seo-geo
Warn
Audited by Snyk on Apr 14, 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 explicitly instructs the agent to fetch and analyze public site content (e.g., checking robots.txt, looking for /llms.txt, assessing passage-level citability and server-side rendering of target web pages) and even suggests using DataForSEO scrapers (ai_optimization_chat_gpt_scraper, ai_opt_llm_ment_search) to read ChatGPT/web search results and platform mentions (Reddit, YouTube, Wikipedia), which are untrusted third‑party sources that the agent will read and use to drive recommendations.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata