freshservice-automation
Warn
Audited by Gen Agent Trust Hub on Aug 24, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the user to add an external MCP server endpoint (https://rube.app/mcp) to their client configuration. This domain is not a recognized or well-known service provider, introducing a dependency on a third-party server.- [REMOTE_CODE_EXECUTION]: By registering the remote MCP server at rube.app, the agent relies on tool definitions and execution logic provided by that external source. This effectively grants the remote provider the ability to define the tools and code the agent executes when using this skill.- [DATA_EXFILTRATION]: The tool FRESHSERVICE_CREATE_TICKET_OUTBOUND_EMAIL allows the agent to send emails to arbitrary addresses. When used alongside tools that retrieve sensitive ITSM data (such as FRESHSERVICE_GET_TICKET), it provides a functional mechanism for exfiltrating internal ticket information.- [PROMPT_INJECTION]: The skill processes external ticket data (descriptions and custom fields) which may contain malicious instructions. The skill lacks instructions for the agent to sanitize this content or use boundary markers to prevent indirect prompt injection attacks.
- Ingestion points: Freshservice ticket bodies and custom fields via FRESHSERVICE_LIST_TICKETS and FRESHSERVICE_GET_TICKET.
- Boundary markers: None provided to isolate untrusted data from the agent's core instructions.
- Capability inventory: Includes tools for sending outbound emails (FRESHSERVICE_CREATE_TICKET_OUTBOUND_EMAIL) and bulk modifications (FRESHSERVICE_BULK_UPDATE_TICKETS).
- Sanitization: No evidence of input validation or content filtering for ingested data.
Audit Metadata