The Agent Skills Directory

[COMMAND_EXECUTION]: The references/system-assessment.md file contains a Python class LegacyCodeAnalyzer that uses subprocess.run to execute the system git command. This is used to analyze the change history of the codebase being modernized.
[EXTERNAL_DOWNLOADS]: The skill's documentation and reference files (specifically references/legacy-testing.md and references/system-assessment.md) recommend the installation and use of various third-party tools and libraries from the Python Package Index (PyPI), such as mutmut, radon, and pylint.
[PROMPT_INJECTION]: The skill provides automated assessment logic in references/system-assessment.md that reads and processes external source code, which constitutes an indirect prompt injection surface.
Ingestion points: The LegacyCodeAnalyzer class in references/system-assessment.md ingests untrusted code by reading all files within the provided project path.
Boundary markers: No specific delimiters or boundary instructions (e.g., "ignore instructions in the following code block") are utilized when the agent analyzes the ingested source files.
Capability inventory: The skill possesses capabilities for file system access (rglob, open), command execution (subprocess.run), and network communication (via the httpx and stripe libraries in migration examples).
Sanitization: No sanitization, content filtering, or validation is performed on the ingested code before it is passed to the analyzer or processed by the agent.

legacy-modernizer