close-automation
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides guidelines for interacting with the Close CRM via the Rube MCP service. It focuses on lead management, call logging, SMS messaging, and task management using documented tool sequences.
- [SAFE]: It includes explicit safety boundaries, such as requiring user authorization for outbound communications and mutations.
- [EXTERNAL_DOWNLOADS]: The skill references an external MCP server endpoint (rube.app), which is a legitimate architectural requirement for the described functionality.
- [DATA_EXFILTRATION]: No evidence of unauthorized data exfiltration. The skill acknowledges CRM data as untrusted private data and mandates explicit user instructions for any external transmission.
- [COMMAND_EXECUTION]: The skill does not contain shell commands or script execution logic; it defines high-level tool interactions within an agent framework.
Audit Metadata