The Agent Skills Directory

[SAFE]: The analysis of the skill's instructions and reference files did not reveal any malicious patterns, prompt injections, or unauthorized data access. The content is focused on providing architectural knowledge and operational workflows for Proxmox VE management.
[CREDENTIALS_SAFE]: The skill demonstrates safe credential management practices by explicitly instructing agents to store Proxmox API tokens in environment variables (PROXMOX_TOKEN_NAME, PROXMOX_TOKEN_VALUE) or secrets managers, and to use dedicated users with minimal required permissions.
[INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing data from the Proxmox API (logs, tasks, and system statuses). While these ingestion points (proxmox_node_log, proxmox_node_task) could theoretically contain untrusted data, the skill includes verification and monitoring workflows that mitigate the risk of automated obedience to embedded instructions.
[COMMAND_EXECUTION]: Command execution capabilities are limited to documented administrative tools (such as proxmox_agent_exec for QEMU VMs) and standard Proxmox CLI tools (pct exec), which are appropriate for the skill's stated purpose of infrastructure automation.

proxmox-admin