Free tools

/experiment

Optimises anything you can measure

What is it?

A free skill for AI agents such as Claude Code. Point it at something you want to improve, and it works out how.

What problem does it solve?

You know something could work better, but not which change would do it. Guessing takes time, and a change that looks better on a few tries can be worse overall. This skill lets the agent find the changes itself. It comes up with ideas, tests each one fairly, keeps what makes the result better and drops what doesn't.

How can I use it?

Tell your agent: "Install the experiment skill from github.com/antopaulogist/experiment." Then tell it what to improve. For example: "Improve our website chatbot's rules so it gets fewer customer questions wrong." Give it some real questions with the right answers. It tries adding, changing and removing rules, keeps each change that cuts the wrong answers, and drops the rest.

/ai-brain

A proper memory for your AI tools

What is it?

A free add-on that gives AI tools such as Claude Code, Codex and Cursor a proper brain, with a short-term and a long-term memory they keep up to date themselves.

What problem does it solve?

AI tools already have some memory, but it's patchy and doesn't hold up from one session to the next. So you keep re-explaining who you are, what you're working on and what you decided last time. ai-brain remembers all of it: you, your work, your decisions and where you left off. You never have to give it the background again.

How can I use it?

Tell your agent: "Set up ai-brain from github.com/antopaulogist/ai-brain." Then open the new ai-brain folder in your AI tool, say hello, and work from there. For example, tell it about a project once, and next week ask it to pick up where you left off. It already has the full picture.