hallmark

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

• Third-party content exposure detected (low risk: 0.10). The workflow extracts design DNA from user-selected URLs or screenshots, representing low-risk active fetching of external web content.

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill executes an installer command fetching a third-party GitHub repository via npx, constituting a common unofficial operational dependency.