The latest tagged 0.x release receives fixes. Older tags are not patched.

Scribe parses untrusted documents (PDF and ZIP/EPUB containers), so parsing bugs with security impact — memory unsafety in the zip reader, crashes on crafted inputs, path traversal via archive entry names — are in scope and taken seriously.

Please do not open a public issue for security reports. Use GitHub's private vulnerability reporting ("Report a vulnerability" under the Security tab), or email kegbenk@gmail.com with a description and a reproducing input if you have one.

You'll get an acknowledgment within a week. Fixes ship as a patch release with credit unless you prefer otherwise.

• Extraction performs no network I/O (enforced in CI by tools/privacy-audit.sh).
• The zip reader rejects encrypted entries and zip64 archives rather than attempting partial parses.
• Archive paths are normalized (./.. components resolved) before lookup; entries are read from the central directory, never extracted to disk.