ai-agents-crash-course
Free 42-page AI agents crash course v5.0 — 5 days, 3 learning paths, 6 playable games, 13 Anthropic Academy courses, interactive quizzes, XP gamification. From zero to multi-agent systems. MIT licensed.
Install / Use
claude mcp add itallstartedwithaidea -- npx -y github:itallstartedwithaidea/ai-agents-crash-courseIf the server publishes to npm under a different name, use that package instead — check the repo README.
MCP Server
Model Context Protocol server
Quality Score
Category
AI & Machine LearningSupported Platforms
Skill content
View source on GitHub🤖 About
A free, open-source, 42-page crash course on building AI agents — written by John Williams, Senior Paid Media Specialist at Seer Interactive with 15+ years managing $48M+ in digital advertising spend, and creator of googleadsagent.ai.
Every concept explained with accessible analogies. Every step verified with "What You Should See" callouts. Every day ends with a playable video game built through natural-language prompting. Formatted in APA 7th Edition with 16 formal references.
What's New in v5.0 (March 2026)
- Anthropic Academy Integration — 13 free courses with certificates linked throughout, including Claude 101, Claude Code in Action, MCP courses, AI Fluency, and Agent Skills
- Interactive MCQ Quizzes — 6 graded quizzes (21 total questions) replace text-area assessments with instant feedback and scoring
- XP & Gamification System — Earn XP for completing days and quizzes, level up from Novice to Master, earn 6 achievement badges
- Daily Streak Tracking — Return daily to build your streak; 3-day streaks unlock badges
- New Products Coverage — Claude Code, Claude Cowork (Jan 2026), Agent Skills, MCP updates
📚 What's Inside
Document Structure
| Section | Contents | |---------|----------| | Cover | Branded title page with author credentials | | Table of Contents | 5-page manual TOC with all 87+ headings | | Preface | Why This Course Exists, Audience & Paths, How to Use, Acknowledgments | | Executive Summary | 1-page overview for quick reference | | Section 0 | Complete A-Z setup: Claude, ChatGPT, Cursor, Kaggle, GitHub, Python | | Days 1–5 | Each with: Learning Objectives → Content → Key Takeaways → Self-Assessment | | Appendix A | Complete Resource Library (white papers, courses, frameworks, platforms) | | Appendix B | Glossary of 22 key terms | | Appendix C | googleadsagent.ai Technical Reference (28 actions, 6 sub-agents) | | References | APA 7th Edition bibliography (16 entries) | | About the Author | Full bio and credentials |
Three Learning Paths
| Path | Who | Focus | Time/Day | |------|-----|-------|----------| | 🟢 Observer | Executives, strategists | Concepts only | 2–3 hrs | | 🔵 Builder | Marketers, analysts | Hands-on labs + games | 4–6 hrs | | 🟣 Architect | Developers, engineers | Deep theory + production | 6–8 hrs |
Five Days + Capstone
| Day | Topic | White Paper | Game | |-----|-------|------------|------| | 0 | Complete Environment Setup | — | — | | 1 | Introduction to AI Agents | Introduction to Agents | 🎮 Tetris | | 2 | Agent Tools & MCP | Agent Tools & MCP | 🎮 Zelda Adventure | | 3 | Context Engineering | Sessions & Memory | 🎮 Football (Tecmo Bowl) | | 4 | Agent Quality & Observability | Agent Quality | 🎮 Super Mario Bros | | 5 | Production & Multi-Agent | Prototype to Production | 🎮 Duck Hunt | | Cap | Everything combined | All five | 🎮 AI Game Master RPG |
🤖 The googleadsagent.ai Through-Line
googleadsagent.ai is woven throughout as a living reference implementation:
| Component | Implementation | Day |
|-----------|---------------|-----|
| Brain | Claude by Anthropic (Opus 4.5 + Sonnet 4.5) | 1 |
| Tools | 28 Python API actions with live Google Ads API access | 2 |
| MCP | Adapter layer: tool_executor.py | 2 |
| Context | CEP Protocol + Session State Manager | 3 |
| Quality | Top-Down Reporting + CONFIRM write-safety | 4 |
| Multi-Agent | 🦁 Simba · 🐠 Nemo · ❄️ Elsa · 🧞 Aladdin · 🌊 Moana · 🤖 Baymax | 5 |
| Production | CLI, REST API, Docker, Python SDK | 5 |
📥 Download
| Format | Description | Link |
|--------|-------------|------|
| 📄 Word (.docx) | Full course with formatted callouts, tables, code blocks | docs/AI-Agents-Crash-Course-V4.docx |
| 📕 PDF | Print-ready, 42 pages | docs/AI-Agents-Crash-Course-V4.pdf |
| 📝 Markdown | Read on GitHub | docs/COURSE.md |
🎮 Game Projects (Vibe Coding)
Each day includes a game built entirely through natural-language prompting. Copy the prompt, paste into Claude or Cursor, play instantly.
