hubspot-crm-data-hygiene
Pass
Audited by Gen Agent Trust Hub on Jun 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious code, unauthorized network access, or credentials were found in the skill instructions.
- [PROMPT_INJECTION]: The skill was analyzed for indirect prompt injection risks, as it processes external CRM data from HubSpot. The risk is mitigated by explicit instructions that prohibit autonomous data merges or overwrites and require a human-in-the-loop confirmation table for all write operations.
- [DATA_EXFILTRATION]: No evidence of unauthorized data exfiltration or external network calls to untrusted domains was found. The skill operates strictly within the context of HubSpot CRM tools.
- [COMMAND_EXECUTION]: The skill uses defined CRM search and fetch tools and does not attempt to execute arbitrary shell commands or system-level operations.
Audit Metadata