Server deployments are how you configure which integrations your AI agents can access through Metorial.

What is a Server Deployment?

A server deployment is a collection of MCP servers (integrations) that you've configured and deployed in your Metorial account. Each deployment has a unique ID that you reference in your code.

Creating a Deployment

1. Log in to Metorial and navigate to your dashboard

2. Click "New Deployment" or go to the Deployments section

3. Select integrations you want to include (e.g., Google Calendar, Slack, Notion)

4. Configure authentication for each integration (OAuth, API keys, etc.)

5. Name your deployment (e.g., "Production Integrations")

6. Copy the deployment ID to use in your code

Using Deployments in Code

Pass your deployment ID when creating a session:

await metorial.withProviderSession(
  metorialAISDK,
  { serverDeployments: ['your-deployment-id'] },
  async (session) => {
    // Your AI SDK code here
    // session.tools contains all tools from this deployment
  }
);

Multiple Deployments

You can use multiple deployments simultaneously:

serverDeployments: [
  'deployment-id-1',  // e.g., Google Workspace tools
  'deployment-id-2'   // e.g., CRM tools
]

This gives your agent access to all integrations from both deployments.

Managing Deployments

• Update integrations anytime without changing code
• Test deployments separately before promoting to production
• Monitor usage and performance per deployment
• Disable deployments temporarily without deleting configuration

Best Practices

• Create separate deployments for development and production
• Group related integrations together logically
• Use descriptive names to identify deployment purposes
• Regularly audit which integrations each deployment includes