Models & technology

MCP (Model Context Protocol)

People & organizations
Anthropic

MCP is an open standard for connecting AI applications to tools and data. It gives different AI clients a common way to use compatible services.

MCP standardizes connections between AI and external systems

MCP stands for Model Context Protocol. It is an open protocol for connecting AI applications to tools and data sources. Anthropic introduced MCP in late 2024, and adoption spread to other AI companies and software providers. It is sometimes described as “USB-C for AI.”

What problem does it solve?

Without a common protocol, every AI application and external service may need a separate custom integration. If there are M tools and N AI clients, that can create M × N combinations to build and maintain.

An MCP server exposes a standard set of capabilities. Compatible AI clients can then connect through the same protocol instead of requiring a completely different integration each time.

What can MCP enable?

  • Read a calendar and create an event
  • Search internal files or a database
  • Read and write information in services such as Slack or Notion
  • Combine several tools into a multi-step agent workflow

Function calling lets an AI invoke tools; MCP provides a shared way to describe and connect many of those tools across products.

Permissions still matter

Connecting an MCP server grants an AI access to some external capability. Check who operates the server, what data it can read, what actions it can perform, and whether important changes require approval.

MCP support is becoming a practical factor in choosing an AI application. A common protocol can make tools portable, but it does not make every server trustworthy or every permission safe.

Related terms

Sources and review information

Last reviewed July 17, 2026

Back to the AI Glossary

Search this site