skills/shinyorg/skills/shiny-push/Gen Agent Trust Hub

shiny-push

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill documents procedures for ingesting and processing data from external, untrusted push notification providers (APNs, FCM, and Azure Notification Hubs), which could be leveraged if the implementing code does not apply proper validation.
  • Ingestion points: Push notification payloads enter the application context through the PushNotification.Data dictionary within the IPushDelegate.OnReceived and IPushDelegate.OnEntry handlers (found in both SKILL.md and reference/api-reference.md).
  • Boundary markers: The documentation and code examples do not specify boundary markers or provide instructions to treat the incoming payload data as untrusted input.
  • Capability inventory: Handlers for push notifications typically trigger actions such as navigation, user alerts, and data logging, which are sensitive to the contents of the notification payload.
  • Sanitization: The provided examples do not include sanitization, validation, or filtering of the Data dictionary content before it is used to perform application logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 11:41 PM
Security Audit — agent-trust-hub — shiny-push