Zendesk MCP Server by Fruggr
This MCP server provides per-user OAuth-scoped Zendesk Support and Guide operations for agents and end users, exposed as MCP tools.
Create, view, update, list, and search tickets; manage status, priority, assignee, tags, custom fields, followers, and email CCs.
Read and paginate ticket comments, add public replies or private notes, inspect/download attachments, view change history, metrics, and linked incidents.
Work queues via Zendesk views, preview macros, list SLA policies/alerts, and configure personal SLA alert thresholds.
Browse and search Help Center categories, sections, and articles; create/update/reorder/archive articles and edit large articles section-by-section.
Manage Help Center translations, drafts, labels, content tags, attachments, permission groups, and user segments; audit translation gaps.
Look up the current user, search/get users, list/get organizations, and run unified search across tickets, users, and organizations.
Support end users: list request forms, submit requests with attachments/custom fields, list/get/reply to their own requests, and mark requests solved.
Serve admin-authored context/customizations such as instructions, prompts, skills, tone, lexicon, procedures, and workflows.
Integrates with Zendesk Support & Help Center APIs, providing tools for managing tickets, articles, users, and organizations with per-user OAuth authentication.
Zendesk MCP Server
A Model Context Protocol (MCP) server that puts Zendesk inside your AI assistant, for both sides of the conversation.
For agents: it finds answers in the Help Center; drafts, updates and translates articles while keeping the languages in sync; and handles Support tickets end to end, comments, triage and image attachments included.
For your customers: the Help Center's "Submit a request" form, in plain language: pick the request, answer its questions, follow the replies, close it. See End-user mode.
For your Zendesk admins: teach the assistant how your team works, with house tone, lexicon, procedures and ready-made workflows, from a dedicated back-office, with no redeploy. See Context and customizations.
It all happens in plain language, without switching apps.
It does roughly what the Zendesk agent for Microsoft 365 Copilot does, minus the tie to one vendor: it drops into any MCP client (Claude Desktop, Claude Code, Cursor, VS Code, and the rest). And it always acts with each user's own Zendesk permissions, never a shared admin key.
What your assistant can do
Ask in natural language; the assistant works out the context and the intent, then calls the right tools on your behalf.
Find answers in the Help Center. "How do I request a software license?" or "what's the time-off policy?" surfaces the right article, matched by meaning rather than by keyword.
Create, view and update tickets without leaving the conversation: open a ticket, check its status, add a public reply or an internal note, change the priority or the assignee, mark it solved.
Summarize a ticket for a report or a quick decision. The assistant pulls the details and the full comment thread and gives you the gist in a sentence.
Read the screenshots and photos attached to a ticket. Error dialogs, UI captures and product photos are handed to your assistant's own model as images, so it can describe them or act on what they show.
Search and triage your queue in plain language: "show me my open tickets about billing from this week."
Draft and maintain knowledge-base articles. You can write a new one, or revise a large one a single section at a time, so the whole HTML body never has to round-trip through the model.
Follow your team's playbook. Instructions, prompts and skills written by your admins reach every user's assistant, whatever MCP client they use.
Submit and follow a request as a customer, not an agent. Choose between the kinds of request the vendor offers, answer only the questions that form asks, and then track it: what was replied, what you replied, and whether it's done.
Related MCP server: Zendesk MCP Server
Why this server
Most Zendesk integrations run on a shared admin API key, which hands every user full access to every ticket, and bolt on a fixed set of tools. This one is built differently.
Per-user authentication, OAuth only. Both transports use OAuth 2.1 PKCE: each user signs in with their own Zendesk credentials, so the assistant sees and touches exactly what that person is allowed to, the same scoping you get by signing into Zendesk directly. Static API tokens are deliberately not supported (why).
Two audiences, one server. Because auth is per-user rather than a shared admin key, the same install serves your agents and your customers — the end-user surface is a namespace you switch on, speaking the API path Zendesk reserves for requesters. Most Zendesk MCP servers are agent-only by construction.
Section-based article editing. For large Help Center articles, read and rewrite one section at a time (parsed by
h1/h2/h3headings) instead of shuffling the full HTML body through the assistant. On a targeted edit that cuts tokens by a factor of 10 to 100.Native multimodal attachments. Ticket images come back as native MCP image content, so the client's own model (Claude, GPT, Gemini, whichever) sees the pixels directly. No server-side vision model, no extra API key, and nothing tying you to one provider. Non-image attachments come back as text references, and both image caps are configurable.
A tool surface you can cap. Expose every operation as its own tool, group them into namespace proxies, or collapse everything into a single unified tool. You can also filter by namespace or down to read-only operations, so each context loads only the surface it needs (see Tool surface).
Your practices ship with the server. Most teams hand-distribute a skill or a rules file per AI client. Here admins author instructions, prompts and skills once, inside the server, behind a back-office reserved to Zendesk admins, and every MCP client receives them. Skills fall back to prompts on clients that don't support them yet.
Two deployment shapes, same auth story, on a lean stack. Run it on your laptop as a stdio MCP server, or deploy it as a private remote MCP server reached over HTTP, each client carrying its own user's token; underneath, the official MCP TypeScript SDK plus
zod, speaking to the Zendesk Support and Help Center (Guide) APIs.
Look elsewhere when:
You need Zendesk products outside Support and Guide (Talk, Explore analytics, Sell). Those endpoints aren't covered.
You need a single shared service account, or static API-token auth. This server supports neither, by design (see below).
What this server does not do
There is no API-token authentication. The server speaks OAuth 2.1 PKCE and
nothing else: no ZENDESK_EMAIL + ZENDESK_API_TOKEN (Basic auth) mode, in any
transport. That is deliberate, for two reasons.
API tokens are insufficiently secure. A Zendesk API token is a long-lived, static, shared secret that carries the full rights of the user who issued it. There is no per-user scoping, no short expiry, and no per-user consent or revocation. OAuth 2.1 PKCE issues per-user, revocable tokens instead.
API tokens don't scale. A single static credential can't attribute actions to individual users, and it can't be revoked granularly. It also makes a multi-user remote deployment unsafe: over HTTP it would expose the issuing user's rights to every caller.
If you specifically need an API-token or service-account mode (headless CI with a shared account, say), use one of the other Zendesk MCP servers that support it. A few are listed under Inspiration & related projects.
Quick start: local (stdio)
The default shape: one developer, one Zendesk account, OAuth 2.1 PKCE in the browser. You need Node.js >= 22 and a Zendesk instance (Support or Suite).
