analyze

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • Data Processing Surface: The skill is designed to ingest and analyze data from external sources, such as connected data warehouses or user-provided files (CSV, Excel). This introduces a surface for indirect prompt injection where malicious instructions could potentially be embedded within the data being processed. Standard validation steps are included in the workflow to mitigate data quality issues.
  • Dynamic Visualization and Code Generation: As part of its presentation workflow, the skill may generate Python code for visualizations or construct HTML dashboards. This dynamic execution is a typical feature for data analysis agents but should be used within the agent's established execution environments.
  • SQL Query Generation: The skill generates and executes SQL queries based on user questions. This involves interpreting schemas and user intent to perform data retrieval operations.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 03:04 PM
Security Audit — agent-trust-hub — analyze