The Agent Skills Directory

[SAFE]: The skill functions as a security analysis utility, providing documented heuristics to identify common malware distribution patterns (e.g., password-protected archives, lack of source code) in game modification tools.
[COMMAND_EXECUTION]: Provides Python templates that use standard library functions like os.listdir, pathlib.Path.rglob, and subprocess to inspect local repository structures for suspicious file extensions such as .exe, .dll, and .bat.
[PROMPT_INJECTION]: Contains an indirect prompt injection surface because the skill is designed to ingest and process untrusted data from external repositories.
Ingestion points: Reads content from files within specified repository paths in scan_repository.py and legitimate_checker.py, and parses README strings in analyze_readme (SKILL.md).
Boundary markers: None explicitly implemented in the provided logic to isolate untrusted data from agent instructions.
Capability inventory: File system reading using rglob and open, directory listing using os.listdir, and subprocess execution capability in scan_repository.py.
Sanitization: No sanitization or escaping of external content is present before the data is processed by the heuristics.

malware-game-cheat-detection