The Agent Skills Directory

[SAFE] (SAFE): The skill documentation and scripts follow secure coding practices. Specifically, the YAML parser in quick_validate.py uses yaml.safe_load() to prevent arbitrary code execution during the validation process.
[DATA_EXFILTRATION] (SAFE): No hardcoded credentials, sensitive path access, or network calls (e.g., via curl or requests) were identified in the source code.
[COMMAND_EXECUTION] (SAFE): The Python scripts interact with the filesystem using standard libraries (pathlib, zipfile) and do not use unsafe functions like eval(), exec(), or subprocess.run() on user-controlled inputs.
[PROMPT_INJECTION] (SAFE): Documentation files contain instructional templates for structuring agent responses, but these do not include malicious bypass markers or safety filter override attempts.

skill-creator