dotnet-realtime-communication
Pass
Audited by Gen Agent Trust Hub on Mar 7, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Potential for indirect prompt injection via real-time data ingestion.
- Ingestion points: The methods 'NotificationHub.SendMessage', 'ChatHub.SendToRoom', and 'UploadHub.UploadData' in SKILL.md accept untrusted string and stream data from external clients.
- Boundary markers: The code snippets do not implement delimiters or 'ignore' instructions to prevent the agent from interpreting message content as commands.
- Capability inventory: The skill facilitates high-frequency communication which, if connected to an agent with shell or file system access, increases the risk of automated instruction execution.
- Sanitization: There is no evidence of input validation or content sanitization for the message payloads in the provided implementation examples.
Audit Metadata