The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides an assessment script in references/system-assessment.md that executes git log via subprocess.run. This is used for 'hotspot' analysis to find frequently changed files in the legacy system. This is a legitimate development use case.- [EXTERNAL_DOWNLOADS]: The reference documentation suggests the use of several standard Python testing and quality tools, including pytest, mutmut, hypothesis, coverage.py, syrupy, and approvaltests. These are common, well-known libraries in the software engineering community.- [INDIRECT_PROMPT_INJECTION]: The skill includes code to analyze user-provided legacy codebases. * Ingestion points: The LegacyCodeAnalyzer class in references/system-assessment.md reads all files within a provided path. * Boundary markers: The static analysis logic does not use specific boundary markers or 'ignore' instructions. * Capability inventory: The skill demonstrates capabilities for executing shell commands (git), reading/writing local files, and proxying network requests to localhost. * Sanitization: The analyzer uses ast.parse for structural analysis, which is a safe, non-executable method of processing code content.

legacy-modernizer