Getting Started with Informatica MCP Servers > Introduction to Informatica MCP servers
  

Introduction to Informatica MCP servers

MCP (Model Context Protocol) is an open-source standard for connecting AI applications to external data sources, tools, and workflows. You can use MCP to connect your AI applications to Informatica’s MCP servers. This enables your AI agents to ground their responses using your IDMC organization’s trusted data.
Note: Effective in the November 2025 release, Informatica MCP servers are available for private preview. Preview functionality is supported for evaluation purposes but is unwarranted and is not supported in production environments or any environment that you plan to push to production. Informatica intends to include the preview functionality in an upcoming release for production use, but might choose not to in accordance with changing market or technical circumstances. Note that if you are working on a preview POD, all data is excluded from SOC 2 compliance coverage. For more information, contact Informatica Global Customer Support.
MCP eliminates the need to create complex point-to-point integrations for each tool that your AI application interacts with. You no longer need to handle the API connectivity, authentication, data access, and code manually. MCP acts as a universal translator for each external tool that you want your AI application to interact with.
When you connect to Informatica MCP servers, your AI application acts as an MCP host, and each of your AI agents acts as an MCP client. Each MCP client maintains a one-to-one connection with an Informatica MCP server.
The following image illustrates the process:
The image shows an AI application, which is the MCP host. Two AI agents, which are the MCP clients, exist within the AI application. Each AI agent maintains a one-to-one connection with an Informatica MCP server.
Each MCP server exposes the tools that are available to your AI agent. Informatica calls these tools "actions." For example, the CDGC Metadata Search server provides actions for getting asset details (get_asset_details) and searching metadata (search_metadata). When you configure your AI agents, you can select the specific actions the agents need to interact with.
Informatica MCP servers use MCP protocol 2025-06-18 and streamable HTTP transport.
You can connect to Informatica MCP servers using a third-party tool like Claude Desktop, Cursor, VS Code, or Python. You can also use the MCP tool in AI Agent Engineering.