The Agent Skills Directory

[SAFE]: The skill is entirely educational and instructional, focusing on promoting high-quality Python code and identifying security vulnerabilities in other codebases.
[COMMAND_EXECUTION]: The skill includes many standard shell commands for development, linting, and security scanning (e.g., uv, ruff, bandit, pip-audit). These are used as intended for auditing and project management.
[PROMPT_INJECTION]: No attempts to override agent behavior or safety filters were found. The mention of 'CRITICAL' or 'HIGH' severities refers to finding classifications for code audits, not an attempt to escalate privileges.
[DATA_EXFILTRATION]: There are no network operations targeting suspicious domains. All external references are to well-known repositories like Astral (Ruff) and standard pre-commit hooks.
[REMOTE_CODE_EXECUTION]: The skill specifically warns against dangerous patterns like pickle.loads or eval on untrusted data, demonstrating a strong security posture.

sota-python