skills/openclaw/skills/tor-browser/Gen Agent Trust Hub

tor-browser

Pass

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The setup instructions recommend the use of sudo for installing the Tor service and managing its lifecycle on the host system.
  • [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection by processing untrusted data from the web.
  • Ingestion points: The skill retrieves content from external web sources (including .onion sites) via the navigate and get_snapshot methods in scripts/tor-browser.py.
  • Boundary markers: Extracted content is not wrapped in specific markers or accompanied by warnings to ignore instructions embedded in the data.
  • Capability inventory: The script allows for browser interactions like clicking and form filling, which could be manipulated by malicious web content.
  • Sanitization: There is no evidence of sanitization or filtering of the text and element metadata extracted from web pages.
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the download of browser binaries from the official Playwright repository managed by Microsoft during the environment setup process.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 11, 2026, 01:20 PM