Model Context Protocol
em-see-PEE
An open protocol that standardizes how AI applications connect to external data sources and tools.
MCP (Model Context Protocol) is an open protocol created by Anthropic that standardizes how AI applications connect to external tools and data sources. It defines a common interface so that AI models can interact with databases, APIs, file systems, and other services in a structured, secure way.
Before MCP, every AI application had to build custom integrations for each tool it needed to access. MCP provides a standard so that a single integration works across multiple AI applications. This is similar to how USB standardized hardware connections: instead of a custom cable for every device, one standard works for all.
For developer tools, MCP matters because it is how AI coding assistants connect to your product. If your product has an MCP server, any AI assistant that supports MCP can interact with it directly. This is the infrastructure layer of agentic commerce.
Examples
A database provides an MCP server.
The database publishes an MCP server that lets AI assistants query data, inspect schemas, and run migrations. A developer using Claude Code or Cursor can interact with the database through natural language, and the AI uses MCP to execute the actual operations.
A company builds MCP support into its API.
The developer tools company publishes an MCP server alongside its REST API. AI coding assistants can now create deployments, check build status, and read logs through MCP. Developers who use AI assistants get native integration without any custom setup.
MCP enables multi-tool AI workflows.
An AI assistant uses MCP to connect to three tools simultaneously: a project management tool (to read tasks), a code repository (to write code), and a deployment platform (to ship changes). MCP provides the standardized connections between the AI and all three tools.
In practice
Read more on the blog
Frequently asked questions
Who created MCP?
Anthropic created and open-sourced MCP. It is designed to be an open standard that any AI application and any tool can implement. The goal is interoperability: one protocol for connecting AI to external systems.
How is MCP different from a regular API?
An API defines how one specific service can be accessed. MCP defines a standard protocol for how AI applications access any service. A company still has its API, but the MCP server wraps it in a standard format that AI assistants understand natively.
Related terms
A commerce model where AI agents evaluate, compare, and purchase products on behalf of human decision-makers.
An AI system that can plan, reason, and take autonomous actions to complete tasks. Goes beyond chat to actually do things.
The network of tools, libraries, integrations, and community resources that surround a developer platform.
A file published on websites that helps AI language models understand the site's content structure and key information.
Want the complete playbook?
Picks and Shovels is the definitive guide to developer marketing. Amazon #1 bestseller with practical strategies from 30 years of marketing to developers.