Model Context Protocol (MCP)
The Model Context Protocol (MCP) is an open-source standard that defines how AI applications connect to external systems such as data sources, tools and workflows through a common interface.
Also called: MCP, MCP server
Why it matters
Before a shared standard, every AI application needed its own custom integration for each tool. MCP separates the two sides: an MCP server wraps a system such as a CRM, database or search service, and any AI application with an MCP client can use it. Servers expose capabilities such as tools the model can call, resources it can read, and reusable prompts.
For go-to-market teams, MCP means an AI assistant can read CRM records, query enrichment data or send a Slack message from one conversation, as long as a server exists for each system. MCP is a protocol, not a product: it does not decide what the model should do, and access control still depends on the credentials and permissions each server is given.
Example
A RevOps manager connects an AI assistant to two MCP servers, one for the company CRM and one for a data provider. She asks it to list open deals with no activity in 30 days and to look up the current title of each main contact. The assistant calls the CRM server, then the data server, and returns one table.
Turn the definition into a working system.
We build the GTM Engineering systems behind terms like this one. Tell us where your pipeline breaks and we will show you what we would fix first.
