review-scraper

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

• Third-party content exposure detected (high risk: 0.95). The required runtime path fetches outsider-authored review text from public review sites via Apify (e.g., run_apify_actor() calls Apify dataset items, then normalize_review()/filter_reviews() ingest review["text"]/reviewBody into the agent’s context as output), which is free-form text authored by third parties.

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

• Potentially malicious external URL detected (high risk: 0.80). The script makes runtime requests to the Gooseworks Apify proxy (https://app.gooseworks.ai/v1/proxy/apify) to start platform-specific Apify actors (e.g., zen-studiog2-reviews-scraper, agentstrustpilot-reviews), which executes remote scraping code and is required for the skill to function.