interactive-learner

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's workflow explicitly fetches and ingests public web content — e.g., the find-videos.py script "scrapes YouTube HTML" and the research phase instructs the agent to gather resources from community forums, blogs, and arbitrary URLs (play-with-k8s.com, Stack Overflow, Reddit, etc.) which the agent is expected to read and incorporate into lesson content.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 13, 2026, 07:39 PM