hello-boss
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill documentation includes examples of user queries that trigger business research workflows, such as mapping industries and building prospect lists. These are internal skill operations and do not involve unauthorized system commands.
- [DATA_EXPOSURE_AND_EXFILTRATION]: While the skill involves gathering business emails and registry data, it emphasizes using public sources (e.g., Companies House, SEC EDGAR, national registries) and strictly instructs the agent to avoid sensitive or administrative mailboxes. It includes guidance on platform anti-abuse policies to ensure compliant data gathering.
- [PROMPT_INJECTION]: No patterns of prompt injection, role-play bypasses, or instructions to ignore safety filters were detected. The skill maintains a professional, task-oriented tone.
- [REMOTE_CODE_EXECUTION]: The skill mentions using existing mailing tools rather than hand-rolling SMTP loops, which is a recommended safety practice to manage throttling and unsubscribes. No unauthorized remote script execution was found.
Audit Metadata