The Agent Skills Directory

[COMMAND_EXECUTION]: The skill defines quality gates in SKILL.md and references/tooling.md that execute shell commands such as pytest and ruff. Running pytest involves executing the logic contained within the project's test files, which creates a risk of arbitrary code execution if the agent is tasked with reviewing or testing malicious or unvetted code. Evidence Chain: 1. Ingestion points: Python source code and test files in src/ and tests/. 2. Boundary markers: No explicit boundary markers or instructions to ignore embedded malicious instructions in processed code are present. 3. Capability inventory: Use of subprocess execution for ruff, mypy, and pytest. 4. Sanitization: No sanitization or content validation of the code is performed prior to execution.

python-data-engineer