helium-browser

Pass

Audited by Gen Agent Trust Hub on Oct 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: Provides shell, PowerShell, and Node.js commands to programmatically read the browser's DevToolsActivePort file from the local filesystem to establish a debugging session.
  • [DYNAMIC_EXECUTION]: Includes a Node.js verification script executed via the CLI (node -e) that reads local configuration files and opens a WebSocket connection to the browser's debug port.
  • [INDIRECT_PROMPT_INJECTION]: By facilitating agent control over an active browser session, the skill creates a pathway for the agent to process untrusted web content, which is a known vector for indirect prompt injection attacks.
  • [EXTERNAL_DOWNLOADS]: The configuration examples utilize npx to fetch and execute the chrome-devtools-mcp package directly from a public registry.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 7, 2026, 09:01 PM