Jesse Vincent

Skills were released in Claude Code, what, yesterday? I doubt there's a simple answer to this -- it'll depend on the model, task, etc.

You could try to get your agent to test its own skills. From https://blog.fsck.com/2025/10/09/superpowers:

> As Claude and I build new skills, one of the things I ask it to do is to "test" the skills on a set of subagents to ensure that the skills were comprehensible, complete, and that the subagents would comply with them. (Claude now thinks of this as TDD for skills and uses its RED/GREEN TDD skill as part of the skill creation skill.)

> The first time we played this game, Claude told me that the subagents had gotten a perfect score. After a bit of prodding, I discovered that Claude was quizzing the subagents like they were on a gameshow. This was less than useful. I asked to switch to realistic scenarios that put pressure on the agents, to better simulate what they might actually do.

Actually, I _started_ from an ErgoDox. :)
( http://blog.fsck.com/2013/01/pinkies-and-your-brain.html )

The ErgoDox is really the two halves of a Kinesis flattened.

Keyboard.IO is a little bit different. There's a splay of a few degrees between the columns that map to each finger. A bunch of the work you do with your pinkies on the ErgoDox or just about any other keyboard moves to your thumbs. And yes, it ended up as a single keyboard. I really, really wanted to do something that was two separate pieces. Between the risks it would add to the mechanical design and the fact that I found myself intensely frustrating that the two halves of the ErgoDox were never in quite the places I expected them to be, we've decided to run with a single-piece unit, at least for the Model 01.

As a long time user, I'd recommend checking out https://github.com/obra/superpowers#the-basic-workflow. If that reasonates with you, try using it to develop a few features or capabilities.

It works very well for the way that I work (interactively and iteratively, not "one-shot"), and it helps me to better work in less time. Superpowers is one of the few skill/agent suites I use for all software development projects.

If you like building skill/agents, the posts at https://blog.fsck.com/ are a great resource for learning how to do well. The effectiveness of my project Axiom (a skill/agent suite for Apple OS developers) has benefited enormously from the knowledge that Superpowers' creator Jesse Vincent has been kind enough to share.

TLDR: You owe it to yourself to try it.

Most of these keyboards based on the ErgoDox (and the ErgoDox itself) make the mistake of considering that the thumb is a finger positioned like the others. And the keys you're supposed to hit with your thumbs are just placed in a crazy way (totally non-ergonomic IMHO).

To me one of the rare person who really "got it" is "Jesse" (he's both here, on GeekHack and on Deskthority) with its "butterfly" keyboard.

That's one of the only split (and ergonomic) keyboard that takes into account the fact that the thumb's opposition-apposition makes it unlike any other finger.

Look at his "butterfy" keyboard here and how smartly the keys you're to hit with the thumbs are positionned:

http://blog.fsck.com/2013/12/better-and-better-keyboards.html

https://raw.githubusercontent.com/obra/dotfiles/6e088092406cf1e3cc78d146a5247e934912f6f8/.claude/CLAUDE.md contains the following entry:

"- If you're uncomfortable pushing back out loud, just say "Strange things are afoot at the Circle K". I'll know what you mean"

Most of the rules seem rationale. This one really stands out as abnormal. Anyone have any idea why the engineer would have felt compelled to add this rule?

This is from https://blog.fsck.com/2025/10/05/how-im-using-coding-agents-in-september-2025/ mentioned in another comment