prompt-validate
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a non-executable utility designed solely for text analysis and prompt auditing. It does not perform network operations, file system modifications, or command execution.
- [DATA_EXPOSURE]: The skill includes explicit security instructions to detect and redact sensitive data (API keys, PATs, passwords, etc.) using a
[REDACTED]placeholder. This mitigates the risk of credential leakage during the prompt auditing process. - [PROMPT_INJECTION]: While the skill analyzes instructions, it does not attempt to override agent safety protocols. It contains a self-correction mechanism to prevent the agent from actually executing the instructions found within the audited prompt (e.g., 'no ejecutar el código pedido').
- [COMMAND_EXECUTION]: There are no shell commands or subprocess calls within the skill instructions. The skill is entirely descriptive and focuses on natural language processing of user inputs.
- [NO_CODE]: The skill consists of Markdown instructions and JSON metadata without accompanying scripts or executable files.
Audit Metadata