eda-report

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and analyze external datasets (e.g., data.csv). This creates a theoretical surface for indirect prompt injection where malicious content within the data could attempt to influence the agent's generated report.
  • Ingestion points: Data is loaded from data.csv using pandas.read_csv in Step 1.
  • Boundary markers: There are no explicit instructions provided to the agent to ignore or delimit instructions found within the data rows.
  • Capability inventory: The skill is restricted to data profiling and saving visualization files (distributions.png, correlation_heatmap.png) locally.
  • Sanitization: No specific filtering or sanitization of column values is performed before profiling and summarizing.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 06:28 AM
Security Audit — agent-trust-hub — eda-report