seo-research

Warn

Audited by Snyk on Mar 19, 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). The SKILL.md explicitly requires the generated research prompt to gather and analyze untrusted public web content (e.g., "Recent statistics", "Competitor article analysis", and "Community questions (Reddit, forums)") and to cite URLs (see Step 2 and the "Research sections" / anti-hallucination instructions), so the agent will be instructed to fetch and act on third-party, user-generated sources.

Issues (1)

W011
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 19, 2026, 09:20 AM
Issues
1