expressen-se

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

• Third-party content exposure detected (low risk: 0.10). The agent reads articles and search results from expressen.se, which is a public news site where outsider-authored text can appear in articles and comments, representing low-risk ingestion since the agent actively searches or opens specific pages.

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 an external remote MCP server and execute queries via third-party service Unbrowse (unbrowse.ai), representing an unverified third-party operational dependency.