Connect an AI assistant to your help center

Nov 11, 2026 18 min read

View as MarkdownView this page as plain text LLMs fullRead this help center as Markdown for LLMs Open in ClaudeAsk questions about this page Open in ChatGPTAsk questions about this page

Download Markdown Use with AI

Connect Claude, ChatGPT or an agent you build to your help center through the HelpCenter.io MCP server. Claude and ChatGPT install HelpCenter.io from their own directories, so there is nothing to paste. This guide covers each client, what you approve when you sign in, how to work with several help centers, and how to take access back.

Before you start

Plan: the help center must be on the Growth or Catalyst plan, or a free trial of one. On other plans, the AI agents card in Settings says Not included in your current plan. See plans and pricing.
Where to find it: Claude installs the HelpCenter.io connector from its Connectors Directory, and ChatGPT the HelpCenter.io plugin from its plugin directory.
Server URL: both connect to https://mcp.helpcenter.io for you. You need the URL for Claude Code's command, another MCP client or your own agent. In your dashboard it is in Settings → AI agents, under Connector URL, with a Copy button.
Signing in: with Claude and ChatGPT you sign in with your HelpCenter.io account, and there is no key to copy. You can connect the help centers that belong to your account.
Your own agent: needs an API key from the help center it should work on. See Create an API key.

Connect Claude

HelpCenter.io is in Claude's Connectors Directory: see the HelpCenter.io listing. Connect it once and it works in Claude on the web, in Claude Desktop and on mobile, and in Claude Code when you sign in to it with your Claude account.

In Claude, open Customize → Connectors, select Discover and search for HelpCenter.io. Or open its directory page directly.
Select Connect to Claude.
Claude opens the HelpCenter.io sign-in: in the same tab on claude.ai, and in your browser from Claude Desktop. Sign in, check the permissions on the Authorize access screen, tick your help centers if you have more than one, and click Allow.
Back on the Connectors page, HelpCenter.io is listed under Your connectors as Connected.
In a chat, select + → Connectors and check that HelpCenter.io is turned on. Then ask something that needs your help center, for example "Which of our articles explain refunds, and what do they leave out?"

On a Claude Team or Enterprise plan, an Owner adds HelpCenter.io for the organization first: for them the button reads Connect for your team. Members who can't add connectors see Request, which asks the Owners, or a note to ask an Owner. Each member then connects with their own HelpCenter.io account.

Claude can ask for your approval before a HelpCenter.io tool runs, with Allow once or Always allow. Tools that can change or remove something that already exists, or change what readers see, are marked destructive, and those always ask; read-only tools can run without asking. To change what needs approval, open HelpCenter.io under Customize → Connectors and set its Tool permissions.

Add it by its URL instead

You can also add https://mcp.helpcenter.io by its URL, for example to pin a connection to one help center with ?site= (see Work with more than one help center, below). On Free, Pro and Max plans, use Customize → Connectors → Add custom connector; on Team and Enterprise, an Owner adds it in Organization settings → Connectors. A connector you added by URL keeps working. If its address differs from the directory's, for example because it ends in ?site=, adding HelpCenter.io from the directory gives you a second connection: remove the one you don't want under Customize → Connectors.

Screen recording. Under AI agents, copy the Connector URL. In Claude, add it as a custom connector, then sign in and pick this help center. Using Claude Code? Copy the one-line command instead. — The AI agents card in Settings has the connector URL and the Claude Code command.

Connect Claude Code

If you connected HelpCenter.io in Claude and sign in to Claude Code with the same Claude account, Claude Code has it already: /mcp lists it with the connectors that come from claude.ai. Claude Code loads those only with a claude.ai login, not when it runs on an API key or another model provider.

To add it to Claude Code on its own, run this command. The AI agents card shows it too, under Claude Code, with a Copy button.

claude mcp add --transport http helpcenter https://mcp.helpcenter.io

You sign in to HelpCenter.io the same way as in Claude, on the same Authorize access screen. If you have both, Claude Code connects to HelpCenter.io once, using the one you added with the command, and /mcp shows the claude.ai connector as hidden. That only happens when both use the same address: a connector with ?site= on the end is a separate server.

Connect ChatGPT

HelpCenter.io is in ChatGPT's plugin directory as the HelpCenter.io plugin, by HelpCenter Solutions. It brings the HelpCenter.io app, which signs in to your help center through this MCP server, and skills that show ChatGPT how to use it.

In ChatGPT on the web, desktop or mobile, select Plugins in the sidebar and find HelpCenter.io. Or open its directory page directly.
Select Install plugin if it is shown. If ChatGPT prompts you, select Connect for the HelpCenter.io app.
ChatGPT opens the HelpCenter.io sign-in. Sign in, check the permissions, tick your help centers and click Allow.
Ask your question. ChatGPT uses the plugin when your request needs it. To choose it yourself, @-mention HelpCenter.io, select + and enter its name, or select Plugins in the message box.

