I agree with you. I don't see people hyping them and I think a big part of this is that we have sort of hit an LLM fatigue point right now. Also Skills require that your agent can execute arbitrary code which is a bigger buy-in cost if your app doesn't have this already.
Currently I've been working on https://terragonlabs.com which is a way to orchestrate Claude Code and other agents (Amp, Codex) as background agents.
I feel like I am locally constantly bouncing between different agents for different tasks and really wanted to be able to do the same in a remote environment.
When I worked for a large tech company I spent a few months building a similar service for internal use. I can confirm, managing all of those instances and making it reliable can be a huge headache. I really wish using browserless was an option for us