websocket-realtime-builder

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill implements a real-time chat feature where untrusted user input is ingested and broadcast. While standard for chat applications, this creates a surface where the agent could be influenced by malicious instructions embedded in messages if it is tasked with processing the conversation history.
  • Ingestion points: The send-message event in namespaces/chat.ts accepts arbitrary string content from users.
  • Boundary markers: Absent. The template does not include logic to wrap user messages in protective delimiters.
  • Capability inventory: The code includes capabilities for database persistence and network broadcasting via WebSockets.
  • Sanitization: No sanitization is performed on the message content before it is stored or emitted to other clients.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 09:30 AM
Security Audit — agent-trust-hub — websocket-realtime-builder