Control MCP use in a Google Cloud organization

By default, any principal with the Service Usage Admin (roles/serviceusage.serviceUsageAdmin) role on a project can enable an MCP server in that project. If you want to restrict MCP enablement within your Google Cloud organization, then you can create a custom organization policy using the gcp.managed.allowedMCPServices constraint to control who can use which MCP tools on specific resources. This document includes some common use-cases for restricting MCP use with custom organization policies.

Required roles

To get the permissions that you need to to set custom organization policies, ask your administrator to grant you the Organization Policy Administrator (roles/orgpolicy.policyAdmin) IAM role on the Google Cloud organization. For more information about granting roles, see Manage access to projects, folders, and organizations.

You might also be able to get the required permissions through custom roles or other predefined roles.

Common use cases for restricting MCP use within an organization

The following sections show common use cases for custom organization policies to control MCP use in a Google Cloud organization. If you copy any of these examples, replace the following:

  • ORG_ID: the Google Cloud organization ID.
  • PROJECT_ID: the Google Cloud project ID.

For more information about custom constraints and organization policies, see Creating and managing custom constraints.

Prevent MCP server usage in an organization

The following organization policy prevents MCP usage within a Google Cloud organization.

{
  "name": "organizations/ORG_ID/policies/gcp.managed.allowedMCPServices",
  "spec": {
    "rules": [
      {
        "enforce": true,
        "parameters": {
          "allowedServices": []
        }
      }
    ]
  }
}

Allow use of a specific MCP server in an organization

The following organization policy allows users to enable the BigQuery MCP server endpoint in all projects within a Google Cloud organization:

{
  "name": "organizations/ORG_ID/policies/gcp.managed.allowedMCPServices",
  "spec": {
    "rules": [
      {
        "enforce": true,
        "parameters": {
          "allowedServices": [
            "bigquery.googleapis.com",
          ]
        }
      }
    ]
  }
}

The endpoint for the specified MCP server is still disabled by default in all projects. You must enable the MCP endpoint at the project level before using it.

Restrict MCP use to a specific folder

The following organization policy restricts the use of MCP tools to the folder experimental and permits use of the BigQuery and Pub/Sub MCP servers.

{
  "name": "folders/experimental/policies/gcp.managed.allowedMCPServices",
  "spec": {
    "rules": [
      {
        "enforce": true,
        "parameters": {
          "allowedServices": [
            "bigquery.googleapis.com",
            "pubsub.googleapis.com",
          ]
        }
      }
    ]
  }
}

Conditionally permit MCP use

The following organization policy uses resource tags to prevent enabling MCP endpoints in Google Cloud projects tagged with the resource tag prod as the environment key within the experimental folder, while allowing MCP enablement in all other projects:

{
  "name": "folders/experimental/policies/gcp.managed.allowedMCPServices",
  "spec": {
    "rules": [
      {
        "enforce": true,
        "parameters": {
          "allowedServices": [],
        },
       "condition": {
          "expression": "resource.matchTag('environment', 'prod')",
        }
      }
    ]
  }
}

Set up a custom constraint

Console

To create a custom constraint, do the following:

  1. In the Google Cloud console, go to the Organization policies page.

    Go to Organization policies

  2. From the project picker, select the project that you want to set the organization policy for.
  3. Click Custom constraint.
  4. In the Display name box, enter a human-readable name for the constraint. This name is used in error messages and can be used for identification and debugging. Don't use PII or sensitive data in display names because this name could be exposed in error messages. This field can contain up to 200 characters.
  5. In the Constraint ID box, enter the name that you want for your new custom constraint. A custom constraint can only contain letters (including upper and lowercase) or numbers, for example custom.disableGkeAutoUpgrade. This field can contain up to 70 characters, not counting the prefix (custom.), for example, organizations/123456789/customConstraints/custom. Don't include PII or sensitive data in your constraint ID, because it could be exposed in error messages.
  6. In the Description box, enter a human-readable description of the constraint. This description is used as an error message when the policy is violated. Include details about why the policy violation occurred and how to resolve the policy violation. Don't include PII or sensitive data in your description, because it could be exposed in error messages. This field can contain up to 2000 characters.
  7. In the Resource type box, select the name of the Google Cloud REST resource containing the object and field that you want to restrict—for example, container.googleapis.com/NodePool. Most resource types support up to 20 custom constraints. If you attempt to create more custom constraints, the operation fails.
  8. Under Enforcement method, select whether to enforce the constraint on a REST CREATE method or on both CREATE and UPDATE methods. If you enforce the constraint with the UPDATE method on a resource that violates the constraint, changes to that resource are blocked by the organization policy unless the change resolves the violation.
  9. Not all Google Cloud services support both methods. To see supported methods for each service, find the service in Supported services.

  10. To define a condition, click Edit condition.
    1. In the Add condition panel, create a CEL condition that refers to a supported service resource, for example, resource.management.autoUpgrade == false. This field can contain up to 1000 characters. For details about CEL usage, see Common Expression Language. For more information about the service resources you can use in your custom constraints, see Custom constraint supported services.
    2. Click Save.
  11. Under Action, select whether to allow or deny the evaluated method if the condition is met.
  12. The deny action means that the operation to create or update the resource is blocked if the condition evaluates to true.

