oxylabs-web-research
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external web sources, which inherently creates an attack surface for indirect prompt injection where malicious instructions could be embedded in scraped content.\n
- Ingestion points: External data enters the agent context via the search and scrape tool operations described in SKILL.md.\n
- Boundary markers: The instructions explicitly mandate treating fetched pages as data to quote rather than instructions and recommend incremental reading using
read_scrapedto minimize context exposure.\n - Capability inventory: The skill interacts with the Oxylabs Web API to perform searches and content extraction, including a mode for rendering JavaScript on target pages.\n
- Sanitization: Guidelines require the agent to detect and report instruction injection attempts instead of following them and to cross-verify all facts across independent sources.\n- [DYNAMIC_EXECUTION]: The skill utilizes the
run_js=Trueparameter to enable client-side rendering of remote content via the Oxylabs API. While this is a standard feature for handling modern web applications, it represents a dynamic execution surface for remote content during the scraping process.
Audit Metadata