## The problem MCP solves

The Model Context Protocol (MCP) is an open standard that connects language models to tools and data sources in a uniform way. Before MCP, every integration was re-coded per application — combinatorial and brittle.

MCP flips this: an **MCP server** exposes capabilities, and an **MCP client** discovers and calls them. The same integration then serves any compatible client.

## The three primitives

- **Tools**: functions the model can call.
- **Resources**: contextual data exposed to the model.
- **Prompts**: reusable templates the user can trigger.

```json
{ "jsonrpc": "2.0", "id": 1, "method": "tools/list" }
```

Write the integration once, reuse it everywhere — MCP is to AI tools what LSP was to code editors.