publication-link-audit

Warn

Audited by Snyk on Aug 14, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (medium risk: 0.30). In SKILL.md the Publication Link Audit workflow extracts arbitrary URLs from the user-authored document and then ingests outsider-authored free text at runtime by fetching each Reddit URL’s <title> (and similarly reads HF/GitHub page content) to compare against surrounding labels.

Issues (1)

W011
MEDIUM

Third-party content exposure detected (indirect prompt injection risk).

Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 14, 2026, 07:57 AM
Issues
1
Security Audit — snyk — publication-link-audit