last30days
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). This skill explicitly fetches and ingests public user-generated content from Reddit, X (Twitter), and arbitrary web pages (via the last30days script and WebSearch), so the agent will read and synthesize untrusted third-party content that could carry indirect prompt injections.
Audit Metadata