demand-mining
Warn
Audited by Snyk on Feb 15, 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 explicitly instructs using WebSearch and WebFetch to scrape and ingest user-generated content from open/public sites (e.g., Reddit, X/Twitter, Hacker News, Zhihu, 小红书) and to read/quote those posts as part of its analysis, exposing the agent to untrusted third‑party content.
Audit Metadata