What's inside
An operating system, not a prompt pack
Four pieces that turn an AI coding assistant into a team that plans, tests, reviews, ships against your roadmap — and refuses to build things nobody would miss.
Plans agents execute — including the marketing
The PLAN.md and MARKETING_PLAN.md templates that run the lab: lanes with validation gates and 'done when' criteria, a KPI morning routine, channel cadences, and the rule that a week without distribution actions is a failed week. The part every other AI kit leaves out.
A validation gate your AI enforces
No feature gets build hours until four questions are answered in writing: what problem, who feels it daily, why current solutions fail, who would miss it tomorrow. Your agents refuse to build ungated ideas — the discipline that stops you shipping things nobody would miss.
47 agent definitions + layered rules
Planner, TDD guide, code reviewer, security reviewer, build-error resolvers, per-language reviewers — backed by a common layer of standards with overrides for TypeScript, Python, Go, Swift, PHP, and the web. Delegation instead of chat.
Quality-gate hooks
Commits linted for secrets and message format, edits auto-formatted and type-checked, hook-bypass flags blocked. The standards stop being optional — for you and for your AI.
How it works
Installed before your coffee cools
-
1
Install the rules and agents
Copy the rule layers you use and the full agent roster into ~/.claude. Two commands, no configuration.
-
2
Turn on the hooks
Merge the hooks config and scripts. From the next session, commit quality, secret detection, and format/type checks run automatically.
-
3
Drop the plan into your repo
Fill in the PLAN.md template — lanes, priorities, acceptance criteria — and open Claude Code. Your agent reads the plan and starts executing.
Proof
The system's output is this site
Every product below was planned, built, tested, and deployed with the exact rules, agents, and hooks in this kit.
- FoodTracker — iOS nutrition app with StoreKit 2 subscriptions, live on the App Store
- MicroReps — free macOS fitness app, shipped as DMG and on the App Store
- Close Idle Tabs — Edge extension, live on the store
- ScriptPal and Accent Cracker — two more products in the pipeline, same system
- All of it built and shipped by one person orchestrating AI agents
FAQ
Questions, answered
Has this system made you rich?
No — and you should distrust anyone selling you otherwise. The lab's revenue is early and I publish it, including the ugly numbers: my first app launched to 12 downloads. What the system verifiably did is let one person ship five production apps — and then forced me to confront that building was never the bottleneck. That's why it now schedules validation and distribution as ruthlessly as code quality. I'm selling the operating system I actually run, not an income claim.
What exactly do I get?
A downloadable kit: the layered rules directory, 45+ agent definition files, the hooks config with all backing scripts, the PLAN.md and MARKETING_PLAN.md templates, and a written operating manual covering setup, your first agent-run feature, and the daily loop.
Do I need Claude Code?
The kit is built for Claude Code (CLI or desktop). The rules and plan templates are plain Markdown and adapt to other agentic tools, but the hooks and agent definitions target Claude Code directly.
Is this a course or a prompt pack?
Neither. It is a working configuration — the same files used to run this lab — plus the manual that explains how the pieces fit. You install it and your next session behaves differently.
Will it stay up to date?
The kit reflects the live setup at Patako Labs and is updated as the lab evolves. Buyers get updates through the same download link.
Build less. Distribute more. Ship what's missed.
One download. The next session, your agents follow the plan — and the plan includes the marketing.
Get the system — $39 launch price (opens in new tab)