broken-authentication

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONCREDENTIALS_UNSAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides pre-configured command-line strings for the hydra tool, enabling the agent to perform brute-force and credential stuffing attacks against specified targets.\n- [COMMAND_EXECUTION]: An embedded Python script is included to automate the collection of session tokens and analyze their entropy, requiring the execution of Python code and network interaction via the requests library.\n- [CREDENTIALS_UNSAFE]: The skill includes wordlists of common default credentials (e.g., admin:admin) and weak passwords to be used as payloads for authentication testing.\n- [SAFE]: All identified command execution and network operations are directly aligned with the skill's primary function as an offensive security testing guide and are documented with clear intent.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 03:18 PM
Security Audit — agent-trust-hub — broken-authentication