The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides scanner scripts (analyze_maintainability.py, analyze_app_resilience.py, analyze_mockability.py) that read and process the content of arbitrary files within a user-specified project directory, which introduces a potential surface for indirect prompt injection. Ingestion points: These scripts read file contents using Path.read_text() and os.open(). Boundary markers: The instructions do not define delimiters or 'ignore instructions' directives for the processed content. Capability inventory: The skill has the ability to read the filesystem and execute internal scripts. Sanitization: No sanitization is performed on file contents before analysis via regex or AST.
[COMMAND_EXECUTION]: The regression test suite in scripts/test_analyzers.py uses subprocess.run() to execute the skill's own scanner scripts.
[DYNAMIC_EXECUTION]: The scripts/test_analyzers.py file uses importlib.util to dynamically load the maintainability scanner module during its testing phase.

maintainable-code