← All docs
Getting started with MCP
What is the Model Context Protocol?
MCP is an open standard for connecting AI assistants to external tools and data sources. Think of it like USB for AI: instead of every assistant needing a custom integration for every tool, MCP provides one universal plug.
A MCP server exposes tools (actions the AI can take), resources (data the AI can read), and prompts (reusable templates) through this standard protocol.
Why does it matter?
Without MCP, your AI assistant is smart but isolated — it can only work with what's in its training data. With MCP servers, it can:
- Read and write files on your computer
- Query databases and APIs
- Control browsers and applications
- Access your GitHub, Slack, calendar, and more
Your first 5 minutes
- Pick a server from our directory. Start with something official (look for the ✓ Verified badge).
- Install it following the server's setup guide (usually `npm install` or a download).
- Connect it to your AI client — see our connection guide.
- Ask your assistant to use it. Try: "List the files in my Documents folder" (with a filesystem server).
What's next?
Once you're comfortable, explore servers by category, save your favorites (create a free account), and submit servers you've built.