aeon
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWEXTERNAL_DOWNLOADS
Full Analysis
- External Downloads (LOW): The skill documentation describes functions like load_classification and download_all_regression in datasets_benchmarking.md which fetch datasets from well-known academic repositories such as timeseriesclassification.com and forecastingdata.org. These are trusted sources for benchmarking and do not involve executable code.
- Indirect Prompt Injection (LOW): The skill provides tools to ingest and process external time series data from formats like .ts, .arff, and .csv. Evidence: Ingestion points: load_from_ts_file, load_from_tsv_file (datasets_benchmarking.md); Boundary markers: Absent; Capability inventory: The skill enables model training and prediction but does not expose subprocess execution or network-write capabilities to the data; Sanitization: Absent. This represents a low-risk surface for indirect prompt injection via data processing.
Audit Metadata