helpdesk-automation

Pass

Audited by Gen Agent Trust Hub on Apr 17, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection through the processing of untrusted HelpDesk data.
  • Ingestion points: Ticket bodies and subjects via HELPDESK_LIST_TICKETS, and template content via HELPDESK_LIST_CANNED_RESPONSES.
  • Boundary markers: The instructions lack delimiters or system-level warnings to ignore embedded commands in retrieved data.
  • Capability inventory: The agent is instructed to use RUBE_MANAGE_CONNECTIONS and RUBE_SEARCH_TOOLS to interface with the HelpDesk toolkit.
  • Sanitization: No sanitization or validation of the external content is performed before processing.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes an external MCP server endpoint at https://rube.app/mcp for tool discovery and execution.
  • [NO_CODE]: No scripts or binaries are included with the skill; it consists entirely of markdown-based instructions and configuration.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 17, 2026, 06:45 AM