The Agent Skills Directory

[PROMPT_INJECTION]: No prompt injection or behavior override patterns detected. The instructions follow standard educational patterns for testing frameworks.
[DATA_EXPOSURE]: No sensitive file paths or credential exposure found. Network operations are absent from the executable snippets.
[REMOTE_CODE_EXECUTION]: Dependencies listed (pytest, pytest-cov, pytest-asyncio, pytest-xdist, pytest-mock) are standard, well-known packages in the Python testing ecosystem.
[COMMAND_EXECUTION]: Command patterns are restricted to standard testing and python execution (pytest, python, uv run) as defined in the allowed-tools configuration.
[DATA_EXFILTRATION]: No patterns for exfiltrating data to external domains were identified.
[INDIRECT_PROMPT_INJECTION]: While the skill involves processing and running test files (untrusted data), it lacks the capability chains or dangerous interpolation patterns typically associated with indirect injection attacks.

pytest-advanced