The AI Playbook
A guide to keep, not a course to finish. 12 pages and 93 short points on how to think about, ask, steer and check AI agents. Free to read, no sign-up.
Start here
What an agent is, and how this Playbook works.
Think like a manager
Your chat is a manager with a whole team behind it. Work with it that way.
Asking well
What to put in a request so the first answer is the one you wanted.
How models differ
Each model family reads your words differently. Write for the one you are using.
Your agent's team
Helpers, second opinions and parallel work, all from one chat.
Going in circles
Spot the loop early and change something, instead of asking the same thing again.
Checking the work
How to trust what an agent hands you, without redoing it yourself.
Getting more done
Ways to get more useful output for the same time and effort.
What the agent knows
Where the agent's knowledge comes from, and how to give it what it needs.
Agents in your day
Changing how you work, not just adding a new tool.
Working as a team
Getting more from agents together than each person alone.
Staying safe
Good judgment about what you give the agent and what you let it do.
Want the long way through?
The Playbook is the quick reference. The courses go deeper, with quizzes. Join to unlock them, and to hear when Interweave OS opens to more teams.