The Agent Skills Directory

PROMPT_INJECTION (LOW): The skill is vulnerable to indirect prompt injection due to its core function of processing external Zendesk data.
Ingestion points: Untrusted data enters the agent context through ZENDESK_LIST_ZENDESK_TICKETS, ZENDESK_GET_ZENDESK_TICKET_BY_ID (ticket comments/descriptions), and ZENDESK_SEARCH_ZENDESK_USERS (user-provided metadata).
Boundary markers: Absent. The instructions do not provide delimiters or specific warnings to the agent to disregard instructions found within ticket content.
Capability inventory: The skill provides significant capabilities, including the ability to delete tickets (ZENDESK_DELETE_ZENDESK_TICKET), modify ticket statuses, and send outbound communications via ZENDESK_REPLY_ZENDESK_TICKET.
Sanitization: Absent. There is no instruction to escape or validate ticket bodies before processing.
EXTERNAL_DOWNLOADS (LOW): The skill directs users to configure an external MCP server at https://rube.app/mcp and references tool documentation at composio.dev. While these are established services in the agent ecosystem, they are not included in the predefined list of trusted organizations.

zendesk-automation