freshservice-automation

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

• Third-party content exposure detected (medium risk: 0.30). Freshservice Automation via Rube MCP can ingest outsider-authored ticket content when the required workflow uses FRESHSERVICE_LIST_TICKETS (and optionally FRESHSERVICE_GET_TICKET) to read tickets’ description and conversations/include fields, which may contain arbitrary user-provided text submitted via email/portal/chat/etc.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires adding and calling the MCP server https://rube.app/mcp at runtime (via RUBE_SEARCH_TOOLS / RUBE_MANAGE_CONNECTIONS) to fetch current tool schemas that directly control available tools/parameters and thus agent behavior.