32dots HEIDELBERG AI
Curriculum › Claude Code course › claude-code-07
Session 7 medium

Connect your tools: MCP

LESSONLesson 7 · ~20 min

🎯Goal. Pull external data and tools into a coding session with the Model Context Protocol (MCP).

  1. 1MCP is an open standard for connecting AI tools to outside data. With it, Claude Code can read design docs in Google Drive, pull tickets from Jira, fetch data from Slack, or use your own custom tooling.
  2. 2Run /mcp to see configured servers and enable or disable them. Tool definitions are deferred by default, so only tool names enter context until Claude actually uses one (run /context to see what's loaded).
  3. 3Prefer a plain CLI (gh, aws) when one exists — it's more context-efficient than an MCP server. The MCP quickstart in the docs connects your first server end to end.

✓You'll see. Claude reading from a connected service — e.g. pulling a spec from Drive — inside the same session where it writes the code.

💳Cost. MCP tool definitions are deferred, so an idle server costs almost nothing; disable unused ones with /mcp to keep context lean.

💡Takeaway. MCP lets Claude reach your real tools and data — connect a server once and it's available to every session.