research
Warn
Audited by Snyk on Feb 16, 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's Information Gathering phase explicitly instructs using web searches (preferencing a gemini web search command and a
WebSearchfallback) and adocs-seekerfor GitHub repos, meaning the agent will fetch and read arbitrary public web pages and user-generated GitHub content as part of its workflow.
Audit Metadata