godot-ai-navigation

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a decision router for Godot 4.7+ AI navigation. It provides architectural guidance on choosing between node-based agents and server-level navigation.
  • [EXTERNAL_DOWNLOADS]: The skill references multiple external scripts and sibling skills located at github.com/thedivergentai/gd-agentic-skills. These are documented as mandatory dependencies for the skill's logic but do not involve automated execution of untrusted code outside the vendor's own context.
  • [PROMPT_INJECTION]: The skill uses instructional language like 'MANDATORY' and 'NEVER' to enforce specific coding standards (e.g., avoiding frame-by-frame target updates) to prevent performance issues. This is standard educational practice and does not attempt to bypass AI safety filters.
  • [COMMAND_EXECUTION]: The provided GDScript examples utilize standard Godot API calls (move_and_slide, get_next_path_position) for game character movement. No OS-level command execution or shell spawning is present.
  • [DATA_EXFILTRATION]: No network operations, sensitive file access, or credential usage were detected. The skill focuses entirely on local game engine logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 08:12 AM
Security Audit — agent-trust-hub — godot-ai-navigation