action-cable-patterns

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill implements a chat channel pattern that facilitates the ingestion and broadcasting of user-generated content, which is a standard surface for indirect prompt injection if the content is later processed by an agent. \n- Ingestion points: The ChatChannel#speak method in references/channel-patterns.md accepts a body parameter directly from the WebSocket client. \n- Boundary markers: The provided code snippets do not include explicit instructions or markers to delimit user-provided message content from system instructions. \n- Capability inventory: The skill allows for broadcasting messages to all users in a chat room (broadcast_to) and dynamically updating the user interface via JavaScript (insertAdjacentHTML). \n- Sanitization: The implementation uses ApplicationController.renderer.render to generate HTML partials, which natively applies Rails' automatic HTML escaping and sanitization to user input within the template before it is sent to the client.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 03:35 PM