Stirling-PDF puts 50+ document tools on your own server

A tutorial published on dev.to walks through running Stirling-PDF as a private, browser-based document workbench, replacing the common habit of uploading contracts and scans to whatever free PDF site appears first in search results.

The pitch is simple. According to the dev.to walkthrough, the software bundles more than 50 PDF operations and runs entirely on your own server, so files never leave your control. At its core the application is stateless: documents are processed and returned without being stored permanently, and only configuration plus OCR language data persist in mounted volumes.

Deployment with Docker and Traefik

The setup follows a familiar self-hosting recipe. Stirling-PDF v2.14.2 runs as a single Docker Compose service, placed behind Traefik as a reverse proxy with TLS handled through the configured certificate resolver. The official image includes its own health check, and Traefik forwards traffic only once the container reports healthy — a window of roughly 20 to 30 seconds after start while the Java application warms up.

That statelessness also keeps operations light. The author notes that backups amount to the configuration folder and any custom OCR data, because processed PDFs are not retained on the server.

Login on by default

Version 2 ships with authentication enabled, which the tutorial flags as a meaningful difference from many comparable tools. Initial sign-in uses the well-known default credentials admin/stirling, and the application immediately forces a password change on first login. The advice is blunt: do not weaken or remove the login on a publicly reachable instance, because an open deployment would hand every document that passes through it to anyone who finds the domain.

What the toolbox covers

The interface shows uploaded files in a built-in viewer, with the tool catalog grouped into categories such as signing, document security and verification. The tutorial highlights the operations that come up most in practice:

• Merging multiple PDFs or splitting out individual pages
• Compression for emailing large documents
• Conversion between PDF and image formats, and between Office documents and PDF
• OCR that makes scanned pages searchable and can clean up scans
• Redaction that permanently removes sensitive content rather than covering it with a black box
• Passwords and watermarks

OCR support leans on Tesseract. The stock image includes common languages, and additional ones can be enabled by dropping the matching .traineddata files into a mounted folder.

Caveats worth knowing

Some limits emerge at the edges. Office-to-PDF conversion depends on LibreOffice, which is only present in the larger fat variant of the image. OCR and format conversion are memory-intensive, so a small virtual server can struggle with large files. On maintenance, the author recommends periodically bumping the image tag, since the project is actively developed and new tools arrive regularly.

The tutorial also positions Stirling-PDF as a companion to document setups like Paperless or Nextcloud — a place to prepare files before or after they enter those systems.

Why it matters

Free online PDF utilities are a quiet privacy hazard: people routinely feed contracts, IDs and financial documents into services with unclear retention and handling policies. A self-hosted toolbox removes most of that exposure, and the stateless design means the server is not quietly building an archive of your paperwork. The deployment cost is one container and a subdomain, and because processing happens locally, no document data needs to leave your network at all. For anyone already running Docker and a reverse proxy, Stirling-PDF is close to a drop-in replacement for the search-engine roulette of free PDF sites — with the admin password and OCR languages being the details worth configuring carefully.