run-campaign
Pass
Audited by Gen Agent Trust Hub on May 13, 2026
Risk Level: SAFE
Full Analysis
- External Data Ingestion Surface: The skill processes information from third-party services including QuickBooks, PayPal, and HubSpot. This creates an interface where external data enters the agent's context, which is a consideration for indirect prompt influence. The skill manages this surface with several controls:
- Ingestion Points: Data is retrieved from financial and CRM APIs in Steps 1 and 3.
- Boundary Markers: The skill implements mandatory approval gates ("never auto-progress") that serve as logical boundaries.
- Capability Inventory: The skill is configured with
BashandWebFetchcapabilities for data retrieval and tool interaction. - Sanitization: The workflow relies on explicit human review and confirmation before data is used for asset generation or live campaign sends.
Audit Metadata