noverload
Warn
Audited by Snyk on Feb 13, 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 ingests and displays full content from untrusted public sources—e.g., saved YouTube videos, articles, Reddit threads, X/Twitter posts and arbitrary URLs (see "Content Types Supported", "Save New Content", and "Get Full Content")—which the agent is expected to read and could carry indirect prompt injections.
Audit Metadata