news-sentiment-engine

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

• Third-party content exposure detected (low risk: 0.10). The required workflow ingests articles from public RSS feeds (TechCrunch, The Verge, Ars Technica, Hacker News), which are outsider-accessible publication sources categorized under low risk because the agent fetches specific selected feeds/URLs rather than monitoring an untrusted open submission queue.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs cloning and executing an unverified third-party repository from GitHub (https://github.com/tellmefrankie/news-engine), introducing an inbound operational dependency with weak/unknown provenance.