I’m getting really sick of reading Claudish in my docs.
Does anyone have a skill that’s like a mix of:
- Hemingway editor
- Grammarly
- Microsoft spellcheck
- Weird phrase/analogy detector (“smoking gun!”, “load-bearing”, “it’s not X, it’s Y”, meaningless statements)
@dexhorthy@0xblacklight I feel like gstack is a major culprit when it comes to Claudish haha (“that’s the most insightful thing you’ve said in …”)
This skill looks sick checking it out!
When I wrote this last year, things were way different.
Models were awful, but they were basically free.
Since then, I've changed much of my thinking around hyper engineering. (I'm no longer a tokenmaxxer, I'm basically a tokenminimzer!)
Hope to share another article soon
@cheneypiano Ideally API first. MCPs are more trouble than they're worth
Frontend exists because I don't want to have to burn tokens every time I want to see a data visualization
Introducing the open-source @kortix CompanyOS:
Your entire company as code: agents, skills, connectors and rules in one repo you own. A workspace for your team, a framework for your agents.
Sovereign, multiplayer, extensible AI.