The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs users to add an external MCP server at https://rube.app/mcp. This third-party service is not on the trusted vendors list and is responsible for providing the tool definitions and execution environment for the automation.
[DATA_EXFILTRATION]: The skill accesses and processes sensitive IT Service Management (ITSM) data, including requester names, email addresses, and ticket descriptions. This information is transmitted to the external rube.app endpoint, creating a potential path for data exposure or exfiltration to an unverified third party.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection attacks because it processes user-generated ticket content without proper isolation.
Ingestion points: Untrusted content is ingested from Freshservice ticket fields via FRESHSERVICE_LIST_TICKETS and FRESHSERVICE_GET_TICKET in SKILL.md.
Boundary markers: The skill lacks delimiters or explicit instructions to the agent to ignore potentially malicious commands embedded within ticket descriptions or subjects.
Capability inventory: The skill has access to high-impact tools such as FRESHSERVICE_BULK_UPDATE_TICKETS, FRESHSERVICE_CREATE_TICKET_OUTBOUND_EMAIL, and FRESHSERVICE_CREATE_SERVICE_REQUEST.
Sanitization: No validation or sanitization mechanisms are present to prevent the agent from following instructions found inside the data it processes.

freshservice-automation