MCP tutorials
Step-by-step guides for connecting AI clients to MCP servers, building your own and deciding what each agent can call, with every step run before it was published.
Start here
MCP is the protocol an AI client uses to call tools that run somewhere else.
11 tutorials
- What is an MCP server?
- MCP vs APIs: what is the difference?
- What is an AI agent?
- AI agents vs agentic AI: what is the difference?
Connect AI clients
Setup for the clients people actually run, Claude Code and Cursor among them.
2 tutorials
All 2Build MCP servers
How to write an MCP server yourself, one language at a time, and how to generate one from an OpenAPI spec when the API already exists.
12 tutorials
- How to build an MCP server in Python
- How to build an MCP server in TypeScript
- How to build an MCP server in Go
- How to build an MCP server in Rust
Agent security
How AI agents get attacked, and the controls that hold.
22 tutorials
- What is prompt injection?
- What is MCP tool poisoning?
- The lethal trifecta: how AI agents leak data
- What is an MCP rug pull?
Choose your setup
The decisions you make once and live with: local or remote servers, API keys or OAuth, a gateway or direct connections.
4 tutorials
- Shadow MCP: finding unsanctioned MCP servers on your team's laptops
- Local vs remote MCP servers: which should you run?
- API keys vs OAuth for MCP servers
- What is an MCP gateway?