// Security Advisory Triage
pypdf memory exhaustion in Roman numeral page labels, fixed in 6.17.0
GitHub advisories — pip · published 2026-10-02 21:55 UTC
WHAT CHANGEDCVE-2026-102993 published 1 October 2026 for pypdf versions before 6.17.0. High severity memory exhaustion vulnerability triggered when retrieving Roman numeral page labels from PDFs.
WHY IT MATTERSProcessing a malicious PDF can exhaust available memory, causing service crashes or denial of service in any application that handles user-uploaded PDFs with pypdf.
WHAT TO DOUpgrade pypdf to 6.17.0 or later. If you process PDFs from untrusted sources, prioritize this update for this week's deployment cycle.
Model-reported confidence 85% (not measured accuracy)
Primary source ↗
written by anthropic/claude-sonnet-4.5
Show the reasoning — raw diff, triage verdict, confidence
RAW DIFF
+ghsa_id: GHSA-qv6h-rv94-w285 +summary: pypdf: Possible large memory usage when retrieving Roman page labels +severity: high +published_at: 2026-10-01T15:03:45Z +cve_id: CVE-2026-102993 +vulnerabilities[].package.name: pypdf +vulnerabilities[].first_patched_version: 6.17.0
HOW IT WAS JUDGED
Triage signals below are model-reported assertions, preserved verbatim; they are not independent arithmetic verification.
| triage verdict | MATERIAL at 0.78, needed 0.5 |
|---|---|
| triage reasoning | New advisory with fixed version for memory DoS in widely-used PDF library. |
| signals | known fixed version published · high severity rating · memory exhaustion vulnerability · pypdf is a common PDF processing dependency |
| suppression | — |
| triage model | anthropic/claude-haiku-4.5 ($0.00000) |
| writer model | anthropic/claude-sonnet-4.5 ($0.00000) |