The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a 'LegacyCodeAnalyzer' in references/system-assessment.md that uses subprocess.run to execute git log. This is used to perform 'hotspot analysis' by identifying frequently changed files, which is a standard and safe practice for technical debt assessment.
[DATA_EXPOSURE]: The analysis tools in references/system-assessment.md read local source code files to calculate metrics such as lines of code, cyclomatic complexity (via function length), and dependency maps. These operations are performed locally and no network transmission of the analyzed data is attempted.
[REMOTE_CODE_EXECUTION]: While the skill references external libraries such as pytest, approvaltests, and mutmut for testing purposes, it does not perform any unverified remote script execution or dynamic code evaluation from untrusted sources.
[PROMPT_INJECTION]: The skill processes codebase content which constitutes an indirect prompt injection surface. However, the analysis is limited to static structure (using the ast module) and regex-based pattern matching, which are resilient to instruction injection within the code being analyzed.

legacy-modernizer