The Manual

Client Success
Manual

Twenty-two short chapters plus a glossary. Read in order if it's new to you, or jump to the chapter that matches what you're stuck on. Each chapter is a few minutes long and has one clear point.

For your AI agent

Skip the reading. Point your agent at this.

Every chapter is also available as raw markdown. The agent reads the index, fetches what it needs, and applies the rules to your setup directly.

Chapters

  1. 00

    TL;DR — The Must-Knows

    The five moves, the magic phrases, and the three daily habits that solve 80% of everything. Read this if you read nothing else.

    5 min read
  2. 01

    How to read this manual

    Read in order if you're new. Jump to the chapter that matches what you're stuck on if you're not.

    2 min read
  3. 02

    What an AI agent actually is

    The architectural difference between chatbots and agents, and why it matters for what you build.

    5 min read
  4. 03

    Your first day with the agent

    What to set up, what to say first, and what to ignore on day one.

    6 min read
  5. 04

    How to talk to it (prompting basics)

    Plain English wins. The handful of patterns that get you 90% of the way to a good prompt.

    7 min read
  6. 05

    The magic phrases

    Twenty-something exact phrases that unblock the agent or change its behavior on the spot.

    5 min read
  7. 06

    Memory and always-loaded context

    How the agent remembers things, what "always-loaded" means, and how to make rules stick.

    6 min read
  8. 07

    Skills — teach it once, run forever

    The single biggest compounding habit in the manual. How to make, edit, and audit skills.

    10 min read
  9. 08

    Daily habits of power users

    Morning check-in, follow-up sweep, evening recap. The three habits that replace fifteen productivity apps.

    5 min read
  10. 09

    Braindumps — stop writing notes, start talking

    Voice-message every idea, frustration, and after-call debrief. The agent handles the rest.

    5 min read
  11. 10

    Training it on your voice

    Per-platform voice skills, training on yourself vs competitors, and how to keep the voice fresh.

    8 min read
  12. 11

    When to use sub-agents

    Past three parallel asks per turn, the main agent drops things. Big work goes to sub-agents.

    5 min read
  13. 12

    Making the agent proactive

    Schedule it to wake up on its own — every morning, after every meeting, when new leads land.

    6 min read
  14. 13

    Connecting tools (email, calendar, CRM, browser)

    Just tell it what you use. "Connect to my Gmail / Calendar / CRM / browser — go figure it out."

    7 min read
  15. 14

    Communication channels

    WhatsApp, Telegram, Slack, Discord, email, voice. Which to use when, and what they do differently.

    6 min read
  16. 15

    What the agent remembers and how

    Short-term context, long-term memory, search, and how to keep the right things in front of it.

    6 min read
  17. 16

    Troubleshooting — fix anything in five minutes

    The two real blockers, the phrases that unstick everything else, and a five-minute self-debug.

    7 min read
  18. 17

    Security and privacy

    What the agent sees, what it stores, what it sends out, and how to keep credentials safe.

    6 min read
  19. 18

    Hosting choices and keeping costs down

    Local vs cloud, model choice, and how to keep monthly cost from creeping up.

    5 min read
  20. 19

    The click moments — when you become dangerous

    The shifts that mark when you've gone from "I don't get this" to "this is doing my job for me."

    6 min read
  21. 20

    Common mistakes (and how to skip them)

    Every beginner makes the same fifteen. Here they are, and the one-line fix for each.

    7 min read
  22. 21

    Glossary

    Every term in the manual, defined in one or two sentences. Use it to look things up, not to read end-to-end.

    4 min read

Newsletter

Get our writing on AI agents, weekly.

One short email a week. The lessons we picked up shipping AI deployments, what's working in 2026, what isn't, what we'd build next. No fluff, no AI hype.

No spam, unsubscribe anytime.