CSV
Pass
Audited by Gen Agent Trust Hub on Apr 20, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely instructional and does not include any scripts, tools, or executable commands.
- [DATA_EXPOSURE_AND_EXFILTRATION]: No network access or sensitive file operations are requested or performed.
- [PROMPT_INJECTION]: The content does not contain any instructions aimed at bypassing AI safety protocols or overriding system instructions.
- [INDIRECT_PROMPT_INJECTION]: The skill documents the attack surface of CSV data (Excel formula injection) and provides remediation guidance to prevent the agent from creating vulnerable files.
Audit Metadata