All articles
AI + PM

The Kit Is Out, and the App Can Wait

TL;DR

  • Today ThirtyYearPM shipped its first product, the PM AI Starter Kit: eleven project-management procedures that run inside the AI tools you already use.
  • You type what happened in plain language, it asks the questions a proper register needs, and then it writes the scored risk, the log row, or the status report.
  • It costs $29 once. No subscription, and nothing leaves your machine.
  • I run a studio that builds software, so shipping a file first took some deciding; the full reasoning lives on the ThirtyYearPM blog.
  • The Akigo! suite is still coming, Echo beta first, and what buyers run most will influence what gets built first.

In May I wrote here that AI takes the administrative parts of a project manager’s job first, and that the widening judgment gap is why I built ThirtyYearPM (the post, if you want the argument). Today the first of those tools went out the door. It is called the PM AI Starter Kit, and it is a download.

For most of thirty years I assumed the way to package what I know was software: build the app, put a login on it. What shipped instead is a skill file, a workbook, some templates, and a nine-page guide. That ordering was a decision, and the companion post over on ThirtyYearPM explains why the kit came before the app. This one is the builder’s side of it.

What You Get

The kit turns plain-language updates into the records a project needs. Type “the vendor’s docs slipped two weeks” and it responds the way a good PM would: it asks for probability, impact, owner, and response, then writes the scored risk into the log. The same shape covers issues, change requests, decisions, assumptions, constraints, action items, dependencies, status reports, future-state risk assessments, and lessons learned. Eleven procedures, each one tested on real projects before it shipped, because a procedure that only works when a form forces it was never the point.

The download has five parts. A skill file with prompts for every procedure, built for ChatGPT, Claude, and Gemini, and it works pasted into any other AI if that’s where your team lives. A 17-tab Excel workbook with one log per procedure, a dashboard that does its own math, and tabs for stakeholders, RASI, and communications. A board with KPI tiles, RAG counts, and a ranked list of your top impediments. Branded document templates for when a stakeholder wants a page instead of a spreadsheet. And the guide, which walks through install, your first project, and the daily rhythm.

Everything writes into your workbook, on your machine. The kit doesn’t phone home. That’s the line on the product page, and I mean it the same way I meant it about agent-sleep: plain files, no accounts, no telemetry.

It costs $29, once. One license per person, updates to this version line included, and everything the kit produces belongs to you, client work included. No subscription, because the last thing a status report needs is a renewal date.

Why the File Went Out First

Mocoda has been shipping software for other people since 1996, so when it came time to package thirty years of my own method, shipping a file before an app needed a reason.

The app plan still exists. The Akigo! suite (Echo, Flow, Pulse, Notes) is in development, all local-first, and the waitlist on the site is for it. The kit is a deliberately slimmed-down Akigo! Pulse, the project-intelligence part of the suite, reshaped to run inside the AI you already use. The first app is in testing now; the Echo beta comes after it.

The reason for the order is empirical. Before I harden thirty years of judgment into native software, I want to know which parts of it people reach for. A kit that sells and gets used answers that better than any survey, and buyers hear about the apps first, so what they run most will influence what gets built first. It is the discipline I argued for in the October evidence post, pointed at my own roadmap this time. Usage tells you where the value is; a roadmap only tells you what you assumed.

Built for Directing and Verifying

The October post landed on a claim I keep coming back to: the PM’s job bends toward directing and verifying. The kit is built around that division of labor. You supply the facts and the calls; it handles the record-keeping that used to eat a Friday afternoon. It asks before it writes, and it will not put a risk in your register that you didn’t confirm, because a register full of plausible fiction is worse than an empty one.

Projects rarely fail because nobody knew the framework. They fail when the documentation goes unwritten, or when month nine arrives and nobody can find it. That has been true for thirty years. The AI part is that the writing no longer costs an evening, which means the records can finally keep up with the project.

Try It

The kit is at thirtyyearpm.com/kit: $29, instant download, and the guide covers install and your first project. There is a fifty-seven-second video on the page if you’d rather watch the loop before reading anything. And if the native apps are what you’re waiting for, the waitlist is on the same site; buyers hear about those first.

Shayne Yestal

Project manager, studio founder, and AI advocate with 30+ years of experience. Co-founder of Mocoda. Creator of ThirtyYearPM.