How this tool computes its result
The extractor reads PDF link annotations from every page and classifies each target as external HTTP, email, or internal. Identical targets are grouped into one record containing page numbers and occurrence count. The optional status check sends at most 20 unique external URLs, never the PDF, to a protected server route. That route rejects local and private network destinations, restricts ports, validates DNS results, follows at most four validated redirects, tries HEAD first, falls back to a byte-range GET when HEAD is unsupported, and classifies the final response as reachable, redirected, restricted, broken, blocked, or unverified.
