Readur
Scanned receipts, phone photos of whiteboards, and forgotten PDFs turn into searchable text the moment they land in Readur. Drag files into the browser, point a watch folder at a scanner inbox, or connect WebDAV, S3-compatible storage, or a local directory that syncs on its own schedule. OCR runs in a background queue that finishes small files first, covers over 100 languages, and records a confidence score and processing time for every document. The home screen tracks the processing pipeline, flags documents that need attention, and lists the newest arrivals as thumbnails. Search accepts plain keywords, quoted phrases, fuzzy matches such as invoice~ for sloppy scans, and boolean queries, then narrows results by type, collection, source, date, or status. Open any document in a drawer that shows the original above its extracted text, then add colored labels, leave threaded comments, or create a share link protected by a password, an expiry date, and a view limit. Smart collections keep saved searches current as new files arrive. Admins create accounts and assign roles, anyone can issue personal API keys for scripts, and an OIDC provider such as Keycloak, Auth0, or Azure AD can handle sign-in. The interface speaks English, German, Spanish, and French, in light and dark themes. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.
Deploy