rotate-scraping-proxies
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from untrusted external sources (URLs) which could contain malicious instructions for the agent. \n
- Ingestion points: The
fetcher.get(url)andfetch_with_rotation(url)functions inSKILL.mdretrieve content from arbitrary target URLs. \n - Boundary markers: No explicit prompt delimiters or markers are defined for the scraped content in the provided code snippets. \n
- Capability inventory: The skill performs network GET requests to retrieve web data using the
scraplinglibrary. \n - Sanitization: No sanitization or validation of the retrieved web content is shown before it is potentially returned to the agent context.
- [EXTERNAL_DOWNLOADS]: The skill fetches information from a remote service to verify network configuration and proxy status. \n
- Evidence: The skill makes requests to
https://api.ipify.orginSKILL.mdto confirm the proxy rotation is working. This is a well-known and established service for public IP verification.
Audit Metadata