| Day | Game | Prompt |
|-----|------|--------|
| 1 | Tetris | games/day1-tetris/prompt.md |
| 2 | Zelda Adventure | games/day2-zelda/prompt.md |
| 3 | Football (Tecmo Bowl) | games/day3-football/prompt.md |
| 4 | Super Mario Bros | games/day4-mario/prompt.md |
| 5 | Duck Hunt | games/day5-duckhunt/prompt.md |
| Cap | AI Game Master RPG | games/capstone-rpg/prompt.md |
📖 References (APA 7th Edition)
All sources formally cited in the course:
| Source | Date | |--------|------| | Sundar Pichai — Google I/O 2025 Keynote | May 2025 | | Demis Hassabis — Gemini 2.0 Launch | Mar 2025 | | Google — Agent2Agent Protocol (A2A) | Apr 2025 | | Google — Cloud Next 2025 | Apr 2025 | | Sundar Pichai — NRF 2026 Remarks (UCP) | Jan 2026 | | Alphabet Inc. — Q2 2025 Earnings | Jul 2025 | | Google/Kaggle — 5 White Papers | 2024–2025 | | Anthropic — MCP Documentation | Nov 2024 | | Yao et al. — ReAct Paper | 2022 | | Karpathy — Vibe Coding | Feb 2025 | | Microsoft — AI Agents for Beginners | 2025 |
Full bibliography in resources/REFERENCES.md
🎓 Anthropic Academy — Free Courses
All 13 courses are free with certificates on Anthropic Academy (Skilljar):
| Course | Focus | Level |
|--------|-------|-------|
| Claude Code in Action | AI coding assistant, GitHub, MCP servers, hooks, SDK | Intermediate |
| Claude 101 | Chat, Cowork, Code modes, Projects, Artifacts, Skills | Beginner |
| AI Fluency: Framework & Foundations | The Four Ds — Delegation, Description, Discernment, Diligence | All levels |
| Building with the Claude API | API calls, streaming, function calling, structured outputs | Developer |
| Intro to Model Context Protocol | MCP servers/clients in Python, tools, resources, prompts | Developer |
| MCP: Advanced Topics | Sampling, notifications, file system, transport mechanisms | Advanced |
| Introduction to Agent Skills | Build and share SKILL.md workflows in Claude Code | Developer |
| AI Fluency for Educators | AI Fluency in teaching and institutional strategy | Educators |
| AI Fluency for Students | AI skills for learning and career planning | Students |
| AI Fluency for Nonprofits | AI fluency for mission-driven organizations | Nonprofits |
| Teaching AI Fluency | Teaching and assessing AI Fluency in instructor-led settings | Instructors |
| Claude with Amazon Bedrock | Anthropic models on AWS | Cloud |
| Claude with Google Cloud Vertex AI | Anthropic models on Google Cloud | Cloud |
🎮 Gamification & Interactive Quizzes
The live course includes:
| Feature | Details | |---------|---------| | MCQ Quizzes | 6 quizzes, 21 questions, instant grading with feedback | | XP System | Earn XP for completing days (20 XP each) and quizzes (30-50 XP each) | | 6 Levels | 🔰 Novice → 📘 Learner → 🔧 Builder → ⚡ Practitioner → 🏗️ Architect → 👑 Master | | 6 Badges | 🎯 First Quiz · 💯 Perfect Score · 🏆 All Days · 🔥 3-Day Streak · 🎓 Academy · ⭐ Completionist | | Daily Streaks | Visit daily to build your streak counter | | Progress Tracking | All progress saved locally in your browser |
🔗 Essential Resources
White Papers — All 5 Google/Kaggle Papers · Intro to Agents
Anthropic Academy — Claude Code in Action · Claude 101 · All 13 Courses
Protocols — MCP Docs · A2A Docs
Courses — Microsoft AI Agents · Hugging Face · DeepLearning.AI
Tools — Claude · ChatGPT · Cursor · googleadsagent.ai
📂 Repository Structure
ai-agents-crash-course/
├── README.md ← You are here
├── LICENSE ← MIT
├── docs/
│ ├── AI-Agents-Crash-Course-V4.docx ← Full course (Word)
│ ├── AI-Agents-Crash-Course-V4.pdf ← Full course (PDF)
│ └── COURSE.md ← Full course (Markdown)
├── games/
│ ├── day1-tetris/prompt.md
│ ├── day2-zelda/prompt.md
│ ├── day3-football/prompt.md
│ ├── day4-mario/prompt.md
│ ├── day5-duckhunt/prompt.md
│ └── capstone-rpg/prompt.md
└── resources/
├── REFERENCES.md ← APA bibliography
└── LINKS.md ← All links & resources
👨💼 About the Author
John Williams — Senior Paid Media Specialist at Seer Interactive with 15+ years managing $48M+ in digital advertisin
Truncated for display — read the full file on GitHub.
Related Skills
momen-cursurrules-prompt-file
40.6kCursor rules for building custom frontends with Momen.app as headless BaaS with GraphQL API, actionflows, AI agents, and Stripe integration.
pyspark-etl-best-practices-cursorrules-prompt-file
40.6kCursor rules for PySpark ETL development with code style, joins, window functions, map operations, and Iceberg patterns.
semiotic-react-dataviz-cursorrules-prompt-file
40.6kCursor rules for Semiotic data visualization library with 30+ chart types, MCP server, and AI-assisted chart generation.
claude-mem
90.6kPersistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
