The Agent Skills Directory

[SAFE]: The skill instructions and reference materials were analyzed and no malicious patterns, obfuscation, or unauthorized access attempts were found.
[EXTERNAL_DOWNLOADS]: The skill facilitates the download of datasets and metadata from established academic and research repositories.
Evidence: Fetches metadata from https://timeseriesclassification.com/aeon-toolkit/metadata.csv and benchmark results from https://timeseriesclassification.com/results/ReferenceResults.
Evidence: Suggests fetching citation metadata from https://arxiv.org/abs/2609.00065.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface through external dataset loaders.
Ingestion points: Dataset loading functions in aeon.datasets (e.g., load_classification, load_regression) retrieve data from Zenodo and UCR/UEA archives.
Boundary markers: None present; however, the ingested data consists of numerical time series arrays, which are not suitable for carrying text-based instructions.
Capability inventory: The skill is configured to use Bash, Read, Write, and Edit tools.
Sanitization: Not applicable for the numerical time series data processed by this skill.

aeon