> ## Documentation Index
> Fetch the complete documentation index at: https://dev.writer.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Private endpoints

> Connect WRITER Agent to connectors over a private network using AWS PrivateLink or GCP Private Service Connect

A private endpoint is a network path between Writer's MCP gateway and a service in your VPC. Instead of sending connector traffic over the public internet, Writer connects through [AWS PrivateLink](https://docs.aws.amazon.com/vpc/latest/privatelink/what-is-privatelink.html) or [GCP Private Service Connect](https://cloud.google.com/vpc/docs/private-service-connect) so requests stay on your cloud provider's private network.

<Note>
  For prebuilt connectors, contact your customer success manager to enable
  private endpoints. You can enable private endpoints on custom connectors
  during connector setup in AI Studio.
</Note>

## Prerequisites

Before setting up a private endpoint, make sure you have:

* Administrator access to your Writer organization
* Administrator access to your AWS or GCP account
* A service running in your VPC that you want to expose to WRITER Agent

## Create a private endpoint in your cloud provider

The first step is to create a private endpoint service in AWS or GCP that exposes your service to Writer.

### AWS PrivateLink

Create a VPC endpoint service in AWS that allows Writer to connect to your service privately.

1. Follow the AWS documentation to [create an endpoint service](https://docs.aws.amazon.com/vpc/latest/privatelink/create-endpoint-service.html)
2. When configuring allowed principals, add the Writer AWS account so that Writer can connect to your endpoint service
3. Copy the **Service name** (for example, `com.amazonaws.vpce.us-east-1.vpce-svc-xxxxxxxx`) for use in AI Studio

### GCP Private Service Connect

Create a published service in GCP that allows Writer to connect to your service privately.

1. Follow the GCP documentation to [configure Private Service Connect as a producer](https://cloud.google.com/vpc/docs/configure-private-service-connect-producer)
2. When configuring accepted consumers, add the Writer GCP project so that Writer can connect to your service
3. Copy the **Service attachment URI** (for example, `projects/PROJECT/regions/REGION/serviceAttachments/SERVICE_ATTACHMENT`) for use in AI Studio

## Add the private endpoint in AI Studio

After creating the endpoint service in your cloud provider, add it in AI Studio so it's available when configuring connectors.

1. Navigate to **Connectors & Tools** in [AI Studio](https://app.writer.com/aistudio)
2. Select **Private endpoints**
3. Select **Add private endpoint**
4. Enter the following details:
   * **Name**: A descriptive name for the private endpoint
   * **Cloud provider**: Select **AWS** or **GCP**
   * **Region**: The region where your endpoint service is hosted
   * **Service name** (AWS) or **Service attachment URI** (GCP): The identifier you copied from your cloud provider
5. Select **Save** to add the private endpoint

The new private endpoint appears in your list and is available when you configure connectors.

## Use a private endpoint with a custom connector

After you add a private endpoint in AI Studio, attach it when you create or edit a [custom connector](/home/custom-connectors):

1. Create or edit a custom connector in AI Studio under **Connectors & Tools**
2. In the **Connect** step, toggle **Enable private endpoint** on
3. Select the private endpoint you added
4. Complete the rest of the connector configuration

WRITER Agent routes requests for that connector through the selected private endpoint instead of the public internet.

## Next steps

* [Configure connectors](/home/mcp-gateway): Learn about connector authentication and access control
* [Custom connectors](/home/custom-connectors): Create custom connectors from OpenAPI specifications or MCP servers
* [Set up connectors](https://support.writer.com/article/299-setting-up-connectors): Configure prebuilt connectors in AI Studio
* [MCP gateway overview](/home/mcp-gateway): Learn about Writer's MCP gateway architecture