A Claude Code curriculum that teaches you to fish, not just copy-paste

What it does

This repository is a sprawling reference manual and curriculum for Claude Code, Anthropic’s agentic CLI coding tool. It spans a seven-module learning path, 181 templates, 271 quiz questions, and a dedicated threat database of CVEs and malicious skills. The material targets developers who want to understand the internal logic—context flow, tool orchestration, memory hierarchy—rather than just memorizing slash commands.

The interesting bit

The project ships its own MCP server, turning the guide into a queryable sidecar inside any Claude Code session with 17 tools and a Haiku agent. It also maps its own stargazers on a world map, which is either charming metrics theater or genuine community transparency—possibly both.

Key highlights

• Security-first angle: tracks 28 CVEs and 655 malicious skills, a depth the authors claim no competitor matches
• 48 Mermaid diagrams explaining architecture, model selection, and multi-agent patterns
• Role-based entry points for tech leads, CTOs, product managers, and career pivoters
• Interactive assessment via a 271-question quiz and a /self-assessment skill
• Published under CC BY-SA 4.0 with PDF and EPUB exports

Verdict

Worth bookmarking if you are trying to move beyond vibe-coding and actually architect AI-assisted workflows. Skip it if you just want a one-command plugin to drop into your repo; the authors explicitly send those users elsewhere.