content-monitor

Warn

Audited by Snyk on May 18, 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 fetches untrusted public content (e.g., discover-reddit.js calling https://www.reddit.com/r/.json, discover-github.js scraping github.com/trending, and discover-rss.js ingesting YouTube RSS, Hacker News, blogs, arXiv, etc.) which the Extract/Summarize/Evaluate pipeline consumes and uses to produce actionable digests in .content-monitor/digests/, so third-party content can materially influence agent decisions and actions.

Issues (1)

W011
MEDIUM

Third-party content exposure detected (indirect prompt injection risk).

Audit Metadata
Risk Level
MEDIUM
Analyzed
May 18, 2026, 04:07 PM
Issues
1