large-scale-map-visualization

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection due to the ingestion of untrusted external data in combination with high-privilege tool access.
  • Ingestion points: The skill ingests untrusted location data via the locations array in React components and geographic data results returned from Supabase RPC database calls.
  • Boundary markers: There are no explicit boundary markers or instructions to the agent to ignore potentially malicious content embedded within geographic data properties (e.g., place names or metadata).
  • Capability inventory: The skill is granted access to high-privilege tools including Bash, Write, Edit, WebSearch, and WebFetch.
  • Sanitization: While the code uses standard PostGIS functions for spatial queries, it does not explicitly demonstrate sanitization of user-controlled properties before they are rendered in HTML-based icons (divIcon) or processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 11:38 AM