vp-chrome-profiles
Pass
Audited by Gen Agent Trust Hub on May 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a local helper script 'scripts/chrome-profilectl' to manage profile lifecycles. The script implements robust validation for profile names and ports to mitigate command injection risks.
- [EXTERNAL_DOWNLOADS]: The skill references 'chrome-devtools-mcp@latest' via npx, which fetches and executes the package from the official NPM registry. This is a standard procedure for MCP-based browser control.
- [DATA_EXFILTRATION]: The helper script uses curl to probe the local Chrome DevTools debugging endpoint (127.0.0.1) for connectivity checks. No non-whitelisted external network traffic was identified.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection when interacting with external websites. 1. Ingestion points: Profile names and target URLs. 2. Boundary markers: The skill relies on natural language instructions for safe browsing rather than formal data delimiters. 3. Capability inventory: Includes shell command execution, file management, and local network probing. 4. Sanitization: Profile names and ports are strictly validated using regular expressions.
Audit Metadata