What it does
unlazy is an add-on "skill" for AI coding agents such as Claude Code and Codex. Its goal is to stop an agent from declaring work finished too early. Before starting, the agent writes a list of goals, called gates, in a simple text file.
Each gate can include a check command and the result it should produce. A small checker tool runs these checks and marks a gate as met only when the check succeeds. You must approve each check command before it is allowed to run. Finished work can also be re-checked later.
For bigger jobs, it can split a task into smaller pieces that run in parallel, each with its own gates. An optional hook for Claude Code keeps the agent from stopping while gates are still unmet. The makers note that approval is consent, not a safety sandbox: checks run with full access to your computer.
Who it's for
Developers who use AI coding agents for large tasks and want proof that each part of the work was actually done and tested.
Can I use it without coding?
Developers only. It is installed with a command-line tool and requires writing check commands and reading shell code. New to this? Start here.
Can I run it?
- Runs on
- Mac, Windows, Linux
- Computer
- An ordinary computer is fine
Try it
- Install the skill into your AI coding agent using the skills command-line tool, or copy the project folder into the agent's skills folder.
- Ask your agent for a substantial task using the /unlazy trigger.
- Review and approve the check commands in the gates file before letting the checker run them.
Using Claude Code, Cursor or another AI coding assistant? Paste the prompt and it will walk you through the setup.New to this?
Before you install
- Only download from the official GitHub page. Copies with similar names exist for many popular projects.
- Plain-AyI is an independent guide and doesn't host or review the software it lists. Please check each project's official page before installing.
- License (MIT): Very permissive: free to use, change and share, including for business.
- More on staying safe