apify-ultimate-scraper

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

• Third-party content exposure detected (high risk: 0.85). The skill runs Apify Actors that scrape third-party web/social content at runtime (e.g., apify actors call ... then apify datasets get-items ... --format json), and those datasets can include outsider-authored free text (captions, comments, review bodies, page text) that is then ingested into the agent’s LLM context for “quick answer mode”/summarization.

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

• Potentially malicious external URL detected (high risk: 0.90). This skill makes runtime calls to the Apify platform (via apify CLI commands such as "apify actors info" and "apify actors call") and references console endpoints like https://console.apify.com (e.g., https://console.apify.com/settings/integrations and https://console.apify.com/actors/runs/RUN_ID/log); those runtime fetches of Actor READMEs/schemas and the apify actors call execute remote Actor code and thus directly control agent behavior.