helpdesk-automation

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). The required workflow calls HelpDesk “list/read” tools (e.g., HELPDESK_LIST_TICKETS, HELPDESK_LIST_VIEWS, HELPDESK_LIST_CANNED_RESPONSES, HELPDESK_LIST_CUSTOM_FIELDS) which at runtime return ticket/view/template content authored by other users/agents, and that returned free text is then ingested into the agent’s LLM context for processing.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill requires adding and calling the MCP endpoint https://rube.app/mcp at runtime (via RUBE_SEARCH_TOOLS) to fetch current tool schemas, so remote content from that URL will directly control the agent's available tools/prompts.