MCP
Model Context Protocol — an open standard that lets AI models connect to external tools, data sources, and services through a unified interface.
MCP (Model Context Protocol) is an open protocol developed by Anthropic that standardises how AI models connect to external systems. Before MCP, every integration between an AI and an external tool (a database, a file system, a web API) required custom code. MCP defines a standard "plug" shape that any tool can implement once and be usable by any compliant AI.
Key concepts: - MCP servers expose tools, resources, and prompts that an AI can call - MCP clients (AI applications) discover and call those tools at runtime - The same MCP server can work with Claude, GPT, Gemini, or any compliant model
Practical examples: an MCP server for a company's internal database lets Claude query it directly; an MCP server for GitHub lets Claude read and create pull requests; an MCP server for Notion lets Claude read and write pages.
MCP is increasingly a selection criterion for enterprise AI tools — platforms that support MCP can be composed into multi-tool agentic workflows without custom integration work.
Example
A developer connects Claude to their codebase via an MCP server. Claude can now read files, run tests, and check git history without the developer copying and pasting code into the chat.
Related terms
Tool Use / Function Calling
Enabling an AI to call external functions or APIs to perform tasks beyond text generation.
Agentic AI
An AI system that can break down goals into steps, use tools, and iterate until a goal is achieved.
Agent Orchestration
Coordinating multiple AI models, tools, or agents to work together on a complex goal.