Wahyu

Is it just me, or do agents always seem to push back whenever I ask them to assess a large migration task? I’m talking about things like:

“Vet the feasibility of replacing framework X with Y.”
“Can we use validation library B instead of A?”

But AFTER I tell them to just do it and then assess the result, they always say it’s a net positive. :/

honestly how do people manage multiple agents concurrently? I am already having trouble keeping track of 3 threads of agents working. My brain is hitting context compaction way too often.

I feel like jev can really be used as agents tool selection to a point the ast-grep / rg / codegraph etc will build a mini digest to equip the agent to complete the task end to end

i have always wanted the meta quest to work with muse, can you imagine being a god? just prompt using your voice and worlds around you get built and torn?

should metamucil sue meta's muse bot for trademark infringement?