flutter-firebase-messaging
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions and reference materials describe how to ingest and process external data from Firebase Cloud Messaging (FCM) payloads, which represents a potential surface for indirect prompt injection.
- Ingestion points: Remote notification data (
RemoteMessage.data) and payload content processed within background and foreground listeners (identified inSKILL.mdandreferences/production-guide.md). - Boundary markers: The skill includes 'Non-Negotiable Rules' stating that payload data and notification routes must be treated as untrusted input and require authentication and authorization before any navigation or state mutation occurs.
- Capability inventory: The implementation involves background execution isolates, interaction streams that can trigger application-wide navigation, and potential persistence of received metadata.
- Sanitization: The architecture recommends a dedicated 'Notification parser' layer to perform schema validation and domain mapping, ensuring that only expected values and whitelisted action types are processed.
Audit Metadata