agency-it-service-manager
Pass
Audited by Gen Agent Trust Hub on Aug 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of Markdown-formatted instructions and text templates for IT Service Management (ITSM). No executable scripts, binaries, or configuration files are present.
- [DATA_EXPOSURE]: While the skill describes processes for handling organizational IT data (Incidents, CMDB records, SLAs), it contains no commands to access local sensitive files (e.g., SSH keys, environment variables) or network tools for exfiltration.
- [COMMAND_EXECUTION]: There are no shell commands or subprocess calls defined in the instructions. All technical deliverables are provided as static Markdown templates for the agent to use in its responses.
- [PROMPT_INJECTION]: The 'Critical Rules' section defines operational constraints for the persona (e.g., incident classification logic). These are domain-specific instructions that do not attempt to override the underlying AI's safety guidelines or extract system prompts.
Audit Metadata