Improved stats output

[adityasingh345]

Sir now lx can estimate total token count, I was wondering if it would make sense to add features like :-

Two small, opt-in ideas:

1. Token summary block
   A short footer showing:
   • number of files
   • total lines
   • estimated tokens

This would stay fully deterministic and optional, and help users
budget LLM context more precisely.

Happy to work on this if it aligns with lx’s direction.

[rasros]

It is a great idea, very nice! Thank you for the suggestion and willingness to do it.

Give me a day or two to finish merging v2 into main and the repo will be in a good state for this. I'll let you know and give some input on how to do the feature.

[adityasingh345]

Okk sir

[rasros]

@adityasingh345 v2 is merged now.

The feature here is fairly straightforward to implement, depending on what you want.

There is a new debug mode activated when output is written to clipboard -c or to file -o context.md. You can also activate it with -v or disable it with -q. It always writes to stderr. If this is sufficient just improve the template used to render the debug output. The template is called DefaultDebugTemplate in template.go.

If you want a header/footer to always display in stdout then it's a bit more involved but not hard. Just add two more optional templates in template.go and to config.go, one for header and one for footer. I think default should be empty, so you have to configure config.yaml to use it.

The new functionality is not released yet, I need to complete some minor tasks first.

[adityasingh345]

Thanks a lot for the detailed explanation!

That makes sense. I’ll start by improving DefaultDebugTemplate to include a clearer token summary (total + per-file breakdown) in the debug output, and open a PR for that.

If needed later, we can explore the header/footer approach, but I’ll keep this iteration focused and minimal.