Field guide · Edition 001 · Kickstarter Sept 26, 2026

The field guide to using AI agents responsibly at work.

A living subscription guide of tested workflows you can run. Each one shows the real tool output, the safeguards, and a 15-minute Try-it. Recipes get updated as the tools change.

Free preview + first access at launch · 3 emails, no spam

01 / Stays current

Rewritten as the tools move

Models and products change by the week. We update recipes when a step breaks or a better path shows up, often weekly or monthly. The $47/yr subscription is what pays for that work.

02 / Authored, then open

Co-authors fill it before launch

Founding writers are on contract. They write and test the first library. Outside contribution opens later, once there is a real catalog to join, not an empty page waiting for strangers.

03 / Responsible by design

Safeguards sit above the steps

Every recipe names what never goes in the prompt, who reviews before anything leaves the building, and when you stop. That block is read before Method.

Read an actual recipe

Every entry is a workflow you can run.

Here’s a real one from the founding library — the whole thing, exactly as it appears in the guide. This is the format behind every workflow.

recipe RE-01 · v1.2 · ~15 minReal Estate

Draft a property-market brief with an AI agent — without leaking client data.

Problem
You need a sharp neighborhood market brief for a client meeting, fast — but the listing sheet and client notes can’t go into a public model. Job: turn messy local data into a credible one-pager in minutes, safely.
Method
  1. Strip identifiers from the source data locally (a provided script does this).
  2. Prompt the agent with the anonymized sample and a fixed brief template.
  3. Review the draft against the checklist before anything leaves your machine.
Safeguards
Redaction runs on your machine — raw client data never reaches the model.
A human review step is required before the brief is shared. The checklist is part of the recipe.
Try it
Run the same brief on the included sample dataset (no real client data) in about 15 minutes — then swap in your own, safely. ◷ cost to try ≈ $0.03
A note on why

Most AI advice is hype or theory, and neither survives contact with real work. We are writing the guide we wish we had: workflows that actually run, with the safeguards built in, kept current by the people who use them every day as the tools move. For a lot of us, fluency with these tools is how you build the life you are working toward. Fund the first edition and you help decide what goes in it.

The co-authorsAiBook.fun founding team
Real estate · finance · leadership · tech
Who’s writing this

Practitioners, one per vertical.

Each chapter is owned by someone who does that work for a living. Names go up as co-authors confirm.

RE

Real Estate & Finance

Market briefs, compliant summaries
Co-author confirmed
CS

Fractional C-Suite

Strategy & board workflows
Co-author confirmed
TE

Technology & RAAI

Agent patterns & safe defaults
Co-author confirmed
HE

Higher Education

Teaching & research integrity
To be announced
CO

Communities

Moderation & member programs
To be announced
Your name?

Contribute a workflow after launch and earn a byline in the guide.

Be first

Get the preview chapter free.

Join the pre-launch list. You’ll get one full workflow to keep, a heads-up the morning we launch on Kickstarter, and early-backer pricing. Three emails. No spam, ever.

Launching Sept 26, 2026 · unsubscribe anytime · we track sign-ups, never your work

When we launch

What backing will look like.

Preview of planned pledge tiers. Final pricing is confirmed at the Kickstarter launch. Nothing here is a live checkout yet.

Supporter
$10
one-time
The preview chapter, community updates, and an early look at the founding library as it is written.
The one to get
The Book
$47 /yr
annual subscription
The full guide in PDF and web reader, plus ongoing recipe updates while you are subscribed. Billed yearly. No lifetime SKU in v1.
Founding Sponsor
Founding
limited · final terms at launch
A named acknowledgment in the guide, a private cohort seat, and a hand in shaping the first edition.

Frequently asked.

Is this AI theory, or something I can actually run?
Something you run. Every workflow shows the real tool output, lists the real steps, and ends in a ~15-minute Try-it on non-confidential sample data. If you can't run it, it doesn't ship.
How is this different from a course?
A course teaches concepts. This is a field guide of tested recipes you apply to your job this week. The recipes get rewritten as the tools change, not left sitting until next year.
What does "responsible by design" actually mean?
Every recipe includes data-handling rules, human-oversight steps, failure modes, and escalation paths. They sit in the template above Method, not in a footnote.
Can I contribute a workflow after backing?
Yes. Open contribution starts once the founding library is live. How contributors get paid is still being finalized before launch.
Why subscribe instead of buy once?
A one-time AI book goes stale in weeks. The $47/yr subscription funds ongoing updates as models and products move, often weekly or monthly.
Why frame this around the American Dream?
For most of the people writing and reading this guide, fluency with these tools is a practical ladder. Nobody in this book inherited their edge. They built it, one real skill at a time. We think this is the next one worth building.