Model Context Protocol (MCP)
Model Context Protocol (MCP) is an open protocol that lets AI applications connect to external tools and data sources through a uniform client-server interface.
In plain English
MCP — Model Context Protocol — is the USB-C of AI integrations. Anthropic introduced it in late 2024 as an open standard, and the major AI app vendors rapidly adopted it. The idea: stop writing one-off integrations between each AI app and each data source. Instead, every integration speaks the same protocol. Any MCP-speaking client can use any MCP-speaking server.
A protocol sounds boring until you feel the leverage. Once a tool is exposed as an MCP server, every MCP-speaking AI app can use it: Claude Desktop, Cursor, Zed, custom agents. Once an AI app speaks MCP as a client, it inherits the entire ecosystem of servers without custom code. That is how standards compound — one integration unlocks every combination on both sides.
Why it matters for Black Box
Black Box speaks MCP as a client. Any MCP server the user installs — a CRM connector, a database proxy, a Gmail tool — becomes available to the specialist agents without Black Box shipping custom code. Owners keep their integration investment even if they change AI vendors.
Examples
- Installing a Notion MCP server so agents can read and write pages.
- Using a database MCP server that exposes read-only SQL over a Postgres instance.
- Running a filesystem MCP server that gives agents scoped read/write access to a workspace directory.