react-map-gl
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions facilitate the processing of external map data, which represents an attack surface for indirect prompt injection. Maliciously crafted map features (GeoJSON or vector tiles) could potentially influence the agent's behavior during interactive map sessions.
- Ingestion points: Data enters the agent's context through map event handlers (onClick, onMouseMove) which return feature properties as described in references/interactive-layer-ids.md and references/map-event-handlers.md.
- Boundary markers: The provided patterns do not include explicit delimiters or instructions to ignore potential commands embedded within feature properties.
- Capability inventory: The skill provides instructions for the agent to modify the browser URL state (references/map-url-state.md), call map instance methods like fitBounds and flyTo (references/initial-view-state.md), and update feature states (references/feature-state.md).
- Sanitization: There are no instructions for sanitizing or validating feature properties before they are processed or used to update application state.
Audit Metadata