carto-trade-area-analysis

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches geographic data samples from storage.googleapis.com/carto-workflows-examples. These downloads target a well-known cloud storage service and originate from a repository associated with the skill's author, which is consistent with the intended functionality for providing example datasets.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing external data through workflows that have write capabilities and support SQL execution.
  • Ingestion points: Data is ingested via ReadTable nodes in commercial_hotspots.json, identify_best_billboards.json, and isochrones_from_points.json from external URLs.
  • Boundary markers: None identified in the workflow templates to delimit untrusted data.
  • Capability inventory: The workflow components include native.saveastable for writing results to a database, and native.customsql or native.selectexpression for executing SQL strings.
  • Sanitization: There are no explicit sanitization or validation steps mentioned for the ingested data before it is used in SQL expressions or saved to tables.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 02:36 PM