Install
# Run without installing
npx -y @fruggr/zendesk-mcp-server <your-subdomain>
# Or install globally
npm install -g @fruggr/zendesk-mcp-server
zendesk-mcp-server <your-subdomain>Signing in needs a Zendesk OAuth client, so register one first (next section).
Zendesk OAuth setup
Go to Admin Center → Apps and integrations → APIs → OAuth Clients
Create a public client:
Identifier:
<your-subdomain>_zendesk(or setZENDESK_OAUTH_CLIENT_ID)Redirect URL:
http://localhost:27439/callback(change the port to matchOAUTH_CALLBACK_PORT/--callback-portif you override it; Zendesk accepts several redirect URLs, one per line)
If the client restricts its allowed scopes, it needs read — plus write
unless the server runs with --read-only, which asks
Zendesk for the read scope alone.
On the first tool call the server starts the sign-in flow: it opens a browser
window and returns the authorize URL in a tool message. The call does not block
waiting for sign-in, so authenticate in the browser and then retry the request.
The token is persisted to an owner-only file and reused across restarts, so you
don't authenticate again every time your MCP client respawns the server (path
and overrides: OAUTH_TOKEN_FILE).
Several instances can run side by side — a read-write one and a --read-only
one, say — without trading credentials.
MCP client wiring
Add to your claude_desktop_config.json:
{
"mcpServers": {
"zendesk": {
"command": "npx",
"args": ["-y", "@fruggr/zendesk-mcp-server", "<your-subdomain>", "--mode", "single"]
}
}
}claude mcp add zendesk -- npx -y @fruggr/zendesk-mcp-server <your-subdomain> --mode singleAdd to your .vscode/mcp.json:
{
"servers": {
"zendesk": {
"command": "npx",
"args": ["-y", "@fruggr/zendesk-mcp-server", "<your-subdomain>", "--mode", "single"]
}
}
}Something not working? See Troubleshooting.
Quick start: remote (HTTP)
Experimental. The HTTP transport ships, but it has not yet been exercised end-to-end against a real Zendesk tenant from every MCP client. Local stdio is the supported path.
You can also deploy a private remote MCP server for one Zendesk account.
It is its own OAuth authorization server: web clients such as claude.ai and
ChatGPT add it by URL and register themselves, each user signs in to Zendesk
through it, and no shared admin key or per-client Zendesk setup is involved. Each
release also ships a hardened, distroless container image for it
(ghcr.io/fruggr/zendesk-mcp-server), signed and attested. The full guide covers
the image, the one redirect URL to add to your Zendesk OAuth client, the
master secret and grant store, --public-url behind a reverse proxy, trusted
clients, MCP client wiring, CORS and what stays the operator's job:
docs/http-deployment.md.
Tool surface
Tools are grouped into namespaces: Tickets, Help Center, Users & Organizations, Search, Connector preferences — this server's own per-user settings, such as SLA alert thresholds, never written to Zendesk — and Requests — the end-user surface, which is opt-in and covered under End-user mode. The server registers them in one of three modes, so you can trade granularity against context budget:
all: every operation as its own tool, for clients with good tool selection;namespace(default): one proxy tool per namespace, a balanced middle ground;single: a singlezendesktool, for models with limited tool slots (the connector's own preferences keep theirconnector_preferencesproxy beside it).
Proxies take { "operation": "<tool_name>", "params": { … } } and validate
params through the original schema. --namespace, --tool and --read-only
filter tools before the proxies are built, so each proxy describes only the
operations that survive.
There's one way to pick the inventory (--namespace / --tool), --mode
packages it, and --read-only narrows it. When the combination isn't obvious,
don't guess — --print-tools has the server answer what it would expose, with
no credentials needed.
Every tool with its description and its read/write mode:
docs/mcp-tools-reference.md. The flags and
worked examples: docs/configuration.md.
End-user mode
The audience for everything above is a Zendesk agent. This section is about the other one: your customers.
On the web, a customer opens a ticket through the Help Center's "Submit a request" form — they pick a kind of request, fill in the fields that kind asks for, attach a file, and later come back to read the replies. End-user mode is that same journey, in their assistant.
Who it's for
A vendor pointing their customers at an MCP server, so support happens where those customers already work. They install it themselves, sign in with their own Help Center account, and never see anything that isn't theirs.
Turning it on
The end-user tools live in the requests namespace, and it is opt-in: an
agent install shouldn't inherit tools built for someone else, and one of them
(marking a request solved) doesn't work under an agent token at all — Zendesk
accepts it and silently does nothing. help_center is worth serving alongside
it, since a customer who can search the knowledge base often doesn't need to
open a ticket in the first place.
The flags, and how to have the server print what a combination exposes: docs/configuration.md.
The journey it supports
See what kinds of request are available. The forms the vendor offers, by their customer-facing names.
Learn what one of them asks. The questions on that form, which are required, and for dropdowns the exact choices — so the assistant can gather them in conversation instead of guessing at a payload.
Submit it, with attachments.
Follow it. List their requests, read one with its whole conversation (each reply attributed, and support agents marked as such), reply back, and mark it solved when it is.
What it asks of the Zendesk account
Nothing the agent side doesn't already need, plus one thing: at least one ticket form marked visible to end users, because that is what a customer picks between. Sign-in is interactive, through a browser, by design — there is no scripted or headless path to an end-user token, which is the same protection that stops anyone else signing in as your customer.
The prerequisites in full: docs/configuration.md. The step-by-step walkthrough, written for someone who doesn't work in a terminal: docs/end-user-onboarding.md.
What a customer can't do — and shouldn't
A customer can do less than an agent. That's the point, not a gap:
Only their own tickets. Enforced by Zendesk, not by us.
No internal notes, in either direction. Agent-only notes never appear in what this surface returns — Zendesk filters them out of the requester's view of a ticket entirely.
No priority or type. Zendesk drops both when a customer sets them; triage stays with the agents.
Closing a ticket only once an agent has picked it up. Until then Zendesk won't let the requester solve it, so the tool says so rather than pretending.
No search across the ticket base, no user lookups, no views, no macros.
Context and customizations
The assistant gets two kinds of context besides tools:
Built in: the structure of your Help Center (locales, the category → section tree with IDs, segments, permission groups), so it uses real IDs instead of guessing, fetched with the caller's own token. Details: docs/help-center-context.md.
Yours: instructions, prompts and skills your Zendesk admins write from
/mcp/admin, a second connector on the same server. A guidedsetupprompt walks them through it and reminds them what is already built in. Nothing is stored in Zendesk, and no workflow is imposed: each team writes its own. Details: docs/customizations.md.
