The Agent Skills Directory

[SAFE]: The skill provides documentation and code examples for the 'aeon' project, a well-known open-source toolkit for time series machine learning. The provided code snippets follow standard practices for data analysis and do not include any hidden or malicious logic.
[EXTERNAL_DOWNLOADS]: The skill documents functions for loading benchmarking datasets from established academic sources, such as the UCR/UEA Time Series Archives and Monash University's TSER archive. These downloads are standard for the field and originate from reputable institutions.
[INDIRECT_PROMPT_INJECTION]: The skill provides methods for ingesting data from external files (e.g., .ts, .arff, .csv) for processing. While this represents a surface for untrusted data, it is a primary function of a machine learning tool and does not expose higher-risk capabilities like command execution.
Ingestion points: Data loading utilities in references/datasets_benchmarking.md and SKILL.md.
Boundary markers: None specified in the documentation.
Capability inventory: Data is used solely for training and evaluating time series models (fit, predict, transform).
Sanitization: Data is validated by the library's loaders against expected numeric and structural formats.

aeon