como-a-ia-avalia

Pass

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: No prompt injection or bypass attempts detected. The skill contains robust instructions for the agent to maintain a specific persona and explicitly forbids making promises, inventing data, or speculating on internal company criteria.
  • [DATA_EXFILTRATION]: No sensitive data access or exfiltration patterns detected. The skill only provides links to the vendor's official support and blog pages (gupy.io).
  • [COMMAND_EXECUTION]: No shell commands or system-level operations are present. The skill is entirely instructional/conversational.
  • [REMOTE_CODE_EXECUTION]: No external scripts, package installations, or dynamic code execution patterns were found.
  • [EXTERNAL_DOWNLOADS]: All external references are limited to official gupy.io domains for support and documentation purposes, which is consistent with the skill's purpose and author.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 25, 2026, 11:17 PM
Security Audit — agent-trust-hub — como-a-ia-avalia