websocket-engineer
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides a tutorial implementation for a chat application which creates a potential surface for indirect prompt injection via user-supplied messages.
- Ingestion points: The
server.jscode snippet in Workflow 1 processes untrusted input through thesocket.on("message")event listener. - Boundary markers: The provided implementation does not include specific boundary markers or instructions to the LLM to ignore embedded commands in the data stream.
- Capability inventory: The skill demonstrates network communication via Socket.IO for broadcasting messages to rooms.
- Sanitization: The code snippet lacks explicit sanitization or filtering of incoming message data, although the skill's documentation explicitly recommends 'Input Validation' as a security best practice. Because this surface is fundamental to the skill's primary purpose of teaching real-time communication, the risk is considered negligible in this context.
Audit Metadata