Same ideas, two stacks
Every row in these tables is one skill wearing two outfits. The Google column is what you have free for a year; the Internet Menace column is the same idea with the branding sanded off. Learn the idea instead of the product and month thirteen becomes a pricing decision. Learn the product instead of the idea and you start over every time a rebrand ships.
Tool equivalence: learn the idea once, run it in either stack
| Portable idea | Internet Menace | |
|---|---|---|
| A file the agent re-reads every session | GEMINI.md | CLAUDE.md |
| An agent with a filesystem and a shell | gemini-cli | Claude Code |
| A saved persona with standing instructions | Gems | System prompts and playbooks (§2.6) |
| Review the plan before execution | Deep Research "Edit plan"; Antigravity Implementation Plan | Specifications, reviewed (§2.2) |
| A reviewable record of what changed | Antigravity Walkthrough | Git discipline and pull requests (§6.2) |
| Answers anchored to sources you control | @Drive grounding, NotebookLM | Context engineering (§2.4); RAG |
| A protocol for adding tools to an agent | MCP servers via npx add-mcp | MCP, the same protocol |
| Compute you do not own vs. compute you do | Colab GPU tiers, Vertex AI | Self-hosting (§4.4, §4.5) |
| Knowing what a run costs | Compute units, credits, budget alerts | Cost literacy (§4.6) |
Where each part of the guide goes deeper
§2Gemini Advanced and Workspace Extensions
- Curriculum
- 2.1 Requirements: What You Actually Want2.2 Specifications: What The Machine Actually Needs2.4 Context Engineering (the Missing Discipline)2.5 Prompting Based On Data Shape
- Templates
- Product Requirements DocumentFeature SpecificationCLAUDE.md
- Stacks
- RAG Knowledge App
- Glossary
- GroundingRAGContext windowSystem promptHallucination
§3gemini-cli: AI in Your Terminal
- Curriculum
- 2.3 Structured Markdown: The Format Coding Agents Read Best2.6 The Claude Code Playbook3.1 Ides And Editors: Vs Code, Cursor, Anti Gravity3.2 Coding Agents: Claude Code, Openclaude, Others3.6 Crewai For Agents
- Templates
- CLAUDE.md
- Glossary
- MCP (Model Context Protocol)Agent loopTool use (function calling)Coding agentClaude Code
§4Google Antigravity: Agentic Development (Preview)
§5Google Colab AI
§6Choosing the Right Tool
§7Bonus: Google Cloud Platform Credits
- Curriculum
- 4.2 How Models Actually Work (weights, Parameters, Active Parameters)4.6 Cost Literacy: Tokens, Context Windows, And The $200 Question6.4 Secrets, Api Keys, And What Not To Put In A Prompt
- Stacks
- AWS Production
- Glossary
- Foundation modelFrontier modelModel weightsOpen weightsEnvironment variable
§8How the Tools Work Together
- Curriculum
- 2.4 Context Engineering (the Missing Discipline)5.6 Api Design For Vibe Coders6.6 The Deployment Cliff: Making It Run While You Sleep7.4 The Data Analyst Leveling Up (dashboards And Pipelines)
- Templates
- Product Requirements DocumentREST/GraphQL API Skeleton
- Stacks
- RAG Knowledge AppAWS Production
- Glossary
- EmbeddingChunkingVector databaseServerless