Data Quality

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes external data files in SKILL.md, creating a surface for indirect prompt injection.
  • Ingestion points: The Python code in SKILL.md loads data from files named 'sales_data.csv', 'customers.csv', and 'products.csv' using 'pd.read_csv()'.
  • Boundary markers: There are no boundary markers or 'ignore embedded instructions' warnings in SKILL.md to prevent the agent from being influenced by malicious content inside the CSVs.
  • Capability inventory: The skill's frontmatter in SKILL.md requests 'Bash' (command execution) and 'Write' (file system access) tools, which increases the impact of potential injection.
  • Sanitization: The provided implementation in SKILL.md focuses on data quality metrics and type checks but lacks sanitization for LLM prompt injection strings within processed text fields.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 07:27 AM