guia-de-triagem
Pass
Audited by Gen Agent Trust Hub on Jul 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely instructional and intended for human-like conversation. It does not contain any executable scripts, shell commands, or tool invocations.
- [EXTERNAL_DOWNLOADS]: The skill provides links to official Gupy domains (gupy.io) for support and educational blog posts. These are legitimate resources related to the skill's primary purpose and do not involve downloading or executing remote code.
- [DATA_EXFILTRATION]: There is no evidence of credential access, file system reads, or unauthorized network operations. The skill explicitly instructs the AI not to invent or disclose sensitive candidate data.
- [PROMPT_INJECTION]: The instructions include robust 'inegociável' (non-negotiable) rules that reinforce safety and transparency, preventing the AI from making false promises or confirming biases.
Audit Metadata