fix-review

Warn

Audited by Snyk on Feb 17, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 1.00). This skill explicitly fetches and parses external reports from arbitrary URLs and Google Drive links (see "Phase 2: Report Retrieval" and the Quick Reference where it uses the WebFetch tool and gdrive fallback), so the agent will read untrusted third‑party content that could carry indirect prompt injection.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 17, 2026, 10:26 PM