skills/iuliandita/skills/nixos/Gen Agent Trust Hub

nixos

Pass

Audited by Gen Agent Trust Hub on Nov 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill makes extensive use of administrative commands (e.g., sudo nixos-rebuild, sudo dd, systemctl) to manage system state, disk partitions, and services. This is consistent with its stated purpose of NixOS administration.
  • [EXTERNAL_DOWNLOADS]: Fetches tools and configurations from trusted and well-known repositories within the Nix ecosystem.
  • Fetches disk management tools from the Nix Community GitHub organization (github:nix-community/disko).
  • Downloads remote installation scripts from the Nix Community repository (github:nix-community/nixos-anywhere).
  • Uses secrets management tools from established maintainers (github:ryantm/agenix, github:Mic92/sops-nix).
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection as it processes declarative configuration data while having high-privilege execution capabilities.
  • Ingestion points: Processes configuration.nix, flake.nix, and other local Nix files (specified in SKILL.md and references/configuration-and-modules.md).
  • Boundary markers: The skill provides a structured workflow and an 'AI Self-Check' to verify system state and intentions, though it lacks technical boundaries to prevent the execution of malicious Nix code if present in the configuration.
  • Capability inventory: Includes commands for system rebuilds, disk formatting, service manipulation, and secrets handling (detailed across the reference files).
  • Sanitization: No explicit sanitization of Nix expressions is performed, as the agent is expected to interpret these as authoritative system definitions.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 12, 2026, 09:40 AM