external-pattern-integrator

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

• Third-party content exposure detected (low risk: 0.10). The skill fetches external sources via user-provided URLs or local files at runtime, which constitutes active fetching of outsider-authored content.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill processes arbitrary external URLs supplied at runtime as arguments via WebFetch or curl to integrate external patterns, introducing an unverifiable third-party remote dependency tier.