morning

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

• Third-party content exposure detected (medium risk: 0.30). The skill workflow monitors and ingests outsider-authored emails, chat messages, and document comments from connected communication channels.

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill contains instructions to fall back to running npm pack @fontsource/fraunces to fetch and extract font binaries from the npm registry if local assets are missing, constituting a common unofficial operational dependency.