The Agent Operator’s Handbook
An Unofficial Power-User’s Field Guide to Meta’s Personal AI Agent
Most people use an AI agent like a chatbot. They ask a question, get an answer, and start over tomorrow. I use mine like an executive assistant. It runs while I sleep, and it does about 80% of what a human EA would do.
This book is the field guide to doing that. It’s 28 chapters of real systems, not theory. Every chapter is something I actually run.
Inside the book
Inbox triage, overnight
Everything that came in gets sorted three ways: needs a reply, needs my eyes, or noise. The ones that need a reply already have a draft waiting. I approve in one word.
Morning briefing, in five minutes
Overnight news, today’s calendar, inbox triage, and anything with a clock on it. One brief, first thing, and the day is set.
Pre-meeting briefs
Who you’re meeting, why it matters, and what to ask, before you walk in.
Delegation anatomy
How to hand work off so it actually comes back done, with approval before anything commits.
Research delegation
Hand a question off and get an answer, not a pile of links.
Watchlists and monitoring
Your AI agent watches what matters and flags what changes, so nothing slips.
Who it’s for
If you use AI here and there and feel like you’re barely scratching the surface, this is your sign.
It’s written for business owners and team managers, and for the ops person or team lead a manager hands it to.
Why this book is on the Altaris site
Because it’s how I operate. Altaris helps companies make smarter software licensing decisions, and the systems in this book are the ones I run my own business on. If you want to know how I think about leverage, systems, and getting more done with less drag, this is the clearest picture of it you’ll find.
Get the book
Available on Kindle for $4.99.
Questions about the book? Reach me at [email protected].