post-to-campaign

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

• Third-party content exposure detected (medium risk: 0.30). The skill workflow involves reading a LinkedIn post via URL/text, but LinkedIn post contents and comments are authored by third-party platform users rather than trusted first-party administrators, representing an indirect prompt injection vector.

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

• Potentially malicious external URL detected (medium risk: 0.30). An operational dependency on an external Vercel skills repository is evidenced in the installation instructions (npx skills add LaGrowthMachine/gtm-system/...), presenting a third-party unofficial runtime/setup dependency.