wikipedia

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

• Third-party content exposure detected (low risk: 0.10). The agent reads Wikipedia pages and search results via an MCP server, which represents user-initiated searches and article reads (public reference content accessed via search/fetch).

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill connects to a third-party remote MCP server and REST API hosted on unbrowse.ai, which provides the operational tools and behavior-selecting configuration, representing a common/reputable unofficial operational dependency.