Tool to permanently delete a user from Zendesk. Use when you need to remove a user account. This action cannot be undone and requires appropriate admin permissions.

Tool to bulk delete Zendesk organizations. Irreversible — confirm ids/external_ids before calling. Use when cleaning up test data or removing obsolete organizations in one call. Returns a job_status payload (initial status typically 'queued'); deletions complete asynchronously and may require follow-up polling to confirm completion.

Get information about the currently authenticated user in Zendesk. Returns only the caller's own account; to look up other users, use ZENDESK_SEARCH_ZENDESK_USERS. Response nests the user object under data.owner_info.user. Useful for verifying the acting user's identity and permissions before performing bulk operations.

Get all organizations in Zendesk. Returns results nested under an 'organizations' array; an empty list is valid. Accepts no server-side filters — all filtering by name, domain_names (an array field), or organization_fields must be done client-side. Large accounts require pagination; missing pages will undercount results. Multiple organizations may share similar names or domains — always disambiguate using external_id or domain_names and confirm organization_id before acting. Avoid repeated full-list fetches on large accounts; cache or batch client-side instead.

Tool to fetch a single Zendesk user by numeric user_id. Use when you have a user ID from ticket payloads (requester_id, submitter_id, assignee_id, author_id) and need to enrich with full user details (name, email, role, organization_id, etc.).

Get metadata for a specific Zendesk organization by ID. Response fields are available under data.organization_info. Does not return member/user lists.

Get ticket details from Zendesk. Response wraps all data under a top-level data key; access data['comments'] for comments and data['comments'][i]['attachments'] for attachments (attachments unavailable from list endpoints). Each comment has both html_body and plain body fields — choose appropriately. Fields like subject, organization_id, author_id, and body may be null; handle defensively. First comment in data.comments is not necessarily from the requester — compare author_id against ticket requester_id to identify requester messages. For bulk calls, honor Retry-After headers on HTTP 429.

List Zendesk tickets with pagination and filtering. Only server-side filter available is external_id; status, priority, tags, assignee_id, and date-range filters must be applied client-side. Timestamps (created_at, updated_at) are ISO 8601 UTC. ticket.subject may be null — coerce to string before filtering. Attachments are not included; retrieve via ZENDESK_GET_ZENDESK_TICKET_BY_ID. HTTP 429 rate limit responses include Retry-After header; apply exponential backoff. Paginate by following next_page until null to avoid missing tickets.

Tool to list/enumerate Zendesk users via the Users API. Use when you need to export or iterate through the user directory without a specific name/email filter. Supports both cursor and offset pagination.

Action to reply to a Zendesk ticket by adding a comment.

Tool to search for tickets, users, organizations, and groups in Zendesk using query syntax. Use when you need to find resources across Zendesk using flexible search criteria.

Tool to search Zendesk users by email or name. At least one of 'email' or 'name' must be provided; pagination-only calls fail. Results span all roles (end-users, agents, admins); filter by 'role' client-side if needed. No org-scoped filtering — filter by 'organization_id' client-side. A response with count=0 and empty users array means no match, not an error. Use before ticket creation to confirm user identity.

Update an organization in Zendesk. After updating, re-fetch via ZENDESK_GET_ZENDESK_ORGANIZATION if downstream logic depends on current data.

Tool to update a ticket in Zendesk. Use when you need to modify ticket fields like status, priority, or subject after confirming the ticket ID.