helium-browser

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes shell commands (sed) and Node.js (node -e) to parse connection details from the DevToolsActivePort file, enabling tools to identify the dynamic port and WebSocket path for the active session.
  • [EXTERNAL_DOWNLOADS]: The setup instructions recommend using npx -y chrome-devtools-mcp@latest, which downloads and executes the latest version of the tool from the public NPM registry.
  • [CREDENTIALS_UNSAFE]: The skill provides instructions to access the browser's user-data directory and read DevToolsActivePort. This file contains the connection parameters (port and secret path) required to control the browser session.
  • [PROMPT_INJECTION]: The browser automation capabilities configured by this skill create an attack surface for indirect prompt injection from web content.
  • Ingestion points: Untrusted content from web pages accessed via the connected browser tools (SKILL.md).
  • Boundary markers: None provided in the connection setup instructions.
  • Capability inventory: Full control over the browser session (DOM, cookies, network) via Playwright and Chrome DevTools (SKILL.md).
  • Sanitization: No specific sanitization or filtering is defined for the external content processed by the tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 10:35 PM
Security Audit — agent-trust-hub — helium-browser