faster-chrome-devtools-skill

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill is installed from an external repository using npx skills add zeke/faster-chrome-devtools-skill.
  • [COMMAND_EXECUTION]: The skill relies on executing node scripts/cdp.mjs to perform a wide range of browser automation tasks, including navigation and interaction with elements.
  • [DATA_EXFILTRATION]: The skill can extract data from the browser context using commands like get-text, snapshot, and screenshot, which could potentially capture sensitive information if the agent navigates to protected or internal web pages.
  • [PROMPT_INJECTION]: The skill's architecture creates an indirect prompt injection surface by ingesting untrusted data from the web and providing high-privilege interaction capabilities.
  • Ingestion points: Browser page content, accessibility tree, and console logs accessed via scripts/cdp.mjs.
  • Boundary markers: No explicit delimiters or instructions to ignore embedded content are provided in the skill instructions.
  • Capability inventory: High-privilege actions including JavaScript execution (eval), navigation, element interaction, and raw CDP method calls.
  • Sanitization: The skill does not describe any methods for sanitizing or validating content retrieved from external websites.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 07:42 PM
Security Audit — agent-trust-hub — faster-chrome-devtools-skill