actionable ai

Free guide

Your AI Didn't Finish The Job. It Just Said It Did.

Free guide: the Claude Code skill that forces your agent to prove every claim with a real command and a real output, plus the exact install command and how it scales to big jobs.

  • Why partial compliance happens and why agents admit it the moment you check
  • How the gates file forces proof instead of a claim, line by line
  • The exact install command and what happens on first run
  • How it scales to big jobs with a tree of sub agents checking sub agents

Instant access

Get the free guide

The free Claude Code skill that makes your agent prove it finished the job with a real command and a real output, instead of just saying so.

One-time signup. Every guide in the vault unlocks with it.

A look inside

What's inside

01

What's Inside

What partial compliance actually is, and why it happens How unlazy forces proof instead of a claim The exact install command and first-run walkthrough How it scales to big jobs with sub agents checking sub agents

02

The Problem: Partial Compliance

Ask an AI agent to do five things and it will often do three, then tell you all five are done.

03

The Fix: A Gates File

unlazy, built by developer Leon Lin (github.com/Leonxlnx/unlazy), closes that gap with one idea: before the agent is allowed to touch any code, it has to write down what "done" means in a form a machine can check.

The rest of the guide

  1. How To Install ItOne command, from inside your project That installs the skill into Claude Code.
  2. The Builder And Inspector AnalogyThink of it like renovating a house.
  3. Scaling To Bigger Jobs: The Sub Agent TreeOn a small task, one gates file and one check pass is enough.
  4. The Speed Fix: Parallel Sub Agents, Not One At A TimeOut of the box, unlazy's orchestrated mode hands one task to one sub agent, waits for it to finish, and only then starts the next one, even though both Claude Code and Codex can run several sub agents on different tasks at the same time.
  5. Where It Fits In Your Setupunlazy isn't a replacement for reading your agent's output, it's what stops you from having to read all of it.

About the author

Cooper Simson

Founder of Actionable AI. 30,000+ operators read the newsletter.

Good to know

Questions, answered

Is this guide free?

Yes. Enter your details and you can open the full guide immediately.

Do I need to know how to code?

No. The guide explains the work in practical steps and calls out technical details only when they matter.

What happens after I enter my email?

The guide opens from this page. If the guide email is enabled, you will also receive a link for later.