Tools & Data
What Is the Model Context Protocol (MCP)?
An open protocol that gives AI applications a standard way to connect to external tools and data sources.
Definition
The Model Context Protocol, or MCP, is an open protocol that standardises how applications provide context and tools to large language models. It was created at Anthropic and released in November 2024, and is maintained as an open specification with SDKs in several languages.
The problem it solves
Before a shared protocol, every combination of AI client and external system needed its own integration. MCP replaces that with one interface: a server exposes its capabilities in a standard shape, and any compliant client can use them. The integration work becomes linear rather than multiplicative.
What a server exposes
- Tools: operations the model can invoke, such as running a query or calling an API
- Resources: data the client can read into context, such as files or records
- Prompts: reusable templates a user can select
A security note
An MCP server runs with whatever credentials you give it, and a model may be asked to call its tools based on content it has read. Treat data returned from any server as untrusted input rather than as instructions, scope credentials as narrowly as the task allows, and be deliberate about which servers a client is permitted to reach.
MCP is an open protocol that enables seamless integration between LLM applications and external data sources and tools.