noise2blog

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

• Third-party content exposure detected (low risk: 0.10). The agent reads arbitrary web pages via WebFetch when the user provides a URL.

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill executes a curl request to https://api.tavily.com/search to fetch research data, but Tavily is a third-party commercial platform without strong first-party provenance for this project, establishing a common/reputable unofficial operational dependency.