---
sourceDocument: Australia Enable AI
sourceDocumentLink: https://www.servicenow.com/docs/r/australia/intelligent-experiences

 Release :

    - australia

ft:locale :

    - en-US

ft:publication_title :

    - Australia Enable AI

ft:clusterId :

    - platai

bundleId :

    - platai

workflow :

    - Platform


---

# Add an MCP server from AI Control Tower

# Add an MCP server from AI Control Tower {#ariaid-title1}

Release version: Australia  
Updated September 10, 2026  
![](https://www.servicenow.com/docs/portal-asset/ico-clock) 1 minute to read  
Add an MCP server manually in AI Control Tower.

## Before you begin

Role required: sn_ai_governance.ai_steward  
Note:  
The AI Steward (sn_ai_governance.ai_steward) role gets inherited with the following roles:

* (AI admin) sn_aia.admin
* (AIG admin) aig_admin
* (MCP Client admin) sn_mcp_client.admin
{#aict-add-an-mcp-server-from-ai-control-tower__ul_vwn_3g4_lkc}

## Procedure

1. Navigate to AI assetsAI asset inventoryMCP servers.
2. Select Add.
3. Enter the Name (Server name).
4. Enter the MCP server URL.
5. Enter the Description (Optional).
6. Enter the External MCP server documentation URL (Optional).
7. Select the Provider from the list.
8. Select the Vendor from the list.
9. Select Next.
10. Select the Client registration type.  
    Note:  
    If you select Dynamic client registration, AI Control Tower automatically receives the Client ID and Client secret. If you select Manual registration, you must create the client externally on the MCP provider side and enter the Client ID and Client secret manually.
11. If you select Manual registration, review and complete the following details:  
    * Grant type
    * Token authentication method
    * Client ID
    * Client secret
    * Auth scopes
    * Authorization URL
    * Token URL
    * Token revocation URL
    {#aict-add-an-mcp-server-from-ai-control-tower__ul_szv_zl4_lkc}
12. Select Submit.
{#aict-add-an-mcp-server-from-ai-control-tower__steps_cgs_cl4_lkc}

## Result

The MCP server is registered in the AI asset inventory as an unmanaged asset with a status of In review.

## What to do next

MCP server approval workflow.

*[\>]: and then


