carto-import-export-data
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill follows documented procedures for data warehouse management using the CARTO platform. All described actions, including warehouse-native unloads and activity exports, align with the stated purpose of geospatial data movement.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external geospatial data (CSV, GeoJSON, Shapefiles, etc.) from local files and remote URLs, creating a surface for potential instruction injection within data files.
- Ingestion points: Untrusted data enters the agent context via the
import_datatool and thecarto importCLI command using--fileor--urlarguments (references/imports.md). - Boundary markers: No specific delimiters or "ignore instructions" warnings are mentioned for the imported data content.
- Capability inventory: The skill environment includes warehouse query execution (
execute_async_query), file writing (export_data), and various CLI tools for SQL execution and activity logging. - Sanitization: While the skill mentions geospatial parsing and schema inference, it does not specify sanitization or filtering for natural language instructions embedded within the data payloads.
Audit Metadata