In a ChatGPT workspace, admins decide which plugins members can install, under Plugins in the Admin console. If HelpCenter.io shows Disabled by admin, ask an admin to make it available. Whether you can install a plugin also depends on your plan, role and region: OpenAI's guide to plugins in ChatGPT has the current rules.

Add it as a custom MCP server instead

Where OpenAI allows it for your plan and role, ChatGPT can also connect to https://mcp.helpcenter.io as a custom MCP server, with OAuth for authentication; it cannot send an API key. In a workspace, an admin decides who can do that. OpenAI's guide to connecting custom MCP servers in ChatGPT has the steps and the current rules, including whether a server added this way may change content.

Connect another MCP client

A client that supports remote MCP servers over Streamable HTTP with OAuth sign-in can use the same URL, provided it follows the MCP authorization flow as HelpCenter.io implements it:

Discovery. A request without a credential gets 401 Unauthorized and a WWW-Authenticate header naming https://mcp.helpcenter.io/.well-known/oauth-protected-resource. That document names the authorization server, https://helpcenter.io, whose metadata is at https://helpcenter.io/.well-known/oauth-authorization-server.
Registration. Dynamic client registration at https://helpcenter.io/oauth/register. Redirect URIs must be https, or http on a loopback address (localhost, 127.0.0.1, [::1]). Any other scheme is refused with invalid_redirect_uri. The authorization server also accepts a Client ID Metadata Document (its metadata says "client_id_metadata_document_supported": true): a client whose client_id is the https address of its metadata document needs no registration call. Claude connects this way.
Authorization code with PKCE, method S256.
The resource parameter, set to https://mcp.helpcenter.io. Without it, the token is not issued for the MCP server, which refuses it with 401.
Scopes. The challenge suggests content.read content.write analytics.read notes.read notes.write design.read design.write embed.manage: every scope a tool uses. It leaves out webhooks.manage and domain.manage, which no tool uses.

The endpoints, token lifetimes and errors are in OAuth 2.1 for apps and AI clients.

Connect your own agent with an API key

An agent you run yourself skips the sign-in and sends an API key with every request. The key decides everything: it works on the help center it was created in, with its scope (Read only or Read & write), and reaches team notes or the design only if it has that access, which an Owner or Admin sets when creating the key or later with Edit access on it.

Create a key in Settings → API keys of the help center the agent should work on. See Create an API key.
Store it on the server where the agent runs, for example as HELPCENTER_API_KEY. Never put it in browser code.
Send it as Authorization: Bearer <key>, with Content-Type: application/json and an Accept header that lists both application/json and text/event-stream. The X-HelpCenter-Api-Key and apikey headers also carry a key.

This request calls one tool:

curl -s https://mcp.helpcenter.io/ \
  -H "Authorization: Bearer $HELPCENTER_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call",
       "params":{"name":"helpcenter_list_sites","arguments":{}}}'

The result has the text the model reads in content, and the API's response in structuredContent:

{
  "result": {
    "content": [
      {
        "type": "text",
        "text": "# Help centers (1)\n\n- Acme (id: 381) | help.example.com\n\nThis connection reaches one help center, so no site_id is needed."
      }
    ],
    "structuredContent": {
      "status": "success",
      "sites": [
        {
          "id": 381,
          "uuid": "3f9a8112-8fb7-414f-829a-bde2136e0249",
          "name": "Acme",
          "subdomain": "acme",
          "domain": "help.example.com",
          "url": "https://help.example.com",
          "default_language": "en",
          "visibility": "public",
          "publicly_accessible": true,
          "languages": ["en"],
          "created_at": "2026-09-30 08:03:14"
        }
      ],
      "meta": { "items_count": 1 }
    }
  },
  "jsonrpc": "2.0",
  "id": 1
}

Things to know when you write your own client:

Each POST stands alone. The server keeps no session, so you can call tools/call without initialize first. MCP client libraries send initialize anyway, which works too.
A missing text/event-stream in Accept gets 406 Not Acceptable, and a body without Content-Type: application/json gets 415 Unsupported Media Type.
A failed tool call still answers 200 OK. Look for "isError": true in the result; its text says what went wrong and what to do.
A wrong key is not refused when you connect: initialize and tools/list answer with any key. The first tool call fails with Error: Authentication failed. The HelpCenter.io API key is missing or invalid.
The plan applies to keys too. A key that works against the REST API gets the plan refusal through the MCP server when its help center is not on Growth or Catalyst.

