The Agent Skills Directory

[SAFE]: The skill provides legitimate documentation for the Polars library, a high-performance data manipulation tool. All provided code snippets follow standard API usage and best practices for data science workflows.- [COMMAND_EXECUTION]: The skill includes shell commands for environment setup, specifically checking the Polars version and installing packages via pip. These operations target well-known, reputable libraries (polars, duckdb, pyarrow) and are essential for the skill's functionality.- [INDIRECT_PROMPT_INJECTION]: The skill defines an attack surface by facilitating the ingestion of external data from CSV, Parquet, JSON, and SQL databases.
Ingestion points: Methods such as pl.read_csv, pl.read_parquet, and pl.read_database are documented (SKILL.md, references/api_reference.md).
Boundary markers: No explicit boundaries are shown, which is standard for library documentation.
Capability inventory: Data processing, aggregation, and file system writing (write_parquet, write_csv).
Sanitization: Standard library parsing behavior is assumed.
This represents a standard capability for data tools and does not constitute a security violation.

polars