Make Your org's metadata available to AI agents with natural language queries by enabling Salesforce Catalog.
Required Editions
| Available in: All Editions supported by Data 360. See Data 360 edition availability. |
| User Permissions Needed | |
|---|---|
| To access Data Cloud Setup |
Data Cloud Architect permission set |
Before you begin, Set Up and Turn On Data 360 in your org, and review Salesforce Catalog considerations.
- From Data Cloud Setup, under Feature Management, select Feature Manager.
- To turn on Salesforce Catalog, next to the feature name, click Enable.
-
Agree to the terms and service, and confirm that you want to turn on the catalog.
After you turn on Salesforce Catalog, an initial full sync runs to completely extract and ingest your org's metadata. This process can take time depending on the size of your org. After the initial sync completes, Salesforce Catalog keeps your metadata current through incremental metadata sync for supported metadata types and daily full refresh. For more information see, Metadata Extraction and Refresh.
- Next, enable the Metadata Grounding Model Context Protocol (MCP) server in your org by following the steps in Activate MCP Servers.
To connect to the Metadata Grounding MCP server from other AI clients such as Claude, ChatGPT, or Cursor, follow the steps in Connecting an MCP Client. After you enable the MCP server, your AI client can connect to a Salesforce org and act on behalf of authorized users.
For more information on Metadata Grounding MCP server capabilities see, Metadata Grounding MCP Server (Beta).