The Agent Skills Directory

[SAFE]: No malicious behavior, obfuscation, or unauthorized exfiltration patterns were detected in the skill's instructions or provided scripts.
[COMMAND_EXECUTION]: The skill uses local scoring scripts to perform static analysis on Godot project files. These scripts, including score_modernity.py and bottleneck_scanner.gd, are intended to calculate architectural metrics and do not involve network-based execution.
[PROMPT_INJECTION]: The skill ingests untrusted data by reading Godot scripts and documentation (docstrings) for analysis. While this presents a surface for indirect prompt injection, it is a functional requirement of the auditing process. No active exploit patterns were found in the provided rubrics.

godot-analyst