web-scraper

Pass

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill defines several Python scripts (fetch_page.py, search_web.py, crawl_dynamic.py, extract_links.py) intended to be executed in the agent's environment to perform scraping and search operations.
  • [EXTERNAL_DOWNLOADS]: The scripts perform network operations to retrieve content from external URLs provided by the user or found via search. Additionally, the crawl4ai-setup utility downloads browser binaries (Chromium) required for rendering JavaScript-heavy pages.
  • [PROMPT_INJECTION]: The skill processes untrusted content from the public web, creating a surface for indirect prompt injection where instructions embedded in scraped pages could attempt to influence the agent.
  • Ingestion points: Web content fetched via fetch_page.py and crawl_dynamic.py.
  • Boundary markers: The scripts wrap extracted content in Markdown headers and source indicators.
  • Capability inventory: Local script execution and network access.
  • Sanitization: The skill uses html2text to convert HTML to Markdown, which mitigates some direct HTML-based script execution risks but does not filter semantic instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 14, 2026, 09:20 AM