Most folder-cleanup scripts move files first and let you discover the mess afterward. Point one at an old Downloads directory and two seconds later hundreds of PDFs and screenshots have disappeared into subfolders, sometimes overwriting an older file that happened to share the same name.
When I built Smart File Organizer, I built the workflow around three rules so I could trust it on my own machine:
- Dry-run preview first. Scanning a folder builds a table of proposed moves grouped by file type, extension, or modified month. Nothing touches the filesystem until you review and approve the plan.
- No overwrites on collision. Files on disk can change between scanning and clicking apply. Right before moving each file, the tool re-checks file identity and refuses to overwrite an existing destination path, either renaming the incoming file (
report (2).pdf) or skipping it. - Careful undo. Every applied batch is recorded in a local log. Clicking undo moves a batch back only if the moved files have not been modified since; if you opened and saved one of those documents afterward, it leaves your edited file alone and tells you why.
The duplicate finder follows the same habit: it groups identical files by size and SHA-256 hash and shows how much disk space they take up, without deleting anything automatically. Whether I am working on file tools, Automatic Report Generator, or checkout checks in Tiny Git, software that touches user files should always fail on the safe side.