web-research-workflow

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

• Third-party content exposure detected (low risk: 0.10). The workflow ingests content via WebFetch, Tavily API, and browser scraping from arbitrary URLs and competitor sites selected by the user, representing low-risk active fetching of external content.

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill references Tavily API endpoints (https://api.tavily.com/*) as operational search and extraction tools controlled by runtime environment variables, making Tavily a common/reputable unofficial operational dependency.