helpdesk-automation
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the user to configure an external MCP server endpoint at
https://rube.app/mcp. This server is used to broker tool calls and connections to the HelpDesk provider. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from external ticketing systems.
- Ingestion points: Data retrieved from
HELPDESK_LIST_TICKETS(ticket bodies and subjects) andHELPDESK_LIST_CANNED_RESPONSES(template content) enters the agent's context. - Boundary markers: The skill instructions do not specify any delimiters or safety markers to differentiate between system instructions and ticket content.
- Capability inventory: The current toolkit configuration focuses on read-only operations (listing tickets, viewing custom fields), which limits the impact of potential injections.
- Sanitization: There is no evidence of content sanitization or validation of the retrieved ticket data before it is presented to the agent.
Audit Metadata