level-design
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: Extensive analysis of the skill's instructions, reference documents, and bundled Python utility scripts revealed no malicious patterns or security risks.
- [COMMAND_EXECUTION]: The skill includes several Python scripts (
grid_lint.py,grid_puzzle_solve.py,platformer_reach.py,jump_links.py,graph_check.py,dependency_chart.py,beat_lint.py, andlevel_metrics.py) which the AI is instructed to execute locally. These scripts use only the Python standard library to perform deterministic spatial and graph calculations. They do not access the network or perform unsafe operations. The presence ofsubprocess.runis limited to the skill's own unit testing suite to verify the CLI tools. - [INDIRECT_PROMPT_INJECTION]: The skill instructions specify that the agent should ingest and analyze level layouts (ASCII grids and Mermaid graphs) provided by users. This creates a data ingestion surface; however, the skill's core logic is to parse these files using specialized scripts for glyph counting and algorithmic solving, rather than instructing the agent to follow natural language instructions embedded within the user data.
- [EXTERNAL_DOWNLOADS]: The
SOURCES.mdfile contains a detailed bibliography of level design resources. These URLs point to legitimate developer blogs, industry news sites, and academic research (e.g., Game Developer, Wikipedia, arXiv, and well-known studios). These are provided for reference only; the skill does not download or execute code from these remote sources at runtime.
Audit Metadata