I prefer running two or three AI agents locally rather than offloading to remote agents. The feedback loop is faster, I can see what's happening, use the tools I like, and context switching between tasks is easier when everything is on my machine. My...

Markdown supported · ⌘↵ to post
Max Clayton Clowes
Thanks for this, it was really useful! Are you still using this solution as described here?
I was doing investigation myself and it didn't seem like the time constraint is a thing anymore, at least, which is great. I also don't love MCP and having the agents opt in, so I ended up building my own solution.
I'm interested in if you'd developed anything further on your end!