Configuration
The complete reference for the CLI flags (--mode, --namespace,
--read-only, --transport, --public-url, and so on) and the environment
variables (ZENDESK_SUBDOMAIN, OAUTH_TOKEN_FILE, PUBLIC_URL, the
attachment-vision caps) lives in
docs/configuration.md. Every variable has its own
anchor, so you can deep-link a specific setting.
Troubleshooting
Browser not opening during OAuth login, the callback port already in use, having
to re-authenticate every time, and Permission denied on the Guide-admin
endpoints are covered in docs/troubleshooting.md.
Restart with LOG_LEVEL=debug for the full OAuth flow trace.
Development
Setting up the repo, the toolchain, dev mode and how to test a PR branch are
covered in CONTRIBUTING.md. Architecture
and code-style conventions live in AGENTS.md.
FAQ
Which Zendesk products are supported? Zendesk Support (tickets, users, organizations) and the Help Center / Guide (articles, sections, categories, translations, labels, content tags, segments, attachments). Talk, Explore and Sell are out of scope.
Do I need a Zendesk admin API key? No, and the server doesn't support one. Each user authenticates with their own credentials and the server acts with exactly their permissions (why).
Is it safe to run via npx?
Releases are published from CI via npm Trusted Publishing (OIDC), so each version
carries a build provenance attestation you can verify on its
npm page. No secrets
are ever logged by the server.
Contributing
Pull requests are welcome, AI-assisted ones included, as long as the human author has read and validated every line. The guide, the author checklist and the review workflow are in CONTRIBUTING.md.
Versions follow SemVer and are released automatically from
Conventional Commits; the history is in
CHANGELOG.md.
Inspiration & related projects
This project was built with reference to:
License
Built and maintained by Digital4better for the Fruggr project.
Available Tools
63 toolsadd_private_noteAdd Private NoteA
Add an internal note (not visible to requester) to a ticket, optionally with file attachments (uploaded via the Zendesk Uploads API and carried on the note). The note is appended to the ticket thread; use add_public_comment instead when the reply should be visible to the requester.
Inputs: body (required), ticket_id (required), attachments.
add_public_commentAdd Public CommentA
Add a public comment (visible to requester) to a ticket, optionally with file attachments (uploaded via the Zendesk Uploads API and carried on the comment). The comment is appended to the ticket thread and emails the requester; use add_private_note instead for an internal, agent-only note.
Inputs: body (required), ticket_id (required), attachments.
add_request_commentReply on My RequestA
Reply on one of the signed-in user's own requests, optionally attaching files. Returns confirmation with the request's status after the reply, which matters because replying on a request that was already solved REOPENS it — Zendesk moves it back to open, and the user should know that is what their message did. The comment is always public: end users cannot post internal notes, and Zendesk silently ignores any attempt to mark one private. A closed request rejects new comments outright. Agents replying to a requester want add_public_comment, or add_private_note for an internal note.
Inputs: body (required), request_id (required), attachments.
archive_articleArchive Help Center ArticleADestructiveIdempotent
Archive (soft-delete) a Help Center article: it is removed from the Help Center but can be restored from the Guide admin UI. Returns a confirmation message; the article and all its translations become invisible to end users. This is the only removal the Zendesk API offers — permanent deletion is not available via the API (do it from the Guide admin UI). To only hide an article temporarily while keeping it in the knowledge base, use update_article with draft: true (unpublish) instead. Guarded by a required confirm flag.
Inputs: confirm (required), article_id (required).
compare_translationsCompare Article TranslationsARead-onlyIdempotent
Compare two locales of the same article to decide whether the target translation needs work, reporting independent signals instead of one ambiguous verdict. (1) Header — a "Freshness" verdict for the target, derived from the two translations' updated_at timestamps: if the source was edited after the target it is "likely behind, review recommended" (with the day gap), otherwise "up to date". This is the primary staleness signal and is always available. (2) Zendesk's own per-translation "outdated" flag for the target ("yes"/"no"/"unknown"), shown as a secondary overlay: it is only set through Guide's native "mark out of date" workflow and NOT by API edits, so a "no" does not by itself mean current — prefer Freshness. (3) A global structure check (section count and heading-tag sequence); on mismatch the header warns the per-index rows may be misaligned. (4) A per-section table matched by index, status "ok" (present in both), "missing" (present in source, absent in target) or "extra" (present in target, absent in source). (5) Per-section source/target word counts, INFORMATIONAL ONLY: a length difference between languages is normal and is deliberately NOT flagged as a divergence — do not read a word-count gap as an edit regression or staleness. Read-only; performs three Help Center GET calls (both translations plus the translations list for the outdated flag).
Inputs: article_id (required), source_locale (required), target_locale (required).
configure_sla_alertsConfigure SLA Alert ThresholdsAIdempotent
Read or change your personal SLA alert thresholds: how close to a breach a running SLA metric must be to count as WARNING or CRITICAL. Called with no parameters it changes nothing and reports the thresholds in force for you and where each comes from (your preference or the server default). With parameters it saves your own thresholds per metric family (reply: first_reply_time and next_reply_time; wait: every other metric) on this MCP server — nothing is written to Zendesk — and they then drive list_sla_alerts and the Urgency line of get_ticket / search_tickets for you only; reset goes back to the server defaults. When the operator turned personal thresholds off (SLA_USER_PREFERENCES=off), only the read works and the reply says so.
Inputs: reset, wait_warning, reply_warning, wait_critical, +1 more.
create_articleCreate Help Center ArticleA
Create a new article in a section and return the created article with its id. The locale becomes the article's source_locale. Requires a permission_group_id (use list_permission_groups to find available IDs). To add content in other locales afterwards, use create_article_translation.
Inputs: body (required), title (required), section_id (required), permission_group_id (required), +7 more.
create_article_attachmentCreate Article AttachmentA
Upload a file to a Help Center article and return the created attachment (its id, file name, content type, size and content URL). Not idempotent: calling it again uploads another copy rather than replacing the previous one. This is for article assets — for files on support tickets use get_ticket_attachments, and to see an article's existing attachments use list_article_attachments.
Inputs: file_name (required), article_id (required), file_base64 (required), content_type.
create_article_translationCreate Article TranslationA
Create a translation for an existing article in a specific locale. The article must already exist (create it with create_article); this adds a new localized version and returns the created translation (locale, title, draft state). The target locale must not already have a translation — use update_article_translation to modify an existing one, and list_article_translations to see which locales exist. Provide the full HTML body.
