research-docs

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

• Third-party content exposure detected (low risk: 0.10). The workflow parses and reads local user-supplied documents from a data directory ($0), which can contain untrusted external text files or parsed content.

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

• Potentially malicious external URL detected (medium risk: 0.30). The templates/report.html file contains external CSS font stylesheets loaded from Google Fonts and jsDelivr CDN during runtime rendering, constituting an unverified third-party operational dependency.