chief-of-staff
Pass
Audited by Gen Agent Trust Hub on Apr 1, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes local Python scripts (
routing_engine.py,synthesis_generator.py,ecosystem_mapper.py) to perform complex logic. These scripts are self-contained, use only standard library imports, and do not perform any dangerous operations like arbitrary command execution or network calls. - [DATA_EXPOSURE]: The skill documentation mentions loading a local
company-context.mdfile to provide relevant advice. This is a standard practice for specialized AI agents and does not involve hardcoded secrets or unauthorized exfiltration. - [PROMPT_INJECTION]: The
SKILL.mdfile includes robust 'Loop Prevention Rules' and 'Quality Standards' designed to prevent the agent from entering infinite recursive loops or providing low-quality generic advice. These are protective measures rather than malicious injections. - [REMOTE_CODE_EXECUTION]: No remote code execution patterns were found. All script executions are limited to the files provided within the skill bundle.
Audit Metadata