Inputs: body (required), title (required), locale (required), article_id (required), +1 more.
create_content_tagCreate Content TagA
Create a new content tag for Guide articles. Content tags are end-user visible labels that help readers discover related articles; this returns the created tag with its id. Check list_content_tags first (filter by name_prefix) to avoid duplicates, then attach the new id via the content_tag_ids parameter of create_article or update_article. For internal search-ranking labels that are not shown to end users, use article labels (list_labels) instead.
Inputs: name (required).
create_requestSubmit a RequestA
Submit a new support request as the signed-in user — the MCP equivalent of the Help Center's "Submit a request" form. Returns the created request with its number, which the user can then follow with list_requests and get_request. The submission is checked before sending: an unknown form id, a missing field the form marks required, or a dropdown answer the form does not offer is refused here rather than sent, because Zendesk would answer 201 having silently substituted the default form, accepted an empty subject or dropped the unrecognised value. Call get_request_form first to learn which fields to gather. Priority and type cannot be set by an end user — Zendesk drops them — so triage is left to the agents. This posts as the authenticated user; an agent opening a ticket on someone else's behalf wants create_ticket instead.
Inputs: body (required), subject (required), form_id, attachments, +1 more.
create_ticketCreate Zendesk TicketA
Create a new Zendesk support ticket with subject, description, and optional requester, organization, priority, type, assignee and tags. The description becomes the first public comment of the ticket, and the new ticket id is returned. After creation, use update_ticket to change status or assignee, add_public_comment or add_private_note to reply, and manage_tags to adjust tags. Without requester_id the ticket is filed on your behalf; the organization must be one the requester is a member of, and a value Zendesk did not keep is reported as not applied. Look up valid requester_id / assignee_id via search_users, group_id via your Zendesk admin settings. Discover custom field ids and their accepted option values with list_ticket_fields.
Inputs: subject (required), description (required), tags, type, +6 more.
find_translation_gapsFind Help Center Translation GapsARead-onlyIdempotent
Audit the Help Center tree for a target locale and report every category and section that has no translation, or one that is still an unpublished draft. Use it before or after translating articles: an article published in a second locale is unreachable while its parent section only exists in the source locale. Listing sections in that locale cannot answer this — a node with no translation is simply absent, without saying why, and a node whose translation is an unpublished draft may still be listed under its draft name — so this audit reads the draft flag on each node instead of trusting that listing. Costs two listings and covers up to 100 categories and 100 sections; past that only the first page of each level is audited, and the report says so — pass category_id to narrow it. Fix what it reports with set_section_translation / set_category_translation.
Inputs: locale (required), category_id.
get_articleGet Help Center ArticleARead-onlyIdempotent
Retrieve an article by ID with full body content. For large articles, prefer get_article_outline + get_article_section to save tokens. Optionally specify locale for a translated version. Returns body (HTML), metadata, source_locale, and list of available translations.
Inputs: article_id (required), locale.
get_article_outlineGet Article OutlineARead-onlyIdempotent
Return a compact outline of an article (list of sections delimited by h1/h2/h3, with word counts) for the given locale (defaults to source_locale). Includes available translations with their outdated status. Use get_article_section to fetch a specific section.
Inputs: article_id (required), locale.
get_article_sectionGet Article SectionARead-onlyIdempotent
Retrieve the content of a single section of an article in a given locale. Use get_article_outline first to discover section indexes. Default format="html" for round-trip safety. Pass format="markdown" only for human review — the Markdown representation is lossy on some structures ( with , tables with multi- cells are kept as raw HTML to limit the damage, but do not round-trip markdown content back through update_article_section). The "## [n] hN: ..." line at the top is a label, not part of the content.
Inputs: locale (required), article_id (required), section_index (required), format.
get_current_userGet Current Zendesk UserARead-onlyIdempotent
Get the currently authenticated Zendesk user. Useful to verify identity and permissions.
Inputs: no named parameters.
get_linked_incidentsGet Linked IncidentsARead-onlyIdempotent
Get all incident tickets linked to a problem ticket. Returns the list of incidents that reference the given problem (Zendesk problem/incident relationship); useful to gauge a problem's blast radius before resolving it. Each ticket's description is cut to a 200-character excerpt; get_ticket returns it in full.
Inputs: problem_id (required).
get_organizationGet Zendesk OrganizationARead-onlyIdempotent
Retrieve a single organization by its numeric id. Returns full details (name, tags, domains, notes) — more than the name/id that search or list_organizations surface. Use list_organizations to browse or search for a name-based lookup.
Inputs: organization_id (required).
get_requestGet My RequestARead-onlyIdempotent
Read one of the signed-in user's own requests, optionally with the whole conversation on it. Returns the request's status, the form it was submitted on, whether the user may mark it solved, and — with include_comments — every public message on it, each labelled with its author and marked when that author is a support agent. Internal agent notes are never returned on this path, so nothing agent-private can leak through it. A request belonging to someone else answers "permission denied" rather than "not found", so a denial here does not tell you whether that id exists. Find the number with list_requests; agents reading a ticket they do not own want get_ticket instead.
Inputs: request_id (required), include_comments.
get_request_formGet Request FormARead-onlyIdempotent
Read what one request form actually asks for, so the user can be walked through it question by question. Returns the fields on that form in the order the portal shows them, each with its customer-facing label, whether it is required, and for dropdowns the exact values Zendesk accepts — plus any conditional rules ("if Type is Bug, Version becomes required"). Joins the form definition with the field definitions, because the form itself carries only field ids. Use it after list_request_forms and before create_request; omit form_id to describe the account default form. Conditional fields are reported as rules rather than resolved, since which ones apply depends on answers not yet given.
Inputs: form_id.
get_ticketGet Zendesk TicketARead-onlyIdempotent
Retrieve a Zendesk ticket by ID, including its live SLA state (per-metric stage and breach countdown) when an SLA policy applies, plus its comments if requested. Returns ticket details (subject, status, priority, assignee, tags, description) and optionally all comments/internal notes. It also lists the followers and email CCs on the ticket, resolved to names, which is how you tell who a ticket update actually reaches: followers are notified of updates including internal notes, email CCs take part in the public correspondence, and being the requester makes someone neither. Both lists come back empty on an account where the "CCs and followers" setting is not enabled in Zendesk. The per-ticket Show endpoint exposes no SLA, so the SLA block is resolved via a scoped search and may be absent for a very high-volume requester or a just-updated ticket; SLA targets and policy conditions live in list_sla_policies. When the operator turned SLA push on, the SLA block also carries an Urgency line naming the most urgent running metric as BREACHED, CRITICAL or WARNING against the caller's alert thresholds. This returns the ticket as it stands now; for the history of changes behind that state (who changed what, and when), use get_ticket_history. The comment thread is appended in one block — the first page of comments Zendesk returns, cut past the response character limit — so on a long ticket read it with list_ticket_comments, which pages the comments and returns the newest first.
