news-sentiment-engine

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly says it "Collects news from 4+ RSS feeds (TechCrunch, The Verge, Ars Technica, Hacker News)," so the agent ingests open/public third‑party RSS content and uses it to drive summarization, sentiment, ranking, and briefing decisions—exposing it to possible indirect prompt injection from those sources.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill's setup explicitly runs remote code by cloning and executing the repository (git clone https://github.com/tellmefrankie/news-engine then pnpm dev), so fetched content from that URL would be executed at runtime and can directly control prompts/instructions.