mega
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: 0.90). Yes — the skill ingests and uses open third-party content (e.g., exa:web_search_exa search results, infranodus:getGraphAndAdvice calls, and Scholar Gateway lookups in the mcp_integration/mega_mcp_workflow and tool_integration sections) as input to graph construction, gap analysis, and refinement, so untrusted web/social content would be read and interpreted by the agent.
Audit Metadata