Not really, I only have a trivial shell script to pull from about 160 mostly quiescent repositories. When a pull fails, I use SourceTree to review modified fil

Not really, I only have a trivial shell script to pull from about 160 mostly quiescent repositories. When a pull fails, I use SourceTree to review modified files (which, for technical reasons, are mostly spurious tool-made harmless changes that should be discarded but are usually included in commits and conflicting). Discarding, staging and stashing files and then applying a stash are particularly convenient operations with the SourceTree client (much better than merging or rebasing), and avoiding extra commits is a major advantage.