data-wizard

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources (CSV, JSON, and SQL databases), which presents a potential surface for indirect prompt injection attacks.
  • Ingestion points: The skill ingests data using pd.read_csv, pd.read_json, and pd.read_sql_query within the DataPipeline class in SKILL.md.
  • Boundary markers: The instructions lack explicit delimiters or guidance to the agent to disregard instructions that might be embedded within the source data.
  • Capability inventory: The skill has the capability to write to the file system (to_csv, to_parquet, savefig) and interact with databases (to_sql), creating a path for potential misuse if data content influences actions.
  • Sanitization: There is no validation or sanitization of the content within the ingested data before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 09:41 AM
Security Audit — agent-trust-hub — data-wizard