pre-call-research-brief
Pass
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill utilizes official vendor-provided MCP tools (Amplemarket and HubSpot) for data retrieval, consistent with its stated purpose of sales research.
- [SAFE]: The skill implements user transparency regarding credit costs, instructing the agent to warn the user and obtain explicit confirmation before invoking enrichment APIs that incur charges.
- [SAFE]: The skill exhibits robust error handling and graceful degradation; it checks for the availability of HubSpot CRM and skips the relevant sections if the integration is not detected, avoiding errors or empty outputs.
- [SAFE]: The instructions emphasize data integrity, explicitly prohibiting the fabrication of data and requiring the agent to cite the specific source for every finding (e.g., CRM vs. web research).
- [SAFE]: The skill maintains the principle of least privilege by focusing on data synthesis and presentation to the user, with no scripts or commands that perform file system modifications, arbitrary code execution, or network exfiltration to untrusted domains.
Audit Metadata