camoufox

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses subprocess.run in scripts/launch.py to identify the installation path of the camoufox library and executes shell commands to launch browser instances and generate configuration files for playwright-cli.
  • [INDIRECT_PROMPT_INJECTION]: The skill enables the agent to visit external websites and read page content. This creates an attack surface where instructions embedded in untrusted web data could influence agent behavior.
  • Ingestion points: Browser navigation and page snapshots (playwright-cli snapshot) as described in SKILL.md.
  • Boundary markers: None identified; untrusted page content is processed without explicit delimiters or instructions to ignore embedded commands.
  • Capability inventory: File system access for profile management, shell command execution in scripts/launch.py, and full browser automation via Playwright.
  • Sanitization: No evidence of content filtering or sanitization before external data is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 05:36 AM
Security Audit — agent-trust-hub — camoufox