pygmt
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external geospatial data, which constitutes an indirect prompt injection surface.
- Ingestion points: Data enters the system through external grid files and remote PyGMT datasets (e.g., Earth relief) referenced in the instructions.
- Boundary markers: The skill mandates explicit definition of geographic regions, projections, and coordinate systems, which serve as logical boundaries.
- Capability inventory: The code examples include file system writing using the savefig function.
- Sanitization: The skill provides specific instructions to inspect data registration, spacing, coordinate ordering, and NaN masks before processing.
Audit Metadata