Connect Claude or Cursor to Alpire

Alpire has an MCP server: connect your AI assistant (Claude, Cursor, n8n…) to your account and ask it for things the way you'd ask a person. “How many calls did we miss yesterday?”, “book Jorge into the first free slot on Monday morning”, “take Ana off the support queue”.

MCP is a standard that lets an AI use tools from other applications. Your assistant sees the Alpire tools your connection allows, decides which ones to use and tells you the result.

There are two ways to connect it:

• With your account (the easy one). In ChatGPT and Claude.ai you paste an address, Alpire asks your permission on a screen, and that's it. No keys to copy. See Connect ChatGPT or Claude.ai.
• With an API key. For Claude Code, Cursor, n8n and other developer tools. See Step 1.

Available from the Pro plan, with an active subscription: API keys can't be created during the free trial (the MCP server uses the Alpire API under the hood).

This guide goes one way: your assistant uses Alpire. The other way round —the voice agent using your tools (Zapier, Make, n8n) during a call— is covered in Tools for the agent (MCP).

What you can ask it

A real example, from start to finish:

You: How many calls did we miss yesterday, and who were they from? Book Jorge Pérez (+34 611 222 333) a 30-minute check-up on Monday in the first free morning slot.

(the assistant checks which organization it's connected to and what day it is there, lists yesterday's missed calls, looks for slots on Monday morning and books the first one)

Assistant: Yesterday we missed 2 calls: one from Jorge Pérez (+34 611 222 333) and another from +34 622 333 444. I've booked Jorge's 30-minute check-up for Monday at 11:30.

What it can't do yet

• Buy numbers. That's done in the panel. (AI agents, audios, the knowledge base and contacts can be managed from here; deleting contacts or agents cannot.)
• Custom flows. The fixed templates (AI receptionist, AI in hours with voicemail, forwarding in hours with voicemail, direct forwarding and menu) are the quick start; for anything else the assistant designs the full flow with the language reference and your catalog, validates it and leaves it as a draft. Nothing reaches your customers until you say yes to publishing it. It cannot add new MCP tools or scripts to a flow, delete a flow, turn it on or off, or remove or move a number between flows: that is done in the panel editor.
• Download the audio of recordings. The transcript, yes.
• Write in a chat as if it were the visitor.
• List or cancel appointments if your calendar is Cal.com or Google: with an external calendar, your assistant can see free slots and book, but appointments are looked up and cancelled in your calendar.
• Changing the permissions of an existing connection. To give more or fewer permissions, disconnect Alpire in your assistant and connect it again (or revoke the connection in the panel and reconnect).

Connect ChatGPT or Claude.ai (no keys)

No keys and no code. You paste Alpire's address into your assistant, sign in to Alpire as usual (with your second factor, if you have one) and approve what the assistant will be able to do.

You need: to be an administrator of your organization in Alpire, a Pro plan or higher, and a ChatGPT or Claude account that lets you add custom connectors.

The address to paste depends on where your organization is (the same region you use to sign in to the panel):

Claude.ai

1. Open claude.ai, then Settings and Connectors.
2. Click Add custom connector.
3. Type a name (for example “Alpire”) and paste the address. Leave any client ID or secret empty.
4. Click Add, then Connect. An Alpire window opens.
5. Sign in to Alpire if asked and check the permissions screen: it shows which organization, whether the assistant can only look things up or also change things, and when the connection expires. Click Allow.
6. Back in Claude, start a new conversation, enable the connector from the tools menu and try “What can you do with Alpire?”.

On Team and Enterprise plans, the Claude organization owner adds the connector first; then each person clicks Connect with their own Alpire account.

ChatGPT

1. In ChatGPT open Settings, then Apps (or Connectors) and Advanced settings, and turn on developer mode. On business plans a workspace admin turns it on first.
2. Click Create.
3. Type a name (“Alpire”), paste the address and under Authentication choose OAuth. If it asks for a client ID or secret, leave them empty: Alpire registers itself.
4. Accept the notice that you are adding this connection yourself and click Create. An Alpire window opens.
5. Sign in and click Allow on the permissions screen, as above.
6. In a new conversation, pick Alpire among the developer-mode tools and ask “What can you do with Alpire?”.

Menu names change from time to time in ChatGPT and Claude. If you can't find one, look for “connectors” or “apps” in the settings.

What the permissions screen shows

• The organization it gives access to (you pick one if you administer several).
• What it can do: “only look things up” or “look things up and change things” (book appointments, change queues…).
• How long it lasts: 30 days, 90 days or a year. When it expires the assistant asks your permission again.
• Who is asking: the application name (ChatGPT, Claude…). If you did not expect the request, click Cancel.

