apify-competitor-intelligence

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

• Third-party content exposure detected (high risk: 0.85). The skill runs reference/scripts/run_actor.js, which fetches Apify dataset items (free-form scraped text from third-party sources) via https://api.apify.com/v2/datasets/${datasetId}/items?... and then prints them to chat in quick-answer mode or writes them to CSV/JSON—this scraped content is outsider-authored.

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

• Potentially malicious external URL detected (high risk: 0.80). The scripts call Apify API endpoints at runtime (e.g., https://api.apify.com/v2/acts/... and https://api.apify.com/v2/acts/.../runs?token=..., plus dataset URLs like https://api.apify.com/v2/datasets/.../items?token=...) to fetch actor schemas/details and to start actor runs, which executes remote actor code and are required for the skill to function.