Inputs: ticket_id (required), include_comments.
get_ticket_attachmentsGet Zendesk Ticket AttachmentsARead-onlyIdempotent
Download the files attached to a ticket's comments so you can look at them: images come back embedded inline, other files as text references (file name, id, content type, size, download URL). Use it to see an attached image, or a file's name, size and download URL (only images come back with their content); to learn which files a comment carries, list_ticket_comments already shows their ids. Pass those ids as attachment_ids to fetch only them, or omit it to collect the attachments of the ticket's comments, read up to a page cap (10 pages of 100 comments by default); past it the response says later comments were not scanned, and attachment_ids fetches their files. An image over the per-image size limit or past the embedded-image cap comes back as a reference saying why, and once the response size budget is spent the remaining files are counted as omitted.
Inputs: ticket_id (required), attachment_ids.
get_ticket_historyGet Zendesk Ticket HistoryARead-onlyIdempotent
Read a ticket's change history — its audit trail — as a chronological, oldest-first timeline of who changed what and when. Each entry shows the actor (name and id) and the channel, then the field changes that update carried (status, priority, assignee, group, tags, custom fields) as before → after, with assignee/requester/group ids resolved to names. Comments appear as one-line presence markers (public comment vs internal note added), not their text — fetch the bodies with list_ticket_comments (or get_ticket(include_comments=true) for a short thread). Purely system-generated notification events (trigger emails, collaborator/CC notifications, pushes) are filtered out — note this filters notification delivery, not CC-list edits, which are shown as changes — and an update carrying only such events produces no entry, so the timeline stays a readable narrative rather than a raw log. Use it to answer "what happened on this ticket?", "why was it reassigned?" or "when did it go to pending?", reading oldest-first so the founding context is not missed. Read-only, and cursor-paginated oldest-first: pass the returned cursor to page a long-lived ticket toward its most recent changes.
Inputs: ticket_id (required), cursor, page_size.
get_ticket_metricsGet Ticket MetricsARead-onlyIdempotent
Read the elapsed-time metrics (reply, resolution, wait and on-hold times, reopens, replies) of up to 100 tickets in one call. Each ticket gets first reply, first and full resolution, agent wait, requester wait and on-hold times in calendar and business minutes, plus reopen, reply, assignee-station and group-station counts and the key timestamps (assigned, solved, latest comment). These are facts Zendesk records on every ticket, so they are available even on tickets no SLA policy covers; they say how long things took, not how long is left — for the live breach countdown use get_ticket, search_tickets or list_sla_alerts. A duration still running reads "not reached yet". Read-only; tickets the caller cannot see are reported as such rather than failing the call.
Inputs: ticket_ids (required).
get_userGet Zendesk UserARead-onlyIdempotent
Retrieve a single user by their numeric id. Returns the full user record (name, email, role, organization, tags). Use search_users when you only have a name or email, or get_current_user for the authenticated identity.
Inputs: user_id (required).
get_view_ticketsGet Tickets In A ViewARead-onlyIdempotent
Read the tickets inside a Zendesk view, in the view's own configured sort order — the same order the agent sees in the Zendesk UI — which is the natural way to work a named queue like "Unassigned tickets" or "Breaching today". Accepts the view by title or by numeric id (discover both with list_views); a title is matched case-insensitively against the agent's active views, and on no match the available titles are returned so you can retry in one step. Tickets come back with the same fields as list_tickets and are cursor-paginated. Each ticket an SLA policy covers also shows its next SLA breach: only the one metric closest to breaching, read from the view itself at no extra cost (use get_ticket or search_tickets for every metric). sort_by/sort_order override the view's order when you want a different cut. Each ticket's description is cut to a 200-character excerpt; get_ticket returns it in full.
Inputs: view (required), cursor, sort_by, page_size, +1 more.
list_article_attachmentsList Article AttachmentsARead-onlyIdempotent
List all attachments for an article. Returns attachment metadata only (id, file name, content type, size, URL), not the file bytes; both inline and block attachments are included. This is for Help Center articles — for attachments on support tickets use get_ticket_attachments instead. Upload new files with create_article_attachment.
Inputs: article_id (required).
list_articlesList Help Center ArticlesARead-onlyIdempotent
List articles (metadata only, no body). Use get_article for full content. Optionally filter by section ID and locale. Supports sort_by ("title", "created_at", "updated_at") and include_translations: true to show available translation locales per article. Note: include_translations must be re-sent on each paginated request.
Inputs: cursor, locale, sort_by, page_size, +3 more.
list_article_translationsList Article TranslationsARead-onlyIdempotent
List all available translations for an article (metadata only, no body: locale, title, draft, updated_at). Use get_article with locale for full translated content.
Inputs: article_id (required).
list_categoriesList Help Center CategoriesARead-onlyIdempotent
List all Help Center categories. Categories are the top level of the Guide hierarchy (category → section → article); each entry includes its id, name and locale. Results are cursor-paginated. Pair a returned category id with list_sections to drill down, then list_articles to reach articles. Pass a locale to read category names in that translation.
Inputs: cursor, locale, page_size.
list_category_translationsList Category TranslationsARead-onlyIdempotent
List the translations of a Help Center category: for each locale, the localized name, whether a description is set, and whether the translation is published or still a draft. Reach for this when a category looks wrong in a locale, because list_categories with that locale cannot settle it: a category with no translation is omitted from it, while a category whose translation is an unpublished draft may still be listed there under the draft name — so appearing in that listing does not mean published, and the draft flag here is what decides. Fix either case with set_category_translation; to sweep every category and section at once, use find_translation_gaps.
Inputs: category_id (required).
list_content_tagsList Content TagsARead-onlyIdempotent
List Guide content tags, which are end-user-visible labels that help readers find related articles. Results are cursor-paginated (follow the returned cursor to enumerate the full list) and sorted by name by default. Pass name_prefix to look a tag up by the start of its name — do this before create_content_tag to reuse an existing tag rather than fragment the taxonomy. For internal, non-end-user search labels, see list_labels instead.
Inputs: cursor, sort_by, page_size, sort_order, +1 more.
list_labelsList Article LabelsARead-onlyIdempotent
List all article labels, with each label's id and name. Labels improve Help Center search ranking and are not visible to end users. Attach them to an article through the label_names parameter of create_article or update_article, reusing an existing label rather than coining a near-duplicate. Labels are not content tags: for the end-user-visible tags readers browse by, use list_content_tags.
