The Agent Skills Directory

[COMMAND_EXECUTION]: The provided scripts/lint.py script invokes the ruff linter using subprocess.run(). The command is implemented securely using an argument list rather than a shell string, which prevents command injection. This execution is a legitimate and functional part of the skill's code-reviewing capabilities.
[EXTERNAL_DOWNLOADS]: The skill facilitates the use of ruff, a well-known and trusted third-party linter. It provides instructions to the user to install this tool from the official Python Package Index (PyPI) using standard package management tools.
[PROMPT_INJECTION]: The skill analyzes user-provided code, creating a surface for indirect prompt injection via code comments. However, the risk is mitigated by the skill's rigorous instructional framework, which binds the agent's analysis to 90 specific technical items from the 'Effective Python' guidelines, focusing the model on code structure rather than prose instructions embedded in the data.

effective-python