NYC

autoviz

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS] (LOW): The skill documentation suggests installing several common Python data science libraries via pip (autoviz, matplotlib, seaborn, plotly, bokeh, pandas). While these are external dependencies, they are well-established community packages.
  • [Indirect Prompt Injection] (LOW): The skill acts as a data processing interface, which is a common surface for indirect injection.
  • Ingestion points: Processes external CSV files and DataFrames via the filename and dfte parameters.
  • Boundary markers: None explicitly defined in the README context.
  • Capability inventory: Read access to local datasets and write access to the filesystem for saving generated plots and HTML reports.
  • Sanitization: Relies on the security and parsing logic of the underlying pandas and autoviz libraries.
  • [SAFE] (SAFE): A comprehensive review found no evidence of prompt injection, credential exposure, obfuscation, or persistence mechanisms. The skill's behavior matches its stated purpose of data visualization.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:22 PM