data-quality-analysis
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill establishes a methodical workflow for data quality assessment, incorporating discovery interviews, automated profiling, and systematic evaluation of seven quality dimensions. This behavior is consistent with its stated purpose.
- [EXTERNAL_DOWNLOADS]: The HTML report template references stylesheets and fonts from well-known technology providers, specifically Cloudflare (Font Awesome) and Google (Google Fonts). These external resources are used neutrally for presentation and do not introduce security risks.
- [PROMPT_INJECTION]: The skill processes untrusted external data files (e.g., CSV, Excel, Parquet), which presents an attack surface for indirect prompt injection. Malicious content within the ingested data could attempt to influence the agent's analysis or be reflected in the final report. This is a common characteristic of data analysis skills and is addressed through the agent's context and guardrails.
- Ingestion points:
SKILL.md(Phase 1: Data Ingestion & Profiling). - Capability inventory: Data profiling using Python (pandas/numpy), file reading, and HTML report generation.
- Boundary markers: Not explicitly used in the instructions to separate data content from analysis.
- Sanitization: No specific filtering or escaping logic is defined for data samples included in the output report.
Audit Metadata