rebuttal-response-skills

Warn

Audited by Snyk on Aug 24, 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). The runtime workflow ingests the user-supplied “complete review text” and “submitted paper and supplement” as the review source of truth, including reviewer sentences and evidence locations provided via the same input, so outsider-authored free text can be directly consumed at generation time.

Issues (1)

W011
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 24, 2026, 02:18 AM
Issues
1