Model Context Protocol Documentation
The open standard for connecting AI assistants to tools and data, with a specification, SDKs and working examples. If you are building anything that gives a model access to your systems, this is the interface to learn rather than invent.
Guide
AI & Machine Learning
Anthropic
A few hours
Intermediate
Free
What it is
An open protocol that defines how an AI application talks to external tools, data sources and prompts. A server exposes capabilities, a client consumes them, and the transport is ordinary JSON-RPC over standard input and output or HTTP.
Why it matters
Before a standard existed, every assistant integration was bespoke, which meant a tool built for one client did not work with any other. A shared protocol turns integrations into something you write once. The specification is small enough to read in an afternoon and implement directly if you would rather not take a dependency.
What to read first
The core concepts covering tools, resources and prompts, then the transport section. Tools are actions the model chooses to invoke; resources are context a person attaches. Getting that distinction right is most of designing a good server.
Best for: engineers connecting AI assistants to internal systems, and anyone building developer tooling in this space.
Ready to start?
Opens on Anthropic in a new tab.
Stuck on something specific?
Writing only gets you so far. If you want an answer to your situation rather than the general case, book a session and we will work through it together. Every session is free; a few slots open each week.
Follow along
New writing, resources and project ideas land here first.