The Agent Skills Directory

[COMMAND_EXECUTION]: The skill encourages and instructs the agent to use various CLI tools including uv, ruff, pytest, docker, and kubectl for project management, linting, and deployment.
[DYNAMIC_EXECUTION]: The skill identifies metaprogramming and dynamic code generation as core competencies. These are standard advanced Python features but can be exploited to execute arbitrary logic if the agent is directed to process untrusted input.
[PROMPT_INJECTION]: The skill has a surface for indirect prompt injection (Category 8) because its primary purpose involves reviewing external codebases and processing data pipelines.
Ingestion points: Python codebases, data files (via Pandas/NumPy), and API request schemas in advanced-python-patterns.md and data-science-machine-learning.md.
Boundary markers: No specific boundary markers or 'ignore embedded instructions' warnings are present in the skill's instructions for handling external data.
Capability inventory: The skill is capable of file system operations, subprocess execution (via modern tooling), and network requests (via FastAPI/asyncio).
Sanitization: The skill emphasizes Pydantic for data validation but does not provide explicit instructions for sanitizing code comments or documentation that might contain malicious prompts.
[EXTERNAL_DOWNLOADS]: The skill recommends using package managers like uv and pip to install dependencies from well-known services such as PyPI.

python-pro