chrome-cdp
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The
scripts/start-chrome.shscript executes the Chrome or Chromium binary locally with debugging flags enabled. It utilizes environment variables likeCHROME_BINandCHROME_ARGSto configure the browser process, which is standard for local automation tools.\n- [DYNAMIC_EXECUTION]: Multiple scripts, includingscripts/eval.mjs,scripts/download.mjs, and the corescripts/cdp.mjsclient, utilize the Chrome DevTools ProtocolRuntime.evaluatecommand to execute arbitrary JavaScript code within the context of the browser tab. This is the primary mechanism for the skill's automation and inspection capabilities.\n- [INDIRECT_PROMPT_INJECTION]: The skill provides an attack surface for indirect prompt injection because it processes untrusted data from external websites that could contain malicious instructions.\n - Ingestion points: Untrusted data enters the agent context through page navigation in
scripts/goto.mjs, script dumping inscripts/dump-scripts.mjs, and console log capture inscripts/console.mjs.\n - Boundary markers: Absent. The instructions and scripts do not implement specific delimiters or instructions to ignore embedded commands within the processed web content.\n
- Capability inventory: The skill possesses powerful capabilities including executing arbitrary JavaScript in the browser (
scripts/eval.mjs), performing full network captures (scripts/net-capture.mjs), and triggering file downloads (scripts/download.mjs).\n - Sanitization: There is no evidence of sanitization or escaping of content retrieved from the web before it is passed to the agent or used in subsequent automation steps.
Audit Metadata