arcgis-pro

Pass

Audited by Gen Agent Trust Hub on Oct 23, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external spatial data formats which could potentially contain malicious metadata or embedded instructions.
  • Ingestion points: The scripts run_spatial_analysis.py (SKILL.md) and export_map_series.py (gemini.md) accept file paths for feature classes, rasters, and project files (.aprx) as command-line arguments.
  • Boundary markers: The skill does not implement delimiters or explicit 'ignore' instructions when processing data from these external files.
  • Capability inventory: The skill utilizes powerful ArcPy capabilities, including database modification (UpdateCursor), geometry modification (RepairGeometry), and filesystem writes (PDF/TIFF exports).
  • Sanitization: No sanitization or validation of the internal data structure of the input GIS files is performed before they are passed to geoprocessing tools.
  • [COMMAND_EXECUTION]: The skill contains a reference to a non-standard ArcPy function in a production script example.
  • Evidence: In SKILL.md, the script run_spatial_analysis.py calls arcpy.analysis.MultipleURLRingBuffer. This function does not exist in the official Esri ArcPy library; the correct tool name is MultipleRingBuffer. The inclusion of 'URL' within a function name where a geoprocessing operation is expected is non-standard and could indicate poorly vetted code or lead to runtime execution failures.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 23, 2026, 09:13 PM