Inputs: no named parameters.
list_macrosList Zendesk MacrosARead-onlyIdempotent
List the active macros available to the authenticated user. A macro bundles a canned reply and/or a set of field changes (status, priority, assignee, group, tags, custom fields) an agent applies to a ticket in one gesture; this returns each macro id, title, description, availability scope, and its ordered list of actions, offset-paginated. Results are scoped by per-user OAuth to what the current user can see, so no shared admin key is needed. Pass a macro id from here to preview_macro_diff to preview its effect on a specific ticket.
Inputs: page, per_page.
list_organizationsList Zendesk OrganizationsARead-onlyIdempotent
List all organizations with pagination. Returns the name and id of each organization plus basic fields; results are cursor-paginated. Use get_organization with an id for full details (tags, domains, notes), or search for query-based lookups by name. Organizations group end users and can be referenced when creating or filtering tickets.
Inputs: cursor, page_size.
list_permission_groupsList Permission GroupsARead-onlyIdempotent
List all Guide permission groups, which control who can edit and publish an article, with each group's id, name and whether it is built-in. Use this to find the permission_group_id required when creating articles with create_article, or to change it with update_article; to control who can view an article instead, use list_user_segments. Zendesk restricts this list to Guide admins and Help Center managers: without that role the call fails, and you can reuse the permission_group_id of an existing article read with get_article.
Inputs: no named parameters.
list_promoted_articlesList Promoted Help Center ArticlesARead-onlyIdempotent
List the promoted ("featured") Help Center articles — the small, editorially-curated set surfaced at the top of their sections. Returns metadata only (no body); use get_article for full content. COST: the Help Center API has no server-side promoted filter, so this scans article pages (one Zendesk API request per page, up to ARTICLE_RESOURCES_SCAN_MAX_PAGES, default 20) and filters client-side — potentially costly on a large Help Center. Each call performs a fresh, uncached scan, so avoid calling it repeatedly. On a very large Help Center some promoted articles may be omitted, and both the omission and the number of pages scanned are flagged in the output. Lists the default locale. To promote or unpromote an article, use update_article with promoted (requires Help Center admin / Guide admin rights).
Inputs: no named parameters.
list_request_formsList Request FormsARead-onlyIdempotent
List the kinds of request you can submit to this Zendesk — the forms a customer picks between on the Help Center, such as "Report a bug" or "Feature request". Returns each form's customer-facing name, its numeric id, and which one is the account default. Call this first when opening a request so the user chooses the right kind, then get_request_form to learn what that form asks for. Only forms that are active and visible to end users are listed; an account with a single form returns just that one. This is the end-user view: agents picking a form for someone else want list_ticket_fields and create_ticket instead.
Inputs: no named parameters.
list_requestsList My RequestsARead-onlyIdempotent
List the requests the signed-in user submitted, newest activity first by default. Returns each request with its number, subject, status and whether the user can mark it solved; pass a query to search their own requests by free text instead of listing all of them. Scoped to the caller by Zendesk itself — it can never return anyone else's tickets — which is why an agent calling this sees only requests they raised, not their queue. Use get_request to read one with its full conversation. Agents wanting a queue want list_tickets, search_tickets or get_view_tickets instead.
Inputs: page, query, status, sort_by, +2 more.
list_sectionsList Help Center SectionsARead-onlyIdempotent
List Help Center sections. Sections are the middle level of the Guide hierarchy (category → section → article) and group related articles; each entry includes its id, name, category_id and locale. Results are cursor-paginated. Pass category_id to list only one category's sections (ids come from list_categories), then use a section id with list_articles. Pass a locale to read section names in that translation.
Inputs: cursor, locale, page_size, category_id.
list_section_translationsList Section TranslationsARead-onlyIdempotent
List the translations of a Help Center section: for each locale, the localized name, whether a description is set, and whether the translation is published or still a draft. Reach for this when a section looks wrong in a locale, because list_sections with that locale cannot settle it: a section with no translation is omitted from it, while a section whose translation is an unpublished draft may still be listed there under the draft name — so appearing in that listing does not mean published, and the draft flag here is what decides. Fix either case with set_section_translation; to sweep every category and section at once, use find_translation_gaps.
Inputs: section_id (required).
list_sla_alertsList SLA AlertsARead-onlyIdempotent
List the SLA metrics that are breached or about to breach across a set of tickets, most urgent first. Pick the set with exactly one of scope (mine, my_groups or all unsolved tickets), view (a Zendesk view by title or id) or query (any Zendesk ticket search). Each running metric that is breached, or closer to its breach than its CRITICAL or WARNING threshold, is one line under BREACHED, CRITICAL or WARNING, soonest breach first, with the ticket id, metric, time left or overdue, subject, status and priority; paused and achieved metrics never appear. Thresholds are per metric family (reply metrics vs every other metric), the caller's own when set with configure_sla_alerts, else the server's; warning and critical override them for this call only. When no set is given the client may be asked which one, otherwise it defaults to the caller's own tickets and says so; on a user's first call the thresholds are shown too (in that form, or in the reply), and thresholds edited there apply to this call only. Never changes Zendesk; its one side effect is remembering, in this connector's own per-user state, that the user has been shown the thresholds. Results are cached per user for about a minute, and a set larger than a few hundred tickets is read in part, with a notice. For one ticket's full SLA state use get_ticket; for elapsed times use get_ticket_metrics.
Inputs: view, query, scope, within, +3 more.
list_sla_policiesList SLA PoliciesARead-onlyIdempotent
List the configured SLA policies with their filter conditions and per-priority reply/resolution targets. Use this to explain why a given target applies to a ticket and to reconstruct deadlines deterministically instead of hard-coding the policy matrix. Requires an admin token (or a custom role granted the SLA-management permission); a standard agent token gets 403 here, though it can still read live per-ticket SLA via get_ticket / search_tickets.
Inputs: page, per_page.
list_ticket_commentsList Zendesk Ticket CommentsARead-onlyIdempotent
Read a ticket's conversation — public replies and internal notes with their full bodies — one cursor-paginated page at a time, newest comment first. Each entry carries the comment id, the author resolved to a name, the timestamp, whether it is public or internal, and the ids of any attached files. Prefer this over get_ticket(include_comments=true) whenever a thread is long or you only need the latest exchange: get_ticket appends the thread as one unpaginated block and cuts it past the response character limit, which drops the most recent comments first. Keep sort_order "desc" (the default) to read the latest reply first and follow the returned cursor to walk further back in time, or pass "asc" to replay the conversation forward from the ticket's opening description. For who changed which field and when — without comment bodies — use get_ticket_history; to download the attached files themselves, pass the attachment ids shown here to get_ticket_attachments.
