The Agent Skills Directory

Indirect Prompt Injection (SAFE): The skill is designed to ingest data from external websites, which is its primary purpose. While this exposes the agent to untrusted data, no malicious usage of this capability was found.
Ingestion points: url parameters in scrapeProducts and parseArticle (SKILL.md).
Boundary markers: None explicitly defined in the code snippets.
Capability inventory: Network requests and browser automation.
Sanitization: The skill performs standard HTML parsing and data cleaning (trimming, type conversion).

web-scraper