creador-de-habilidades

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • Prompt Injection (SAFE): The skill contains natural instructional language for creating metadata and documentation. No override markers or jailbreak attempts were detected.
  • Data Exfiltration & Exposure (SAFE): No hardcoded credentials, sensitive file paths, or network communication patterns (curl, wget) are present.
  • Remote Code Execution (SAFE): The skill does not download or execute external scripts. It mentions a 'scripts/' directory as a structural recommendation but does not provide or invoke any code.
  • Persistence & Privileges (SAFE): No commands related to system persistence, sudo usage, or permission modifications were found.
  • Indirect Prompt Injection (LOW): While the skill is designed to process user-generated content (new skill definitions), it acts as a generator/assistant and does not automatically execute external code or scripts from untrusted sources in this configuration.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:50 PM