When an app signs in, HelpCenter.io shows Authorize access: the app's name, then Permissions, the list of permissions the app asks for. You approve them together or click Deny: you cannot drop one permission and keep the rest. What you choose is the help centers.

One help center: the screen names it under Help center.
Several: tick them under Help centers to share. Select all ticks every one you can share. Nothing is ticked the first time, and when you approve an app again, the help centers it already reaches are ticked.
Help centers on other plans are shown locked, with their plan and an Upgrade link. When none of your help centers is on Growth or Catalyst, the screen says Your plan doesn't include the MCP server instead, and nothing is shared.

These are the permissions an app can ask for, with the words the screen uses:

Permission	On the screen	What the agent can do with it
`content.read`	Read your help-center categories and articles.	Search and read articles, categories, reader comments, staged changes and versions, change sets, the images uploaded through the API, the widget's install code and the FAQ sections, and list your help centers.
`content.write`	Create and update your help-center articles.	Create, change and delete articles and categories (a deleted article goes to the Trash), upload and delete images, stage, publish and discard changes, restore versions, create, fill, publish, schedule and delete change sets, approve, unpublish and answer comments, and create, change and delete FAQ sections, and add, remove and reorder the FAQs they list.
`analytics.read`	Read your help-center analytics reports.	Use the three analytics tools.
`notes.read`	Read the private notes your team leaves on drafts. Readers never see these.	Read team notes.
`notes.write`	Reply to and resolve the private notes your team leaves on drafts.	Reply to and resolve team notes.
`design.read`	See your help center's design: its theme, page layouts, components, custom CSS and scripts.	Read the design draft, the guide to what can be changed, the themes and the published versions of the design.
`design.write`	`Change your help center's design — theme, layouts, components, custom CSS and scripts — and publish it.`	Apply a theme, edit, discard and publish the design draft, and load an earlier version into it. That includes custom CSS and scripts, which run in every visitor's browser once the design is published.
`embed.manage`	Choose which websites may embed your help center and run its widget. This changes your help center's security headers.	Change the list of websites allowed to embed the help center and run its widget, with `helpcenter_update_embedding_origins`.
`domain.manage`	Connect your help center to your own domain, check its DNS and certificate, and remove the domain.	Nothing through the MCP server: no tool uses it.
`webhooks.manage`	Register webhooks that notify it when your content changes.	Nothing through the MCP server: no tool uses it.
`offline_access`	Stay connected, so you are not asked to sign in again each time its short-lived access expires.	Nothing more. HelpCenter.io gives every app a refresh token, with or without this permission.

Apps may also ask for openid, profile and email, which let them confirm who you are and read your name and email address.

If you already approved the same permissions for the same app, HelpCenter.io skips the screen and connects at once. The footer of the screen reminds you that you can change or revoke access in Account › Connected Apps.

Work with more than one help center

If you approve one help center, the agent works there. If you approve several, the agent chooses one on each tool call:

Tell the agent which help center you mean, by its name or its address.
The agent lists your help centers with the helpcenter_list_sites tool, which shows each one with its id.
It passes that id as site_id on each call. One conversation can work in several help centers.

When your request does not say which help center you mean, the agent is told to ask you instead of choosing. A call that names no help center is refused with site_required, so nothing is written to a help center the agent did not name.

To keep an agent in one help center, add the connector with the id at the end of the URL, for example https://mcp.helpcenter.io/?site=1234. That connection acts on that help center on every call and refuses any other site_id. API keys ignore ?site=, because a key belongs to one help center. The directory connector in Claude and the ChatGPT plugin always use the plain address, so to pin a connection, add HelpCenter.io by its URL (see Connect Claude, above).

Add a permission later

The 401 challenge that starts every sign-in names every permission a tool uses: content.read, content.write, analytics.read, notes.read, notes.write, design.read, design.write and embed.manage. A client that asks for fewer, or a connection made before a permission existed, gets a token without the rest, and the tools that need them are then refused before they run: the server answers 403 Forbidden with WWW-Authenticate: Bearer error="insufficient_scope", scope="analytics.read" (or the permission the tool needs), and the agent reads:

This tool needs the 'analytics.read' permission and the current authorization does not include it. Re-authorize the HelpCenter.io connector and approve that permission — retrying without it will fail identically.

To fix it, authorize the connector again so that the client asks for the missing permission, then click Allow. In Claude, open HelpCenter.io under Customize → Connectors, select Disconnect, then connect it again. HelpCenter.io adds the permission to what it remembers for the app. If you build the client, include the permission in the scope of the authorization request.

