Connect a Client Using an MCP Server

Interact with agents from multiple clients and with Mosaic models in a ChatGPT-like interface connected to an MCP Server. Before you can interact, you must connect the client.

For an overview of using a MCP server, see Interact with Agents and Mosaic Models Using an MCP Server.

Connecting a client is the same procedure for agents and for Mosaic; only the URL differs.

Copy the URL for your server from Server URLs and add it in your client, then sign in to Strategy Library when prompted. Connect to Clients provides steps for common MCP clients.

Prerequisites

• You must have the privileges outlined below.

Action	Required Privilege	Notes
Connect to MCP server	Authenticated user	Clients must have valid credentials and tokens
	Use MCP Server	Beginning in Strategy One (March 2026), must be set to use MCP for agents and Mosaic
List agents	Read access	Only users with permission can list agents
Ask agent	Run AI Bots	Must be allowed to execute actions and questions

• You cannot be an administrator or in the Platform Support Administrator user group.

• Agents must be certified to be used in the MCP server. This allows you to determine which agents are available. For example, if you want to test only production agents, you can certify only those.

• The agent must be added to Library.

Server URLs

In the following URLs, env-domain serves as a placeholder for your environment ID. Replace env-domain with your environment's domain and append the corresponding URL.

• Agent MCP Server URL: https://env-domain.trial.cloud.strategy.com/collaboration/mcp/agent

• Mosaic MCP Server URL: https://env-domain.trial.cloud.strategy.com/collaboration/mcp/mosaic

For clients that use manual OAuth configuration, use the following values:

Connection	URL
Authorization URL	https://env-domain.trial.cloud.strategy.com/collaboration/authorize
Token URL	https://env-domain.trial.cloud.strategy.com/collaboration/token
Refresh URL	https://env-domain.trial.cloud.strategy.com/collaboration/token
Scopes	offline_access

Connect to Clients

Enable the MCP Server before connecting MCP clients to it for interaction with agents and Mosaic.

Select the connection procedure for the client you are working with:

• Claude

• Copilot Studio

• Gemini Enterprise
• ChatGPT
• AWS/Amazon Quick Suite

• VSCode Copilot

Connect to the MCP Server using Claude

• Mosaic Setup Video Example

• Agent Setup Video Example

1. Log into Claude PRO version.

2. In the toolbar in the chat box, click the Search and Tools icon and select Manage Connectors.

3. Click Browse Connectors.

4. In the introductory paragraph, click the link for add a custom one.

5. In the Add Custom Connector window, type a Name for the connector.

6. Type your Remote MCP Server URL using the following syntax:

   https://env-domain/collaboration/mcp/agent

   Replace domain with your MCP server.

   • Beginning in Strategy One (March 2026), you can access Mosaic models in the MCP server. Use this syntax instead: https://env-domain/collaboration/mcp/mosaic.

7. Click Add.

8. If the same window opens again, click Cancel.

9. Check that your new connector displays in the list of connectors.

10. Click Connect next to your new connector.

11. Strategy Library opens. Log in with your credentials.

12. At the message, click Open Claude.

    Your new connector now displays Configure instead of Connect.

13. Click the New Chat icon in the sidebar.

Connect to the MCP Server using Copilot Studio

• Mosaic Setup Video Example

• Agent Setup Video Example

• Teams Setup Video Example

1. Open Copilot Studio.

2. Click Agents in the toolbar on the left.

   

3. Create a new agent by clicking Create Blank Agent in the toolbar or click an existing agent.

   

4. Scroll down to the Tools section and click Add Tool.

   

5. Click New Tool on the right, below the suggestions.

   

6. Click Model Context Protocol.

   

7. Type the Server Name, Server Description, and Server URL including the complete server path. For the URL, use the following syntax:

   https://env-domain/collaboration/mcp/agent

   Replace domain with your MCP server.

   • Beginning in Strategy One (March 2026), you can access Mosaic models in the MCP server. Use this syntax instead: https://env-domain/collaboration/mcp/mosaic.

8. For the Authentication, select OAuth 2.0.

   

9. For the Type, select Manual.

10. Type your Client ID, Client Secret, Authorization URL, Token URL Template, Refresh URL, and Scopes.

    

11. Click Create.

12. Copy the Redirect URL displayed in the Preview window. Provide it to the MCP owner, so they can help to add the redirect_url in the MCP configuration.

    

    • The MCP owner adds it in Workstation, in the environment's properties. It is located on the Library page, in the Model Context Protocol for Agent section, as shown below:

      

13. Click Next.

14. To establish the connection, click Not Connected.

    

15. After you are signed in and connected, click Create.

    

16. A new browser window displays. Strategy Library opens. Log in with your credentials.

You can now interact with your agent or Mosaic model using Copilot Studio.

Connect to the MCP Server using Gemini Enterprise

• Mosaic Setup Video Example

• Agent Setup Video Example

1. Go to the Google Agent Development Kit site and follow the installation guide to set up the Agent Development Kit on your machine. Once installed, you will have a deployment file ready to work with.

   If you have an existing ADK deployment file in your GitHub repository, go to the next step.

