analyze-data-quality

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTION
Full Analysis
  • Cross-Skill Configuration Loading: The skill is designed to invoke a preflight script from a related context skill (../user-context/SKILL.md). This mechanism allows the agent to maintain consistent user preferences and environment settings across different analytics tasks.
  • Dynamic Code Execution for Analysis: To perform data profiling and quality checks, the skill generates and executes SQL or Python code within Jupyter notebooks ($jupyter-notebooks). This approach is used to provide an inspectable and reproducible audit trail of the data quality findings.
  • Data Ingestion Surface: The skill processes external data sources such as tables, files, and query results. As with any tool that interprets untrusted external data, there is a theoretical surface for indirect prompt injection; however, the skill mitigates this by instructing the agent to provide analytical interpretation and evidence-based reporting rather than simply echoing raw data.
  • Local Environment Interaction: The instructions mention the possibility of using local shell access if specific preflight conditions are not met. This is presented as a fallback mechanism for retrieving environment-specific metadata required for the analysis workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 06:12 AM
Security Audit — agent-trust-hub — analyze-data-quality