March 11, 2025
NetApp is excited to announce the launch of Context Protocol (MCP) server for Knowledge Bases. This capability enables customers to utilize MCP to seamlessly connect their enterprise knowledge bases, managed via workload factory, to MCP-compatible AI applications and platforms. By embracing the open MCP standard, customers ensure future compatibility with an expanding ecosystem of AI tools and services.
Many customers struggle to integrate unstructured data to integrate insights from private data sources with AI applications, leading to fragmented knowledge bases and bespoke integrations which lead to increased complexity and scaling inefficiencies. NetApp's local MCP server for knowledge bases addresses these issues by offering standardized tools for LLMs and AI applications to connect to the organizational data and retrieve relevant context needed by AI applications and agents. Using a standard MCP protocol for data integration simplifies and streamlines data access, boosts operational efficiency, and maximizes the potential of generative AI for agentic workflows. NetApp customers can now connect unstructured file data sources residing on on-premises NetApp storage, FSx for NetApp ONTAP, Cloud Volumes ONTAP or any other file storage via SMB or NFS protocols to create structured, searchable knowledge bases that enable extracting context and insights using natural language. The MCP server makes Knowledge Bases available as MCP tools for AI applications and agents to query when the applications need context from enterprise data sources. For example, users of MCP compatible applications such as Claude Desktop, can get more relevant answers that are based on context derived from enterprise data sources, developers using applications such as Amazon Q Developer CLI or Cursor can improve productivity by accessing private data sources such as internal API specifications, design and architectural documents, etc. right from within these applications without any custom integrations.
Discover how NetApp’s MCP server can simplify your private, enterprise data integration with AI applications and boost user productivity by reading the announcement blog and the how to blog.
To get started, explore NetApp's Workload Factory for GenAI documentation and GitHub repository to make the most of this powerful feature and read our release blog here. Once Knowledge Bases have been setup, the MCP server can be deployed locally on the users environment following instructions available here. Users of Anthropic’s Claude Desktop can take advantage of single click deployment using the desktop extension package.