I'm testing this tool and it looks great so far, thank you for creating it!

The main reason why I've decided to try it is because I work with a huge codebase (approx. 17k files) and I hit context compaction frequently enough. I'm trying to avoid it by spawning sub-agents that should implement features in chunks, and I'm succeeding with it to some extent.

But sub-agents are kinda blackbox, pretty hard to check what they do and whether they hit the compaction or not. Your tool does a great job in revealing the internals of the sub-agent sessions, and it helps a lot in understanding what's under the hood.

Tokens for specific operations are greatly helpful, and I'd like to ask you to add the total amount of token context a subagent allocates for itself, if it's technically possible to track that. As well as the compaction note for subagents (if it's not done yet).

I've just started using the tool, but already recommend it to some of my colleagues. Great job indeed!