meta-ad-creative

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

• Third-party content exposure detected (medium risk: 0.30). The workflow reads “result.keyPoints” and “result.ad.copy” extracted from a real app store listing via the Appeeky API using user-supplied app_url, meaning outsider-authored listing text can be ingested into the LLM at runtime.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill sends an App Store URL (https://apps.apple.com/us/app/id123456789) to the generation/analyze API at runtime so the service will fetch the listing and extract content that directly shapes prompts/instructions.