flutter-location-maps
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely instructional and promotes defensive programming practices for mobile development.
- [SAFE]: The guidelines explicitly advise against logging user coordinates or service credentials in diagnostic output or telemetry.
- [SAFE]: External links are restricted to official developer documentation from trusted sources (Google, Apple, Android, and IETF) and standard package registries (pub.dev).
- [INDIRECT_PROMPT_INJECTION]: The skill outlines the handling of external data from geocoding and search providers, which constitutes a potential attack surface. However, it recommends mitigations such as treating results as candidate data and requiring user confirmation.
- Ingestion points: Data enters the agent context via external search results and geocoding responses processed in
references/nearby-search-and-address-picking.md. - Boundary markers: The instructions suggest keeping device fixes, display viewports, and selected points distinct to avoid state confusion.
- Capability inventory: The skill references the use of networking and native platform integrations for standard feature implementation.
- Sanitization: Emphasizes coordinate normalization and unit validation to prevent data swaps or misinterpretation.
Audit Metadata