build-with-arcgis-and-openlayers
Pass
Audited by Gen Agent Trust Hub on Nov 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill promotes a secure-by-default posture by instructing developers to never expose client secrets in browser code and to restrict API keys by referrer. It also mandates the use of PKCE for OAuth flows.
- [INDIRECT_PROMPT_INJECTION]: The skill involves the ingestion of data from external ArcGIS Feature Services, which presents a surface for indirect prompt injection.
- Ingestion points: ArcGIS Feature Service endpoints (
FeatureServer/<layerId>) and Location Services (Geocoding, Places). - Boundary markers: None explicitly for the prompt context, but the skill provides explicit instructions to treat incoming data as untrusted.
- Capability inventory: The skill uses
@esri/arcgis-rest-*packages for network operations and OpenLayers for rendering. It does not include arbitrary code execution or filesystem write capabilities. - Sanitization: The skill explicitly recommends using DOM
textContentand sanitizing any HTML attributes before rendering popups to prevent Cross-Site Scripting (XSS) and injection attacks.
Audit Metadata