News for hackers. Stuff that matters.
Re:Clean up Claude 5's token vomit with a separate LLM by dahdum on Thursday August 20, 2026 @02:19PM (#49378234)
> it's speaking its own dialect, and you get used to it

Same experience. It’s not very “human” but once you have agents talking to each other the shared dialect and verbosity makes things much smoother in my experience. Fighting against the default feels like an uphill battle with no meaningful benefit.

Re:Clean up Claude 5's token vomit with a separate LLM by rudiksz on Thursday August 20, 2026 @03:32PM (#49379072)
"agents talking to each other"? Are you for real dude?
Re:Clean up Claude 5's token vomit with a separate LLM by SyneRyder on Thursday August 20, 2026 @06:18PM (#49381037)
If one agent is spawning several subtask agents to work in parallel, they need to be able to communicate. That's built-in to Claude Code now.
Re:Clean up Claude 5's token vomit with a separate LLM by dahdum on Friday August 21, 2026 @06:18AM (#49386058)
Yes, I run continuous hybrid agent loops with message dispatch, mcp tools, and subagent capabilities. IMHO it’ll be table stakes in a couple years.