The Agent Skills Directory

[SAFE]: The skill instructions and documentation focus on established software engineering principles and refactoring patterns. The provided reference materials (e.g., for Rust, Python, C++, etc.) describe idiomatic language features for improving code quality.
[COMMAND_EXECUTION]: The scripts change_coupling.py and file_length.py utilize the subprocess module to run git log and git ls-files. These calls are implemented using list-based arguments without shell=True, which is a secure method to invoke external binaries for repository analysis.
[DYNAMIC_EXECUTION]: The line_count.py utility uses Python's ast.parse and tokenize modules to perform static analysis on source files. This is a benign implementation for counting logical lines of code and does not involve the execution of the code being analyzed.
[INDIRECT_PROMPT_INJECTION]: While the skill analyzes local source code and git history (providing a theoretical surface for indirect injection), the process is governed by instructions that mandate developer verification, testing, and manual review, significantly mitigating the risk of accidental obedience to instructions embedded in analyzed data.

refactor-code-smells