osint-methods
Warn
Audited by Snyk on Mar 1, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.85). The SKILL.md explicitly directs collection and analysis of public, user-generated third‑party content—e.g., the "Social Media" section (LinkedIn, Twitter/X, Facebook, Instagram, YouTube, Reddit) and "Websites & Domains"/external tools—so the agent is expected to fetch and interpret untrusted web content as part of its workflow.
Audit Metadata