godot-nodes-scenes
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a bundled Python utility (
scripts/check_godot_project.py) to validate Godot project integrity and file references. - [INDIRECT_PROMPT_INJECTION]: The
check_godot_project.pyscript creates a surface for indirect prompt injection by parsing and reflecting content from external, potentially untrusted Godot project files. - Ingestion points: The script reads file contents via
Project.loadandread_binary_uidfor.tscn,.tres,.gd, and.godotfiles. - Boundary markers: Output messages containing data from these files do not use delimiters or boundary markers to isolate it from the agent's instructional context.
- Capability inventory: The skill allows the agent to read the local filesystem via the script; the agent may then act upon the content of the generated reports.
- Sanitization: The script interpolates raw identifier strings from the project files directly into its console output without escaping or validation.
Audit Metadata