prd
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- PROMPT_INJECTION (SAFE): No malicious instructions, bypass markers, or safety-filter override attempts were detected. The skill uses standard instructional formatting to guide document generation.
- DATA_EXFILTRATION (SAFE): The skill does not attempt to access sensitive file paths (e.g., ~/.ssh, .env) or perform network operations to any domains.
- REMOTE_CODE_EXECUTION (SAFE): There are no patterns involving the download or execution of external scripts or packages.
- COMMAND_EXECUTION (SAFE): No system commands, shell scripts, or subprocess executions are present.
- INDIRECT_PROMPT_INJECTION (SAFE): While the skill processes user-provided product ideas, it lacks exploitable capabilities such as file-writing or network access, making it safe from indirect injection attacks.
- NO_CODE (SAFE): This skill is entirely markdown-based and does not include any executable script files.
Audit Metadata