apify-influencer-discovery

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

• Third-party content exposure detected (high risk: 0.75). The required runtime path runs an Apify Actor and then fetches the resulting dataset items via Apify’s API (downloadResults() / displayQuickAnswer() calling https://api.apify.com/v2/datasets/${datasetId}/items?...), where the dataset content is derived from outsider-authored web/social text (e.g., Instagram/Facebook/TikTok bios, comments, captions) and is ingested into the LLM context when the “quick answer” prints those fields to chat.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill issues runtime API calls to https://api.apify.com (e.g., POST https://api.apify.com/v2/acts/{actorId}/runs and GET https://api.apify.com/v2/datasets/{datasetId}/items), which starts and polls Apify actors (causing remote code execution) and is a required runtime dependency (APIFY_TOKEN), so it directly depends on external code execution.