The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/test_skill.py executes internal scripts (validate.py and analyze_maintainable_tests.py) using subprocess.run. These calls are restricted to the skill's own components and are used exclusively for self-validation and testing of the scanner's functionality.
[EXTERNAL_DOWNLOADS]: The skill references external URLs in its documentation (references/source-notes.md) and includes a sample URL in a test fixture (scripts/test_skill.py). These references are for documentation purposes or to test the scanner's detection logic; no network requests are initiated by the skill during its normal operation.
[PROMPT_INJECTION]: As a code review skill, it processes user-provided test files via the analyze_maintainable_tests.py script. While this introduces an indirect prompt injection surface where maliciously crafted test names could theoretically influence the agent's summary, the risk is inherent to the skill's purpose as a reviewer and the script itself does not execute any of the scanned content.

maintainable-tests