drive-chrome-cdp

Pass

Audited by Gen Agent Trust Hub on Sep 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONCREDENTIALS_UNSAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill operates by executing the chrome-cdp CLI tool to communicate with the Chrome DevTools Protocol.
  • [DYNAMIC_EXECUTION]: Features an eval command that allows for the execution of arbitrary JavaScript within the browser's top frame, providing a mechanism for runtime code execution.
  • [CREDENTIALS_UNSAFE]: The skill provides tools to access sensitive browser data, including cookies and local storage. While it implements default redaction for credential-like strings, the accessibility of active session data is a sensitive capability.
  • [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it ingests untrusted data from web pages and possesses high-privilege capabilities to interact with those pages. * Ingestion points: Content is retrieved using snap, find, grid, text, html, and text --article. * Boundary markers: The instructions explicitly warn the agent to treat page content as data and not instructions. * Capability inventory: The skill can perform actions such as click, type, fill, select, eval, and upload. * Sanitization: Implements automated redaction for credential-shaped values in network and storage outputs.
  • [EXTERNAL_DOWNLOADS]: References the sanketsudake/chrome-cdp-cli repository on GitHub for binary installation, which is a resource owned by the skill author.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 1, 2026, 08:53 AM