/v4/accounts/<account-id>/workers/dispatch/namespaces/<your-namespace>/scripts/<script-name>" \
  -H "Content-Type: multipart/form-data" \
  -H "Authorization: Bearer <api-token>" \
  -F 'metadata={
    "main_module": "worker.js",
    "bindings": [
      {
        "type": "kv_namespace",
        "name": "USER_KV",
        "namespace_id": "<your-namespace-id>"
      }
    ]
  }' \
  -F 'worker.js=@/path/to/worker.js'

Now, the User Worker can access the USER_KV binding through the env argument using env.USER_KV.get(), env.USER_KV.put(), and other KV methods.

Note: If you plan to add new bindings to the Worker, use the keep_bindings parameter to ensure existing bindings are preserved while adding new ones.

curl -X PUT \
  "https://api.cloudflare.com/client/v4/accounts/<account-id>/workers/dispatch/namespaces/<your-namespace>/scripts/<script-name>" \
  -H "Content-Type: multipart/form-data" \
  -H "Authorization: Bearer <api-token>" \
  -F 'metadata={
    "bindings": [
      {
        "type": "r2_bucket",
        "name": "STORAGE",
        "bucket_name": "<your-bucket-name>"
      }
    ],
    "keep_bindings": ["kv_namespace"]
  }'