Manage or revoke access

Click Manage connected apps in the AI agents card, or open https://helpcenter.io/app/account/connected-apps.
To change what an app reaches, tick or untick help centers under Help centers it can reach and click Save help centers. The change applies to the app's next request. To leave it none, disconnect it instead. For an AI agent, a help center whose plan doesn't include the MCP server is shown locked, with a View plans link: you can untick it if the app already reaches it, but not add it. Apps that only use the API can be given any of your help centers.
To remove an app, click Disconnect and confirm. Its tokens stop working at once, and the agent is told to reconnect.

Disconnecting HelpCenter.io inside Claude or ChatGPT signs that app out on its side, but does not by itself end the authorization at HelpCenter.io: to end it, disconnect the app here as well.

Each app shows the permissions you approved, which you cannot change on this page. The account owner sees and manages every member's connections, and each member sees their own. For an agent that uses an API key, delete the key in Settings → API keys.

Troubleshooting

The client reports 401 Unauthorized with "Missing HelpCenter.io credentials". That is the challenge that starts the sign-in. A client that supports OAuth for remote MCP servers continues to the HelpCenter.io sign-in. An agent of your own must send an API key.

"Error: Authentication failed. The HelpCenter.io API key is missing or invalid." The key is wrong or was deleted. The message's advice to set HELPCENTER_IO_API_KEY does not apply to https://mcp.helpcenter.io: check the key you send in the Authorization header.

"Error: HelpCenter.io no longer accepts this OAuth authorization". The app was disconnected or its token expired. Reconnect it to sign in again.

"The HelpCenter.io MCP server is available on the Growth and Catalyst plans." The help center's plan does not include the MCP server, for example after a plan change. Reconnecting does not help. It works again on the next request once the help center is on Growth or Catalyst.

The sign-in says "Your plan doesn't include the MCP server". None of your help centers is on Growth or Catalyst. Move one to either plan, then connect again.

A help center you work on is not listed on the consent screen. It belongs to another HelpCenter.io account. Ask its owner to connect it, or use an API key created in that help center.

"Error: This connection reaches more than one help center and this call did not say which". You approved several help centers and the agent called a tool without site_id. Tell the agent which help center you mean. The message lists the ids, and the agent repeats the call with one of them.

"Error: This authorization does not cover help center 1234." The site_id is not one of the help centers you approved. Ask the agent to list your help centers, or approve the app again with that help center ticked.

"Error: This connection is pinned to help center 1234". The connector URL carries ?site=1234 and the agent named another help center. Nothing was sent. Remove ?site= from the URL to let the agent choose on each call, or add a second connector for the other help center.

"Error: This authorization does not cover the help center the connection URL selects with '?site='." The id in the connector URL is not one of the help centers you approved. Fix or remove it, or approve the app again with that help center ticked.

"Error: This connection uses an API key, and a key is issued for exactly one help center". The agent passed a site_id that is not the key's help center. Nothing was sent. Leave site_id out, or use a key created in the other help center.

"This tool needs the 'analytics.read' permission" (or notes.read, notes.write, content.write, design.read, design.write, embed.manage). See Add a permission later, above.

"Error: Permission denied. This action requires the notes.read scope." With an API key, the key has no team notes access. An Owner or Admin can turn on Allow access to editorial notes with Edit access on the key in Settings → API keys. The advice in the message about a write-scoped key does not apply to notes.

"Editorial notes are visible only to team members who can access team comments in the dashboard." The person behind the connection no longer has access to team notes on that help center. Someone with that access has to connect instead.

"Error: This API key was not issued with access to the help center's design." The key has no design access, or it is a Read only key and the tool changes the design. An Owner or Admin can turn on Allow access to the help center design with Edit access on the key in Settings → API keys. A key's scope can't change, so for design changes the agent needs a Read & write key.

"Error: The person this connection signs in as can no longer edit this help center's design." The person who connected the agent, or who created the key, is not an Owner, Admin or Editor of that help center. Someone who can edit the design has to connect instead, or create the key.

A design tool answers that the help center has to switch to the template editor first. The design tools work only with help centers designed in the template editor. The answer carries the link to the template editor, where the owner reviews the current design as a draft and publishes it to switch. See Design API.

"Error: Permission denied. This action requires the content.write scope." The key is Read only. Use a Read & write key.

"Send at most 50 articles per batch; split a larger import into several batches". The agent put more than 50 articles in one bulk import. Nothing was sent. It can send the same articles again in batches of 50 or fewer.

"Staged changes are not included in this help center's plan." Staging an edit needs the Catalyst plan. See Edit live articles safely with staged changes.

New tools do not show up. The server answers each request on its own and cannot notify a client that its tools changed, so the client sees new tools only when it lists them again. The same goes for an argument a tool gained, such as site_id: a client that still holds the earlier list may send it as text and get Expected number, received string. Reconnect the connector, or start a new session in your client.

Your client gets 406 or 415. Send Accept: application/json, text/event-stream and Content-Type: application/json.