The Agent Skills Directory

[SAFE]: The skill provides instructions for acting as a professional Python developer. It focuses on well-known, high-quality libraries and standard ecosystem tools like FastAPI, Pydantic, and SQLAlchemy. The instructions encourage best practices such as type hinting, error handling, and testing.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and review external Python codebases and performance requirements.
Ingestion points: Untrusted data enters the context when the agent is asked to write or review third-party Python codebases as mentioned in the 'Use this skill when' section.
Boundary markers: None identified in the provided text.
Capability inventory: The skill descriptions include project configuration, testing, and performance tuning, though it does not provide explicit scripts for automated execution.
Sanitization: No specific sanitization or escaping instructions for external code content are provided.
Assessment: While an ingestion surface exists, it is inherent to the primary purpose of a code-reviewing persona. The severity is assessed as safe as no malicious intent or exploit patterns are present.

python-pro