netbox-powerdns-integration
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from an external source (NetBox API) and provides tools to create or modify infrastructure resources, creating a potential surface for indirect prompt injection if NetBox contains malicious data.
- Ingestion points: Infrastructure data is retrieved from the NetBox API in
tools/netbox_api_client.pyandtools/netbox_ipam_query.py. - Boundary markers: The agent tools format output as structured JSON or formatted tables using the
richlibrary, providing some separation between data and instructions. - Capability inventory: The skill includes
tools/netbox_vm_create.py, which has the capability to create virtual machines and assign IP addresses in the managed infrastructure. - Sanitization: No specific sanitization of external text fields (e.g., device descriptions) is performed before the data is presented to the agent.
- [EXTERNAL_DOWNLOADS]: The skill utilizes an external Terraform module for VM deployment.
- Source: Downloads the
terraform-bgp-vmmodule from the author's GitHub repository atgithub.com/basher83/Triangulum-Prime. - Purpose: Standardized Proxmox VM deployment configuration.
Audit Metadata