Earlier this week I confessed in here that I'd scoped about $60K of work for a client who paid me under $20K, and that she'd capped it off by sending me a "you didn't deliver" email at 11pm. A bunch of you commented. Some to encourage me. Some of you, I'm pretty sure, just pulled up a chair and grabbed the popcorn. Either way, you wanted to know how it ended. So here's the sequel. Nobody died. I'll lead with that. ------------------------------------ What I walked into ------------------------------------ Monday she blew a gasket. Strongly worded, escalated, the kind of email where you can feel the caps lock breathing through the screen. I did not respond like a calm professional. I spiraled. Half of me concocted a plan to pull up every receipt and go twelve rounds. The other half, the half I'm less proud of, came up with a half-baked plan to just roll over, apologize for stuff that wasn't even mine to apologize for, and turn myself into a doormat so the discomfort would go away. Running on no sleep, nursing a bruised ego, two bad plans and the stress of not having a newborn baby in the house yet... (wife's at 41 weeks, if she hits 42, she's gonna make that baby come out.) ------------------------------------ Then my COO blew up my whole game plan ------------------------------------ The day before the call, I got on with my fractional COO and word-vomited both of my terrible plans at her. She shut them both down. Didn't tell me to fight. Didn't tell me to fold. She handed me an actual plan. First, homework. Go build a point-by-point breakdown of all three agreements. Every deliverable, what's done, what's not, the percentages, and a whole separate column for the work we did that was never even in the scope. That document was her idea, not my heroic late-night brainwave. I didn't have it. She told me to go make it. Then the move. Don't walk in defensive, don't hand her the wheel. Lead with the full picture, so much clarity up front that she can't steer the thing somewhere sideways. She doesn't actually know what a finished marketing blueprint is supposed to look like, so I shouldn't be handing her the power to define "done." Show the whole list first, then talk.

@Jake Van Clief is the reason I was inspired to make this video. The reason I was able to have Claude create the script in my voice. The reason I was able to create an animation from that script with my branding palette. It’s amazing to see every day here what’s being made possible and so much of it is because of the knowledge being shared by everyone. Thank you Jake! https://youtube.com/shorts/9KSaHCcet8c?si=pymly_RgV9OHETag

Sometimes you just have to stop talking, take action, and prove what is possible. Tonight, I’m sitting with my laptop finishing up a PowerPoint presentation. Tomorrow, I am hosting a Global Copilot Cowork Workshop for colleagues across multiple departments and countries. Looking back, things have moved at absolute lightning speed. The Timeline - March: I started experimenting with Claude Code and joined this amazing community shortly after. - May 4th: I had a high-level meeting with our global Head of AI. He gave me the green light to use Copilot Cowork since Claude Code isn't corporate-approved yet. - Today: I'm teaching others how to do what I just did. In just a short month and a half, I’ve built several custom skills focused entirely on our hardest finance bottlenecks. And the timing couldn't have been better. The Ultimate Test: A Compressed Month-End We were recently acquired, which meant our standard 6-day month-end closing window was aggressively cut down to just 4 days. Normally, this would mean pure panic and stress. We had never missed a deadline, but we always had to fight for it. During the May closing, I put my new Copilot skills to work. The result? It went so smoothly that on day 4, my boss literally looked at me and asked: "Did we forget something?" Normally, she works grueling hours, staying up late to log back on after putting her kids to bed. Not this time. I only had to work a truly long day on Day 1. Day 2 was slightly extended, and Days 3 and 4 were completely normal, quiet, peaceful workdays. We smashed a compressed timeline with zero stress. Tomorrow's Mission: Giving Back Tomorrow, I'm stepping up to the whiteboard. I'm running a workshop to show people what this tech can actually do. We are going to build a "light" finance skill together from scratch. The core lesson is simple but powerful: How to feed raw data into an AI agent and get a perfectly formatted, HQ-compliant report out the other side. What the Future Holds

You built an agentic OS. So did the person three seats over. And when you put your architecture next to theirs, the shapes rhyme. Folders for clients. A governing layer up top. Named agents doing scoped work. An orchestrator routing it all. That's not a coincidence and it's not theft. It's convergence. Builders with no access to my internal architecture keep arriving at doctrine I already wrote down. One builder reasons their way to belief over prompting. Another builder names an agent and discovers that's the hook for giving it a soul. And another builds a routing gate to protect the signal. Different vantages, same destination. None of them copied me. They solved the same problem in their own org and ended up in the same place. I used to find that mildly surprising. Not anymore. Because if a stranger reinvents the doctrine without ever seeing it, the doctrine isn't clever positioning I invented. It's just how this actually works. Convergence is the proof that it's true. Which tells me the map was never the moat. If people can rebuild a system from a screenshot of a folder tree, then handing over the clean version costs me nothing and proves the point. So here it is. An outline of the AgencyOS map is attached to this post. Layers, agent roster, sample folder structure, and tool integrations. Take the shape. Start at Layer 1, not Layer 3. The agents are the fun part, so that's where everyone wants to start. The leverage is upstream. The belief layer governs everything the agents produce. Mine didn't come from a build sprint. It started forming in late 2023, with these ideas spoken aloud on a podcast, before any AI was in the picture. Then I spent over a year encoding that conviction into a system. Then I found ICM and ported it over. I'm still refining it. An agent I can spin up in 15 minutes. The belief layer took years because it had to exist before the machine did. Build the workforce on top of nothing and you get a fast machine with no conviction. And the belief layer is the one thing on that map you can't take from me. I can't take yours from you. You can copy the folders. You can copy the agent names. But you cannot copy what I believe into your system. I can go into my AI and ask it "what do I believe?" and it returns my truth. If you load my doctrine and ask the same question, you get what I believe. Not what YOU believe.

I have directed film sets for years. Last night I directed a website the same way, and I never wrote a prompt. On a set you do not operate the camera. You watch the frame and call what it needs. Slower. Dirtier. Again. The crew has the hands. You have the eye. I ran a build the exact same way. A live page open on screen, an agent holding the hands, and me reacting out loud to what moved in front of me. One change. Watch it land. Call the next. DIRECT THE SCREEN, DON'T PROMPT THE MODEL. A prompt is a memo. You write the whole brief, fire it off, and hope what comes back is what you felt. Directing inverts it. You give one small note against a live render, you see it land the instant it changes, and the screen hands you the next note. What came out of it: a wall of twelve CRT monitors, each playing a showreel at a different second, scanlines and flicker and a teal glow. "Make the cursor knock the screens." Now a mouse sweep tears the signal, splits the colour, sometimes cuts a screen to black static, then it recovers on its own. "Dirtier." A pink and teal spark cracks at the pointer. A custom glitch typeface on the headline. None of it was written down up front. It emerged, note by note. This is not magic. Nothing builds itself. An agent built each change, I reviewed every frame, and I called every move. The taste was mine. The hands were not. The loop, if you want to run it: 1. Stand it up live and look. "Warmer" needs something to be warmer than. 2. One move per round, so your reaction maps to one cause, not five. 3. Make every value a named dial. Steer by feel, not by editing code. 4. The agent proves its own edit compiled before it says done. You stay free to just watch. 5. When a look locks, save it. The final dial values are your design system. The skill that runs this loop is open source. Take it: https://github.com/Pushing-Squares/art-direct Here is the part that stuck. You cannot write a feel down. A brief is a guess at an aesthetic you have not seen yet. The eye only knows once the live thing is moving in front of it. So stop trying to specify the feel. Build a loop fast enough to react inside.