research

Warn

Audited by Snyk on Feb 25, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 0.90). The SKILL.md for both research and fact-check explicitly instructs the agent to search, read, and verify claims against external sources (e.g., "Source Type Selection" listing Wikipedia/academic papers and "Phase 3: Source Verification" / "Claim Verification Log"), so it ingests untrusted public web content that can influence follow-up actions.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 25, 2026, 05:34 PM