Anthropic's official "Code with Claude" workshop materials — nine hands-on courses on skills, MCP, evals and Claude Managed Agents.
- official
- workshops
- anthropic
- managed-agents
Model Context Protocol Docs
Official MCP documentation, quickstarts, and examples.
Official Model Context Protocol documentation. Covers the spec, concepts, transport (stdio / Streamable HTTP), security, and debugging guidance. The canonical entry point for MCP.
What it does
- Official MCP documentation with three build tracks: build servers (expose your data and tools), build clients (applications that connect to MCP servers), and build MCP Apps (interactive apps that run inside AI clients)
- Architecture and Security sections cover the core concepts and the security considerations and best practices for MCP in one place
- States the ecosystem pitch plainly: MCP is an open protocol supported across clients — Claude, ChatGPT, VS Code, Cursor, MCPJam among them — build once, integrate everywhere
- Community track documents how to get involved and contribute to MCP itself
When to use it
The canonical MCP learning path — from the intro through architecture and security to whichever build track you need; the reference to cite when explaining what MCP is.
Getting started
No install — documentation. Note the URL carries a dated spec version (2026-07-28): link the version you actually code against when writing your own docs.
Things to know
Docs are versioned with the spec — features described under one dated version may differ in another; keep SDK and docs version aligned when debugging protocol behavior.
Charted
Related on the map
Anthropic's announcement of Fable 5.1 and Mythos 5.1 — one base model, two guardrail postures.
- official
- models
- announcement
- fable
Anthropic's canonical essay on agent design — workflows vs agents, and five composable patterns.
- official
- agents
- engineering
- patterns