Originally, the run.files dictionary was specified to only include entries for files that were scanned in the course of the run. Since then, we have allowed the run.files dictionary to contain many other kinds of files, to enable us to embed the contents of those files within the log file:

• response files
• attachments
• the contents of the standard streams.

The problem is that now we don't have a way to identify, within the run.files dictionary, those files that were scanned. @michaelcfanning and I have discussed this in the past week or so but we neglected to file the issue.