A one-stop study resource for mastering Claude Code. Every concept has a guide, a working skill, a cheatsheet, and a self-test quiz.
No prior Claude knowledge required. Start wherever you are.
Clear explanations from zero. Every term is defined. No assumed knowledge.
Real-world skills for Airflow, PySpark, HTTPX APIs, and ETL pipelines.
Generic versions of every example that work in any stack or language.
Prerequisites: basic Python knowledge. That's it.
Follow in order or jump to what you need. Each module links to its guide, skills, quiz, and cheatsheet.
| # | Module | Guide | Skills | Quiz | Cheatsheet |
|---|---|---|---|---|---|
| 01 | Claude models & ecosystem | 📖 Guide | — | 📝 Quiz | — |
| 02 | Claude Code architecture | 📖 Guide | — | 📝 Quiz | — |
| 03 | Installation & setup | 📖 Guide | — | 📝 Quiz | ⚡ Commands |
| 04 | CLAUDE.md | 📖 Guide | — | 📝 Quiz | ⚡ Template |
| 05 | Skills | 📖 Guide | /fetch-api /migrate-data /visualize | 📝 Quiz | ⚡ Template |
| 06 | Sub-agents & memory | 📖 Guide | /readme-gen | 📝 Quiz | — |
| 07 | Hooks | 📖 Guide | /code-review | 📝 Quiz | — |
| 08 | Scheduling | 📖 Guide | /schedule-task | 📝 Quiz | ⚡ Cron syntax |
| 09 | MCP servers | 📖 Guide | — | 📝 Quiz | ⚡ MCP config |
| 10 | Plugins & marketplace | 📖 Guide | — | 📝 Quiz | — |
Copy any skill into your project's .claude/skills/ folder and trigger it with a slash command.
Fetch data from REST endpoints asynchronously. Saves responses to timestamped directories, handles retries, and logs all HTTP activity.
Convert data files between CSV, Parquet, and JSON. Works on single files or entire directories. Preserves schema and handles encoding edge cases.
Generate charts from data files using matplotlib and seaborn. Auto-detects data shape and picks appropriate chart types.
Auto-generate a README for any project. Scans the codebase structure, infers purpose, and writes a complete markdown README.
Review recent code changes for bugs, style violations, and security issues. Triggered by hooks on file save or manually on demand.
Set up a recurring /loop scheduled job. Configure cron expressions, manage job IDs, and cancel tasks.
Every example ships as a generic version (any stack) and a data-engineering version (Airflow/PySpark flavour).
Async HTTP fetch with httpx. Saves results to timestamped output dirs. Demonstrates error handling and retry logic.
CSV ↔ Parquet ↔ JSON conversion with pandas and pyarrow. Handles large files, custom delimiters, and schema validation.
Auto-generate charts from data files. Bar, line, scatter, and heatmap. Saves PNGs to reports/ directory.
Example hook configurations — PostEdit lint triggers, PreTool guards, and SessionStart injections.
Demonstrates spawning parallel sub-agents for independent tasks and collecting their results.
Local and hosted MCP server configurations. Connect Claude Code to external tools via the Model Context Protocol.
Dense, not verbose. Keep these open while you work.
Three phases. Follow them in order or use as a checklist.
Understand what Claude Code is, get it installed, and write your first CLAUDE.md.
Build real skills, delegate to sub-agents, and configure hooks that fire automatically.
Schedule work with cron, connect external tools via MCP, and explore the plugin marketplace.
Especially new skill templates, additional examples, and corrections — Claude Code evolves fast.
Add to examples/ — include both generic/ and data-engineering/ variants.
Fix anything stale in the docs. Open a GitHub issue with what's wrong and what's correct.