6,918 Paperless matches: a document archive that was meant to replace the filing cabinet

Paperless-ngx scans paper documents, runs optical character recognition over them, and stores the result as searchable files. Teams adopt it to stop losing invoices and contracts, which means the archive ends up holding exactly the records a company must protect.
ZoomEye finds 6,918 assets whose HTML title contains "Paperless". The query used English double-quoted values and ran in all scope. Some of those matches are unrelated projects that share the word, so the Paperless-ngx share of the total is smaller.

Why the archive is sensitive by construction

A document archive is an identity record in disguise. Invoices carry bank details and addresses. Contracts name signatories. Employment paperwork includes national identifiers. Once scanned and indexed, all of it becomes searchable text with a single query box in front of it.
The application also stores metadata that is easy to overlook: correspondent names, tags, and the dates attached to every file. That metadata describes relationships, budgets, and timelines even when the document images are never opened.

Access model and common mistakes

Self-hosted deployments usually authenticate against a local user table, and many rely on a reverse proxy for TLS. The frequent mistakes are familiar. Accounts are created for family members or temporary staff and never removed. The archive sits on a public hostname because remote access was convenient. Backups are written to the same volume, so a compromise takes the copies too.
Older releases of the project have addressed authentication and file-handling issues, and a deployment that has not been upgraded in a year carries the fixes it skipped.

A review that fits the tool

Establish whether the archive must be reachable without a VPN. For most households and small teams the answer is no, and closing the port removes the risk entirely.
Then confirm the version, enable multi-factor authentication where supported, and review the user list. Export a copy of the index and store it separately, so that a lost archive does not also erase the record of what it contained.

Measurement

The count is a starting point for scoping; an operator can restrict the query to their own address space and verify that nothing answers unexpectedly. The link reproduces the query used here.

References

• Paperless-ngx documentation: https://docs.paperless-ngx.com/
• ZoomEye search, query title="Paperless", scope all, retrieved 2026-10-03, count 6,918