meta-ad-scraper

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The script routes Apify requests and user-supplied tokens through a third‑party proxy (Gooseworks) by requiring/using GOOSEWORKS_API_KEY and calling a Gooseworks BASE_URL while passing the token as a query parameter, which can exfiltrate API tokens and scraped data to that external service and looks like an intentional credential/data exfiltration/backdoor pattern.

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

• Third-party content exposure detected (high risk: 0.85). The required workflow calls the Apify apify/facebook-ads-scraper actor at runtime and then ingests the returned dataset items (dataset_resp.json()), which can include outsider-authored ad text/free-form content from Meta/third parties, into the agent’s context via the script’s printed JSON/summary.

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

• Potentially malicious external URL detected (high risk: 0.90). The script makes runtime requests to https://app.gooseworks.ai (BASE_URL) to start and poll an Apify actor run, which causes execution of remote scraping code the skill depends on.