shiny-notifications
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides tools for AI agents to manage reminders, creating a surface for indirect prompt injection where an agent might be manipulated via user chat input to perform unintended notification actions.
- Ingestion points: Untrusted user input can influence the parameters passed to tools like
create_reminderandcancel_reminderinSKILL.md. - Boundary markers: The instructions do not specify the use of delimiters or instructions for the agent to treat notification content as untrusted.
- Capability inventory: The
INotificationManagerallows the agent to send, list, and cancel notifications as specified inreference/api-reference.md. - Sanitization: The
AssertValidmethod provides structural validation for notifications before they are sent.
Audit Metadata