2. Create a Python virtual environment.

3. Install all dependencies from pyproject.toml.

4. Ensure the mcp_url and AUTH_ID match in both agent.py and deploy.py.

5. Deploy your ADK agent to the Vertex AI Agent Engine by running the following command:

   python deploy.py

6. Once deployment succeeds, you can view your agent running in the Agent engine:

   

7. Add a new resource to the Gemini Enterprise Authorization server and register the ADK agent to Gemini Enterprise. Follow the steps below:

   1. Search for Gemini Enterprise.

   2. Navigate to agents.

   3. Click Add Agent.

   4. Select Custom Agent Via Agent Engine.

   

8. Add a new authorization, ensuring that the Authorization ID matches the AUTH_ID deployed in the deploy.py script.

   

9. Click Next.

10. Add your agent's configuration. Follow the steps below:

    1. Use the same Authorization ID as above.

    2. Type the resource name of the ADK engine in the Agent Engine Reasoning box. You can find the resource name in the Agent Engine portal.

    3. Click Create.

    

Connect to the MCP Server using ChatGPT

• Mosaic Setup Video Example

• Agent Setup Video Example

The Callback URL varies depending on the address of the connected MCP server. You need to copy it from ChatGPT into Workstation before completing the connection.

1. Open ChatGPT.

2. In the bottom left, click your name.

3. Click Settings.

   

4. In the navigation pane on the left, click Apps.

5. Click Create App.

   

6. Fill in the information to create the app.

   1. Add an Icon image file (optional).

   2. Type a Name, an optional Description, and the MCP Server URL using the format https://env-domain/collaboration/mcp/agent, replacing domain with your MCP server.

      • Beginning in Strategy One (March 2026), you can access Mosaic models in the MCP server. Use this syntax instead: https://env-domain/collaboration/mcp/mosaic.

   3. From the Authentication drop-down list, select OAuth.

   

7. Click Advanced Settings.

8. In the Registration Method, select User-Defined OAuth Client.

9. Click the Click to Copy icon next to the Callback URL.

   

10. Add the Callback URL to Workstation:

    1. Open Workstation.

    2. Connect to the environment.

    3. Right-click the environment and select Properties.

    4. In the navigation pane on the left, click Library.

    5. Scroll to the Model Context Protocol for Agent section.

    6. Click Add New URL next to Callback URL (for Server Applications) and paste the copied URL.

       

    7. Click OK to save the change.

11. Return to ChatGPT.

12. In the Advanced Settings panel, select Dynamic Client Registration (DCR) from the Registration Method drop-down list.

13. In the New App panel, select the check box next to I understand and want to continue.

    

14. Click Create.

You can now interact with your agent or Mosaic model in ChatGPT.

Connect to the MCP Server using AWS/Amazon Quick Suite

1. Log into the AWS access portal.

2. Click AdministratorAccess.

   

3. Search for Amazon Quick Suite and select it.

   

4. In the navigation bar on the left, click Connectors.

5. Click the Create for Your Team tab. Click Model Context Protocol.

   

6. Click Create New.

7. Type the Name and Description.

8. In the MCP Server Endpoint box, copy and paste your MCP server URL using the following syntax:

   https://env-domain/collaboration/mcp/agent

   Replace domain with your MCP server.

   • Beginning in Strategy One (March 2026), you can access Mosaic models in the MCP server. Use this syntax instead: https://env-domain/collaboration/mcp/mosaic.

   

9. Click Next.

10. From the OAuth Configuration drop-down list, select Default OAuth App.

    • If metadata discovery fails, it falls back to user authentication. In that case, type the Client ID, Client Secret, Token URL, and Authorization URL.

    

11. Click Create and Continue.

12. Review the list of enabled actions. Click Next.

    

13. After publishing, click Try It to test the MCP, or continue with the next step to test.

    

    Start Testing

14. In the navigation panel on the left, click Chat Agents.

    • To interact with Mosaic models, see Mosaic Workflows.

15. Click Create Chat Agent.

    

16. If you want to specify the agent's persona, describe it, then click Generate. If not, click Skip.

    

17. Type a name in the box at the top of window.

18. In the Actions section at the bottom of the window, click Link Actions.

    

19. Select MCP Agent, then click Link.

    

20. At the top of the window, click Launch Chat Agent.

    

21. Once the Successfully Launched Chat Agent message displays, you can interact with your agent using AWS.

    

Connect to the MCP Server using VS Code Copilot

1. Open Copilot.

2. Open a new window using Ctrl+Shift+P.

3. Look up MCP: Add Server.

4. Choose HTTP (HTTP or Server-Sent Events).

5. Paste your MCP server URL using the following syntax:

   https://env-domain/collaboration/mcp/agent

   Replace domain with your MCP server.

   • Beginning in Strategy One (March 2026), you can access Mosaic models in the MCP server. Use this syntax instead: https://env-domain/collaboration/mcp/mosaic.

6. Type a Name for the server.

7. Select Global.

8. Click Allow authentication.

   

9. Click Open the external website.

   

10. Strategy Library opens. Log in with your credentials.

11. In the window, click Open Visual Studio Code.

    

12. Ensure your mcp.json file looks like the image below, with your own URL. Click Start above the mcp line.

    

13. Click Allow to connect.

    

14. Ensure that the MCP connection is running, by checking the mcp.json file for the Running status.