The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional content for pandas developers, focusing on best practices for data cleaning, aggregation, merging, and performance optimization.- [PROMPT_INJECTION]: No patterns of instruction override, safety bypass, or role-play injection were found. The instructional language is professional and task-oriented.- [DATA_EXFILTRATION]: There are no commands that attempt to access sensitive system files or credentials. Network references are limited to legitimate project metadata (GitHub author profile).- [DYNAMIC_EXECUTION]: The skill correctly identifies pd.eval() as a performance optimization tool in references/performance-optimization.md. The usage is presented within the context of safe algebraic expressions for large DataFrames.- [INDIRECT_PROMPT_INJECTION]: The skill processes external data sources through common I/O functions like pd.read_csv (e.g., in references/performance-optimization.md). While this is an attack surface, the skill provides extensive documentation on data validation (Core Workflow in SKILL.md and schema validation in references/data-cleaning.md) which encourages safe data handling.- [COMMAND_EXECUTION]: All code snippets are standard Python/pandas operations designed for data manipulation within a notebook or script environment. No shell commands or privileged operations are included.

pandas-pro