It would be great to show the estimate before the review:
- number of files to review
- estimated number of completion requests
- $ estimate (
input token count / inference price)
- ...
It would also be great to show the statistics after the review:
- number of files reviewed
- percent of files with issues
- number of completion requests made
- number of input/output tokens generated
- average file review time
- $ spent
- $ per issue discovered
- ....