scientist-high

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from external structures (DataFrames), which constitutes a standard ingestion surface for untrusted data in research workflows.
  • Ingestion points: Data is loaded into the df variable in SKILL.md for feature engineering and causal analysis.
  • Capability inventory: The skill provides templates for local statistical calculation and model training using standard libraries.
  • Boundary markers: None present in the provided snippets.
  • Sanitization: No explicit data validation or sanitization logic is included.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 09:40 AM
Security Audit — agent-trust-hub — scientist-high