Glossary

What is MCP (Model Context Protocol)?

MCP (Model Context Protocol) is an open standard that lets AI applications such as Claude connect to external data sources and tools. An MCP server exposes, for example, a document archive or a database, and any compatible AI assistant can use that source without a custom integration being built for every combination.

In simple terms

MCP is like USB-C for AI. Devices used to need their own charging cables; now one connector fits many devices. In the same way, every AI assistant and every data source used to need its own custom connection. With MCP, you build one connector for the data source, and any assistant that speaks MCP can use it. The operator of the server decides which data is exposed.

In the enterprise

Muster Handels GmbH wants its staff to ask questions about supplier contracts in the AI assistant they already use. IT runs an MCP server on the internal network that searches the contract archive. In the assistant, contract search appears as an additional tool. The server only returns documents the user is allowed to open in the archive anyway, and every answer points to the contract and page.

Related terms

Sources

  1. Model Context Protocol: Specification
  2. Model Context Protocol: Introduction (official documentation)
  3. Anthropic (2024): Introducing the Model Context Protocol

How we use this: UNID AI Second Brain · The second brain for your company's knowledge

← All terms

FAQ

Frequently asked questions

Who developed MCP?

Anthropic released the Model Context Protocol as an open standard in November 2024. The specification and software development kits are publicly documented at modelcontextprotocol.io, and MCP is now supported by other AI vendors and tools as well.

See it working on your own material.

In a 30-minute call we show the products live and work out which use case makes sense for you first. No obligation, in German or English, in Vienna or online.

UNID AI · Leonard-Bernstein-Straße 10 · 1220 Vienna · +43 660 476 30 89