webwright-legal-pl

Pass

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches the webwright utility from GitHub and installs the Playwright library from the standard package registry.
  • [COMMAND_EXECUTION]: The skill utilizes shell commands to configure its environment and execute Python scripts that automate browser interactions.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes text content from external legal and government portals.
  • Ingestion points: Web content is retrieved via Playwright's inner_text method in scripts/fetch_eurlex.py and scripts/fetch_orzeczenie.py.
  • Boundary markers: No explicit delimiters or guardrail instructions are included when storing the fetched raw text content.
  • Capability inventory: The skill has the ability to read and write files locally and execute bash commands.
  • Sanitization: Specific metadata fields are extracted using regular expressions, while the full body text is saved to markdown files without further sanitization.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 14, 2026, 08:28 AM
Security Audit — agent-trust-hub — webwright-legal-pl