proxmox-full

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Employs curl and jq to interact with the Proxmox VE REST API for legitimate management tasks, including virtual machine lifecycle operations, snapshotting, and system monitoring.
  • [PROMPT_INJECTION]: Identified an indirect prompt injection surface inherent to the skill's management capabilities.
  • Ingestion points: User-defined parameters for virtual machines and containers such as {node}, {vmid}, hostname, password, and ssh-public-keys within SKILL.md.
  • Boundary markers: None present.
  • Capability inventory: Comprehensive resource management including VM creation, deletion, configuration of authentication credentials, and power state control via curl calls in SKILL.md.
  • Sanitization: Input parameters are passed directly to API endpoints as defined in the instructional examples without explicit validation or escaping.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 03:20 PM
Security Audit — agent-trust-hub — proxmox-full