skills/lev-os/agents/auth-sniffer/Gen Agent Trust Hub

auth-sniffer

Fail

Audited by Gen Agent Trust Hub on Mar 21, 2026

Risk Level: HIGHCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a custom CLI tool, auth-sniff, to perform intrusive system-level operations. This includes identifying running browser processes and forcing their termination using SIGTERM and SIGKILL loops to allow the browser to be relaunched with the Chrome DevTools Protocol (CDP) enabled.
  • [DATA_EXFILTRATION]: The core functionality of the skill is the programmatic extraction of sensitive authentication material from browser profiles. It specifically targets high-value credentials such as auth_token, ct0, and other session-related keys from cookies, localStorage, and sessionStorage. This data represents active login sessions and could be used for session hijacking or unauthorized account access.
  • [CREDENTIALS_UNSAFE]: The skill is designed to automate the discovery and retrieval of private session identifiers and credentials directly from the user's local browser environment, focusing on specific domains like .x.com.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Mar 21, 2026, 06:48 PM