Claude Code Agentic OS Map (Free 2-Page PDF)

A two-page map of five layers that turn Claude Code into a setup that remembers your rules, stays out of folders you protect, and asks before anything leaves your machine.

For each layer, the map says what it does and which of our products or course modules covers it:

• Memory loader: a folder of plain Markdown notes, and the right ones load into every session.
• Safety hooks: refuse writes to folders you list and keep huge tool output out of context.
• Vetted skills: skills read line by line before they reach your machine.
• Subagent routing: a roster of subagents, with routine work sent to cheaper models where that is enough.
• Durable jobs and approve cards: long work survives a crash, and nothing goes out until you tap Approve.

It also marks what we don't cover: voice, and an Obsidian or web front door.

It's free. Enter your email and the PDF is yours.

Written with AI help (Claude) and reviewed by AmagiLabs. AmagiLabs isn't affiliated with Anthropic. Claude and Claude Code are Anthropic's trademarks.

Made by AmagiLabs (amagilabs.ai).