udemy

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

• Third-party content exposure detected (low risk: 0.10). The agent reads Udemy course pages via an MCP server, which exposes it to course descriptions and content authored by third-party instructors (low risk as it requires active search/fetch of specific courses).

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill instructs the agent to connect a remote MCP server hosted at unbrowse.ai, which provides executable tools and behavior-defining configuration for runtime use, representing an unofficial third-party operational dependency.