Inputs: ticket_id (required), cursor, page_size, sort_order.
list_ticket_fieldsList Ticket FieldsARead-onlyIdempotent
List the ticket field definitions configured on this Zendesk (both system fields and custom fields), returning each field's id, type, whether it is active/required, and — for dropdown and multiselect fields — the valid option values. Use this to discover the numeric field ids and accepted option tags that create_ticket and update_ticket expect in their custom_fields argument, so a natural-language intent ("set severity to High") maps to the right id and a value Zendesk will accept instead of a blind guess. Read-only reference lookup; cursor-paginated in Zendesk's default field order.
Inputs: cursor, page_size.
list_ticketsList Zendesk TicketsARead-onlyIdempotent
List tickets with cursor-based pagination, in Zendesk's default order (ascending ticket id), not by recency. Page size is controlled by page_size (not per_page, which is the offset-based parameter used by search_tickets); paginate by passing the returned cursor. To find tickets by recency or any other criterion, use search_tickets with a query. Each ticket's description is cut to a 200-character excerpt; get_ticket returns it in full.
Inputs: cursor, page_size.
list_user_segmentsList User SegmentsARead-onlyIdempotent
List all Help Center user segments, which control article visibility (who can view an article), with each segment's id, name, user type and whether it is built-in. Use the id as user_segment_id when creating or updating articles with create_article or update_article; to control who can edit an article instead, use list_permission_groups. Zendesk restricts this list to Guide admins and Help Center managers: without that role the call fails, and you can reuse the user_segment_id of an existing article from get_article, or omit it on create_article to make the article visible to everyone.
Inputs: no named parameters.
list_viewsList Zendesk ViewsARead-onlyIdempotent
List the agent's active Zendesk views — the saved ticket queues ("Unassigned tickets", "My open tickets", "Breaching today") the agent sees in the Zendesk UI — each with its current ticket count so you can tell at a glance where the workload sits. Views are per-agent scoped, so per-user auth returns exactly the queues this agent can see, with no shared key. Counts come from Zendesk's cache and can lag by up to about an hour (shown as "(count updating)" while a fresh value is still being computed); pass a view's title or id to get_view_tickets to read the tickets inside it.
Inputs: cursor, page_size.
manage_tagsManage Ticket TagsADestructiveIdempotent
Add or remove tags on a ticket. Performs an incremental read-modify-write: it fetches the ticket's current tags, adds those in add and deletes those in remove, then saves the merged set — tags you don't list are left untouched and duplicates are collapsed. Adding a tag already present, or removing one that is absent, is a no-op (idempotent). Returns the ticket's full tag set after the update. Use this for incremental tag edits; to overwrite the entire tag set at once, or to change tags alongside other fields, use update_ticket instead. Find the ticket id via search_tickets or list_tickets.
Inputs: ticket_id (required), add, remove.
mark_request_solvedMark My Request SolvedADestructiveIdempotent
Close one of the signed-in user's own requests, for when they consider it resolved. Returns the request's status read back from Zendesk, not merely an acknowledgement: this operation is refused unless the request is actually solvable by its requester, because Zendesk answers 200 and changes nothing when it is not, which would otherwise be reported as success. A request is only solvable once an agent has been assigned to it, so a brand-new or unassigned one cannot be closed this way — leave it, or reply with add_request_comment to say it is no longer needed. Marking an already-solved request solved again is a no-op. Agents closing a ticket want update_ticket with status solved.
Inputs: request_id (required).
preview_macro_diffPreview a Macro Diff on a TicketAIdempotent
Preview the exact changes a macro would make to a specific ticket, as a before → after diff, WITHOUT saving anything. Orchestrates two reads — the ticket's current state and Zendesk's macro-apply preview (which returns the whole resulting ticket) — and returns only the fields the macro actually changes (status, priority, assignee, group, tags, custom fields) plus the canned reply with its public/internal flag; unchanged and identity fields are omitted. Nothing is committed: to apply it, follow up with update_ticket for the field changes and add_public_comment or add_private_note for the reply. This deliberate two-step keeps the mutation explicit and reviewable rather than hidden. Find macro ids via list_macros and the ticket id via search_tickets or list_tickets.
Inputs: macro_id (required), ticket_id (required).
reorder_articleReorder Help Center ArticleADestructiveIdempotent
Reorder an article within its current section by moving it relative to its siblings (top, bottom, or before/after another article), and return whether the new order was applied. This is the reliable way to satisfy "put this article first/last" requests: it writes the minimal set of article positions needed to make the order deterministic, because Zendesk leaves newly created articles tied at position 0 where a plain position update is silently ambiguous. It does NOT move the article to a different section — use update_article with section_id for that. Zendesk exposes no way to read whether a section is manually or automatically sorted, so when the section is sorted automatically (by date or alphabetically) the position writes are ignored; this tool detects that after the fact and returns guidance to switch the section to manual ordering in Guide. A move may reposition several neighbouring articles; when that count exceeds a configurable safety threshold the call is refused unless confirm is set to true.
Inputs: target (required), article_id (required), confirm, normalize, +1 more.
searchZendesk Unified SearchARead-onlyIdempotent
Search across tickets, users, and organizations. Supports filters like "type:ticket status:open", "type:user role:agent". Returns total count and paginated results (25 per page by default, up to 100); a ticket description is cut to a 200-character excerpt, which get_ticket returns in full. Organization results include name and ID only — use get_organization for full details (tags, domains, details).
Inputs: query (required), page, per_page.
search_articlesSearch Help Center ArticlesARead-onlyIdempotent
Full-text search across Help Center articles (metadata only, no body). Use get_article for full content. Supports locale filtering. Returns total count.
Inputs: query (required), page, locale, per_page.
search_ticketsSearch Zendesk TicketsARead-onlyIdempotent
Search tickets using Zendesk query syntax, returning each result with its live SLA state (per-metric stage and breach countdown) when an SLA policy applies. Examples: "status:open assignee:me", "priority:urgent ticket_type:incident". Returns total count, so queue triage like "breaching today" works without a per-ticket fetch. When the operator turned SLA push on, each SLA block also carries an Urgency line (BREACHED, CRITICAL or WARNING) against the caller's alert thresholds. Each ticket's description is cut to a 200-character excerpt; get_ticket returns it in full.
