Model Context Protocol (MCP)

Model Context Protocol (MCP) is an open, client-server communication standard that lets AI applications connect to external tools, resources, and predefined prompts through a consistent interface.

MCP defines how a client discovers and invokes server-provided capabilities:

  • Tools are callable actions or API requests that the client can ask the server to perform.
  • Resources are structured data or context, such as files, database tables, documents, and schemas, accessible by URI, which the client or model can reference.
  • Prompts are predefined prompt templates that clients can fetch, fill with arguments, and execute to drive structured interactions with language models.

MCP also defines a primitive that runs the other way. Through elicitation, a server can ask the client to gather extra input or a confirmation from the user in the middle of a request. As of protocol version 2026-07-28, MCP is a stateless protocol: every request carries its own protocol version and capabilities, and servers advertise what they support through a mandatory server/discover request.

MCP encodes messages using JSON-RPC 2.0 over one of its two standard transports: stdio for local processes or streamable HTTP for remote servers.

The goal is interoperability and portability across assistants, hosts, and integrations, so developers can build a connector once and reuse it in many AI apps.

Python MCP: Connect Your LLM With the World

Tutorial

Python MCP Server: Connect LLMs to Your Data

Learn how to build a Model Context Protocol (MCP) server in Python. Connect tools, prompts, and data to AI agents like Cursor for smarter assistants.

intermediate ai

For additional information on related topics, take a look at the following resources:

By Leodanis Pozo Ramos • Updated Aug. 27, 2026
AI Coding Glossary Share Feedback