router

Pass

Audited by Gen Agent Trust Hub on Nov 1, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill functions as a router for game development tasks, fingerprinting engines like Godot, Unity, and Unreal. All logic is focused on task classification and progressive disclosure of instructions.
  • [INDIRECT_PROMPT_INJECTION]: The skill analyzes project files to determine the environment, creating an ingestion surface for untrusted data. 1. Ingestion points: Reads project files such as package.json, go.mod, and project.godot (SKILL.md, references/engine-detection.md). 2. Boundary markers: Not present in the routing logic. 3. Capability inventory: Limited to file structure analysis and task-based skill selection; no direct command execution or network exfiltration found in the router code. 4. Sanitization: Not present for static fingerprinting.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 1, 2026, 04:04 PM