Inputs: query (required), page, per_page.
search_usersSearch Zendesk UsersARead-onlyIdempotent
Search for users by name, email, or other criteria using Zendesk search query syntax, returning each match's id, name, email, role, active flag and organization id, with the total count. Use it to turn a person into the numeric user id other tools take, such as the followers of update_ticket; a name can match several users, so confirm the right one. With an id already in hand, call get_user to read that one user, and for your own account get_current_user. Prefer it over search when you want users only: search mixes in tickets and organizations unless the query says type:user.
Inputs: query (required), page, per_page.
set_category_translationCreate or Update a Category TranslationADestructiveIdempotent
Create or update the translation of a Help Center category in one locale, and return the resulting translation (locale, localized name, draft state). Creates the translation when the locale has none and updates it otherwise, so no listing call is needed first; only the fields you pass are written, which makes "publish this draft" a single draft: false. Use it to make a category reachable in a locale where its sections or articles are already translated — a gap find_translation_gaps reports and list_categories cannot explain.
Inputs: locale (required), category_id (required), name, draft, +1 more.
set_section_translationCreate or Update a Section TranslationADestructiveIdempotent
Create or update the translation of a Help Center section in one locale, and return the resulting translation (locale, localized name, draft state). Creates the translation when the locale has none and updates it otherwise, so no listing call is needed first; only the fields you pass are written, which makes "publish this draft" a single draft: false. Use it to make a section reachable in a locale where its articles are already translated — a gap find_translation_gaps reports and list_sections cannot explain.
Inputs: locale (required), section_id (required), name, draft, +1 more.
update_articleUpdate Help Center ArticleADestructiveIdempotent
Update article metadata only (draft, promoted, labels as label_names, content tags as content_tag_ids, visibility as permission_group_id / user_segment_id, section_id, sort position, author, etc.) and return the updated article; fields you omit keep their value. Does NOT update content (title, body) — use update_article_translation for that, or update_article_section to edit one section. To put an article first or last in its section, prefer reorder_article: a plain position write is ambiguous while articles share position 0. To remove an article from the Help Center, use archive_article (a Guide admin can restore it); draft: true only unpublishes it.
Inputs: article_id (required), draft, position, promoted, +6 more.
update_article_sectionUpdate Article SectionADestructiveIdempotent
Replace the body of one article section, heading excluded (the heading is kept as is), in a given locale, leaving the rest of the article intact. The server fetches the current body, replaces the targeted section, and PUTs the full reconstructed body via the Translations API. A leading heading in content that repeats the section heading is removed; any other h1-h3 in content creates new sections and shifts the later indexes (the response warns). section_index 0 is the heading-less intro, which exists only when the body has text before its first heading. Default format="html" for fidelity. Use format="markdown" only when you control the input and know it does not rely on structures that round-trip poorly (code blocks with line breaks, tables with multi-paragraph cells).
Inputs: locale (required), content (required), article_id (required), section_index (required), +1 more.
update_article_translationUpdate Article TranslationADestructiveIdempotent
Update article content (title, body) in a specific locale. For targeted edits on one or a few sections, prefer update_article_section — this tool replaces the FULL body and re-sends the entire article on each write. Use the article's source_locale (from get_article) for the default language, or another locale for translations.
Inputs: locale (required), article_id (required), body, draft, +1 more.
update_ticketUpdate Zendesk TicketADestructiveIdempotent
Update an existing ticket (status, priority, type, requester, organization, assignee, group, subject, tags, custom fields, followers, email CCs). Only the fields you pass are changed, except that Zendesk may move the ticket to an organization of a new requester; the updated ticket is returned. The organization must be one the requester is a member of. A requested requester or organization that Zendesk did not set is reported as not applied; an organization Zendesk moved along with a new requester, when organization_id was not passed, is reported as changed with the requester. Followers and email CCs are incremental instead of replacing: pass { add, remove } lists of user ids, and the response reports who is subscribed afterwards. Zendesk applies those two silently, so a requested change it did not apply comes back listed as unconfirmed rather than raised as an error. Setting tags here replaces the whole tag set — use manage_tags to add or remove individual tags without overwriting the rest. This tool does not post replies: use add_public_comment or add_private_note for that. Find the ticket id via search_tickets or list_tickets.
Inputs: ticket_id (required), tags, type, status, +9 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 63 tools
The tool families are mostly well described, but there are several overlapping areas: generic `search` versus `search_tickets`/`search_users`/`search_articles`, agent-facing ticket tools versus end-user request tools, and article metadata updates versus translation/section updates. The descriptions do a lot to clarify boundaries, but with 63 tools an agent will still face non-trivial misselection risk.
Tool names consistently use snake_case verb_noun patterns such as `create_ticket`, `get_article`, `update_article_section`, `list_sla_alerts`, and `configure_sla_alerts`. The few generic names like `search` do not break the overall convention.
At 63 tools, the surface is far beyond a well-scoped MCP toolset and is likely to overwhelm agents with selection overhead. Even though Zendesk is a broad product, this many tools would benefit from consolidation or higher-level orchestration.
Coverage is very broad: tickets, comments, history, attachments, metrics, SLA, views, macros preview, users, organizations, Guide articles/translations, content tags, labels, permission groups, segments, and end-user requests are all represented. Minor gaps remain around direct user/organization mutation, section/category CRUD, group listing, and applying macros without composing follow-up calls.
Maintenance
Related MCP Connectors
MCP server connecting AI agents to 100+ apps (Gmail, Slack, Notion, GitHub) via one-click OAuth.
MCP server for your apps' tools and custom tools, plus hosted AI agents and approval-gated workflows
Model Context Protocol server for the Apideck Unified API. Connect any MCP-compatible agent framework to 100+ accounting systems, HRIS platforms, file storage providers, and more through one integration. More information https://www.apideck.com/mcp-server
One OAuth connector for your team's remote MCP servers, with Tool Packs, guardrails and logs.
Related MCP Servers
- AlicenseBqualityDmaintenanceA server implementation that provides Claude AI with the ability to interact with Zendesk ticketing systems through various functions including retrieving, searching, creating, and updating tickets.7287 npm16MIT
- AlicenseNot gradedqualityDmaintenanceA lightweight, AI-native server that enables GPT-based AI agents to fetch real-time customer and organization context from Zendesk APIs dynamically.ISC
- AlicenseAqualityBmaintenanceEnables AI-powered ticket analysis and comprehensive Zendesk API integration via the Model Context Protocol, supporting both API token (CLI) and OAuth 2.1 (web) authentication modes.5599 npm5MIT
- AlicenseAqualityCmaintenanceA Model Context Protocol server that enables AI assistants like Claude to interact with Zendesk Support through natural language for searching, creating, updating, and managing tickets.41287 npm3MIT