web-scraper

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection (Category 8) due to the nature of web scraping.
  • Ingestion points: The skill fetches data from external, untrusted URLs via AsyncWebCrawler.arun in main.py and returns the resulting Markdown to the agent.
  • Boundary markers: The execute function in main.py returns the scraped content formatted as Markdown, but it does not wrap the content in protective delimiters or 'ignore' instructions to prevent the agent from obeying embedded commands.
  • Capability inventory: The skill has network access (scraping) and file-read capabilities through the underlying Playwright/Crawl4AI library.
  • Sanitization: While the skill uses PruningContentFilter to remove noise, this is a content-density filter and not a security sanitizer for malicious instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 04:19 PM