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.mdthe 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
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata