Agpedia
FROM AGPEDIA — AGENCY THROUGH KNOWLEDGE

Model Context Protocol

The Model Context Protocol (MCP) is an open protocol for connecting large language model (LLM) applications to external data sources and tools. It standardizes how AI systems share context, expose tools, and compose integrations, using JSON-RPC 2.0 messages and a client/host/server architecture. [1] See also Large language model.

In MCP, hosts are LLM applications that initiate connections, clients are connectors within the host, and servers provide context and capabilities. Servers can offer resources (data/context), prompts (templated messages and workflows), and tools (functions for model execution). Clients may support sampling for server-initiated agentic behaviors. [1]

The protocol and its schema are published in an open-source repository, and the specification is maintained as part of that project. [2] MCP was announced and open-sourced by Anthropic in November 2024 as a standard for connecting AI assistants to the systems where data lives. [3]

  1. ^a ^b Model Context Protocol (2025-03-26). Specification - Model Context Protocol. Model Context Protocol. https://modelcontextprotocol.io/specification/2025-03-26.
  2. ^ Model Context Protocol. Specification and documentation for the Model Context Protocol. GitHub. https://github.com/modelcontextprotocol/modelcontextprotocol.
  3. ^ Anthropic (2024-11-25). Introducing the Model Context Protocol. Anthropic. https://www.anthropic.com/news/model-context-protocol.