arize-dataset

Pass

Audited by Gen Agent Trust Hub on May 5, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for installing and updating the 'arize-ax-cli' package using standard tools like 'pip', 'pipx', and 'uv'. These references point to the official package in the Python Package Index (PyPI), which is a vendor resource for Arize-ai. \n- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data from files (CSV, JSON, JSONL, Parquet) and JSON strings into datasets, representing an inherent surface for indirect prompt injection. \n
  • Ingestion points: Commands such as 'ax datasets create --file' and 'ax datasets append --json' in 'SKILL.md' ingest data from external sources. \n
  • Boundary markers: No explicit instruction delimiters or 'ignore' warnings for data content are specified. \n
  • Capability inventory: The 'ax' CLI tool performs file reading and network communication with the Arize platform. \n
  • Sanitization: The skill recommends schema validation but does not provide specific content-based sanitization for the data.
Audit Metadata
Risk Level
SAFE
Analyzed
May 5, 2026, 05:07 PM