    The allow action means that the operation to create or update the resource is permitted only if the condition evaluates to true. Every other case except ones explicitly listed in the condition is blocked.

  13. Click Create constraint.
  14. When you have entered a value into each field, the equivalent YAML configuration for this custom constraint appears on the right.

gcloud

  1. To create a custom constraint, create a YAML file using the following format:
  2.       name: organizations/ORGANIZATION_ID/customConstraints/CONSTRAINT_NAME
          resourceTypes:
          - RESOURCE_NAME
          methodTypes:
          - CREATE
          condition: "CONDITION"
          actionType: ACTION
          displayName: DISPLAY_NAME
          description: DESCRIPTION
          

    Replace the following:

    • ORGANIZATION_ID: your organization ID, such as 123456789.
    • CONSTRAINT_NAME: the name that you want for your new custom constraint. A custom constraint can only contain letters (including upper and lowercase) or numbers, for example, mcp.managed.allowedMCPServices. This field can contain up to 70 characters.
    • RESOURCE_NAME: the fully qualified name of the Google Cloud resource containing the object and field that you want to restrict. For example, compute.googleapis.com/Instance.
    • CONDITION: a CEL condition that is written against a representation of a supported service resource. This field can contain up to 1000 characters. For example, "resource.matchTag('environment', 'prod')".
    • For more information about the resources available to write conditions against, see Supported resources.

    • ACTION: the action to take if the condition is met. Can only be ALLOW.
    • The allow action means that if the condition evaluates to true, the operation to create or update the resource is permitted. This also means that every other case except the one explicitly listed in the condition is blocked.

    • DISPLAY_NAME: a human-friendly name for the constraint. This field can contain up to 200 characters.
    • DESCRIPTION: a human-friendly description of the constraint to display as an error message when the policy is violated. This field can contain up to 2000 characters.
  3. After you have created the YAML file for a new custom constraint, you must set it up to make it available for organization policies in your organization. To set up a custom constraint, use the gcloud org-policies set-custom-constraint command:
  4.         gcloud org-policies set-custom-constraint CONSTRAINT_PATH
          

    Replace CONSTRAINT_PATH with the full path to your custom constraint file. For example, /home/user/customconstraint.yaml.

    After this operation is complete, your custom constraints are available as organization policies in your list of Google Cloud organization policies.

  5. To verify that the custom constraint exists, use the gcloud org-policies list-custom-constraints command:
  6.       gcloud org-policies list-custom-constraints --organization=ORGANIZATION_ID
          

    Replace ORGANIZATION_ID with the ID of your organization resource.

    For more information, see Viewing organization policies.

Enforce a custom organization policy

You can enforce a constraint by creating an organization policy that references it, and then applying that organization policy to a Google Cloud resource.

Console

  1. In the Google Cloud console, go to the Organization policies page.

    Go to Organization policies

  2. From the project picker, select the project that you want to set the organization policy for.
  3. From the list on the Organization policies page, select your constraint to view the Policy details page for that constraint.
  4. To configure the organization policy for this resource, click Manage policy.
  5. On the Edit policy page, select Override parent's policy.
  6. Click Add a rule.
  7. In the Enforcement section, select whether this organization policy is enforced or not.
  8. Optional: To make the organization policy conditional on a tag, click Add condition. Note that if you add a conditional rule to an organization policy, you must add at least one unconditional rule or the policy cannot be saved. For more information, see Setting an organization policy with tags.
  9. Click Test changes to simulate the effect of the organization policy. For more information, see Test organization policy changes with Policy Simulator.
  10. To enforce the organization policy in dry-run mode, click Set dry run policy. For more information, see Create an organization policy in dry-run mode.
  11. After you verify that the organization policy in dry-run mode works as intended, set the live policy by clicking Set policy.

gcloud

  1. To create an organization policy with boolean rules, create a policy YAML file that references the constraint:
  2.         name: projects/PROJECT_ID/policies/CONSTRAINT_NAME
            spec:
              rules:
              - enforce: true
            
            dryRunSpec:
              rules:
              - enforce: true
            
          

    Replace the following:

    • PROJECT_ID: the project that you want to enforce your constraint on.
    • CONSTRAINT_NAME: the name of the managed constraint that you want to enforce. For example, mcp.managed.allowedMCPServices.
  3. To enforce the organization policy in dry-run mode, run the following command with the dryRunSpec flag:
  4.         gcloud org-policies set-policy POLICY_PATH \
              --update-mask=dryRunSpec
          

    Replace POLICY_PATH with the full path to your organization policy YAML file. The policy requires up to 15 minutes to take effect.

  5. After you verify that the organization policy in dry-run mode works as intended, set the live policy with the org-policies set-policy command and the spec flag:
  6.         gcloud org-policies set-policy POLICY_PATH \
              --update-mask=spec
          

    Replace POLICY_PATH with the full path to your organization policy YAML file. The policy requires up to 15 minutes to take effect.

What's next