proxmox-vm-recovery

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes host-level commands such as qm, pvesh, and qemu-nbd to perform virtual machine management and disk diagnostics.
  • [PRIVILEGE_ESCALATION]: Several workflows require administrative or root-level access to the Proxmox host, including loading kernel modules (nbd), mounting guest partitions, and modifying system-level VM configurations.
  • [INDIRECT_PROMPT_INJECTION]: The skill relies on external inputs such as VM identifiers and node names which are interpolated directly into shell commands, representing a potential vulnerability surface if these inputs are not properly validated by the executing agent.
  • Ingestion points: VM IDs and node names provided via user interaction or discovery are used in SKILL.md commands.
  • Boundary markers: No delimiters or specific instructions to ignore embedded content are provided.
  • Capability inventory: The skill includes instructions for VM control (qm), API interaction (pvesh), and disk mounting (mount/qemu-nbd) in SKILL.md.
  • Sanitization: There is no evidence of input validation or escaping for the variables used in the commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 02:39 AM
Security Audit — agent-trust-hub — proxmox-vm-recovery