thousandflowers

I'm working on something similar for climbing, but it's quite complex without gyroscopes at various points. How did you solve this problem?

Really nice project!

It will be intresing an automation that check online for word like “worker cooperative”, “worker-owned” to find new candidate that fit.

Ofc the hard part is confirming if it's genuinely worker-owned, so keeping that step human makes sense. But the discovery/shortlisting could be largely automated crawl the web, score candidates by how many co-op signals they hit, and present the top ones for you to vet.

Would be happy to help prototype something like this if you're interested (i really like your idea).

The thing that most disconcerts me isn't the runtime pruning, it's the cold loading. Months ago, I added a few skills and MCPs to test them, partly in the frenzy of free shopping, but then I forgot about half of them.

So after I got tired of choosing by hand, and therefore also a bit blindly, I created a small tool that runs locally and analyzes conversations to tell you which skills, MCPs, or other things are always unused.

347 items never used · ~19354 dead tokens/session · ~$25.49/month A lot of ECC that I never used but always loaded.

If anyone's interested, I've put it on GitHub, thousandflowers/skillreaper.