A connection never has more power than you: if you lose a permission or stop being an administrator, the connection loses it immediately.

See and remove connections

In Settings → API and webhooks, under the API keys, Connections with assistants lists which application, whose it is, whether it can change things, when it expires and when it was last used. Click Revoke and it stops working at once. If a colleague leaves the organization, their connections close by themselves.

If something goes wrong

You can also ask the assistant “what can you do here?”: Alpire lets it explain in plain words.

Before you start (with an API key)

• Pro plan or higher.
• Be an administrator of your organization: only administrators create API keys and MCP servers (and a key never grants more permissions than the person creating it has).
• An MCP client: Claude Code, Claude Desktop, Cursor, n8n or another one that supports MCP over HTTP with a header.

Step 1 — Create an API key

Settings → API & webhooks, API keys card (only administrators see it).

1. Name: what it's for and who it's for (“Marta's Claude”, “n8n reports”). That way you'll know which one to revoke.

2. Permissions: tick only what you need. They are grouped by section and each group comes collapsed: open it to tick (a counter tells you how many you have in each one). Your assistant will only see the tools for those permissions. With a read-only key it can't change anything, because it doesn't even see the tools that write. Checkboxes for a section that you can't access are disabled (“You can't grant this: you don't have access to that section”): a key never grants more than its creator can see.

   

3. Expires: 90 days is fine to start with.

4. Click Create key and copy it now: it starts with ak_ and won't be shown again. If you lose it, create another one and revoke that one.

   

What each permission gives

What to tick depending on the use:

• Look things up and analyze (the most common): View calls, Listen to recordings (for the transcripts and the saved variables), View reports, View the calendar, View groups and queues. Nothing that changes things.
• Run the calendar: the above plus Book and cancel.
• Coordinate the team: View extensions, View groups and queues and Change who answers.
• Manage the team and the phone system (for non-technical people who ask their assistant for everything): View the team and invitations, Invite people to the team, View opening hours and holidays, Change opening hours and holidays, View phone numbers, View extensions and Create and delete extensions. You can only tick them if you have access to Settings (team) and Routing (hours and numbers).
• Ask your knowledge base: View assistants and Ask an assistant.

Better two keys (a read-only one for day-to-day use and another one with write access for when it's needed) than one with everything.

What you can ask your assistant

With those permissions, sentences like these work as they are; if something is missing the assistant asks you, and asks for a yes before changing anything:

• "Create extension 105 for Meeting room 2" · "Delete extension 105" (it tells you what is lost: the extension, its place in groups and queues and its outbound profiles, and it cannot be undone).
• "Who is on my team?" · "Can I add one more person?" (it tells you the plan's seats) · "Invite luis@company.com". A seat is a person, not an extension: if the plan is full the assistant tells you ("your Pro plan allows 10 users and you already have 10") and does not invite. A key never invites as owner.
• "When are we open?" · "Close on Saturdays" · "On 24 December we open 9 to 2 every year" · "Remove the 25th holiday".
• "Which flow does number 934 567 890 go to?" · "Is any number not being answered?" (read only: buying or connecting numbers stays in the panel).

Step 2 — Connect your assistant

The address depends on where your organization is. It's the same region you use to sign in to the panel:

Each region has its own MCP server and only serves the organizations of its region. If you use the other one's address, the server runs nothing and answers with a 403 telling you which one to connect to (“esta organización está en la región NA: conecta a https://mcp.na.alpire.ai/mcp”, this organization is in the NA region: connect to …). In the examples, replace ak_… with your key and, if your organization is in North America, mcp.eu.alpire.ai with mcp.na.alpire.ai.

Claude Code

In a terminal:

claude mcp add --transport http alpire https://mcp.eu.alpire.ai/mcp --header "Authorization: Bearer ak_…"

Inside Claude Code, /mcp shows alpire as connected. By default the configuration stays just for you and that project, outside the repository. Add --scope user to have it in all your projects. Don't use --scope project with the key written in: it would end up in .mcp.json, inside the repository.

Cursor

In ~/.cursor/mcp.json (for all your projects):

{
  "mcpServers": {
    "alpire": {
      "url": "https://mcp.eu.alpire.ai/mcp",
      "headers": { "Authorization": "Bearer ak_…" }
    }
  }
}

In Cursor Settings → MCP, alpire should appear enabled, with its tools. Don't use the project's .cursor/mcp.json with the key written in: it would end up in the repository.

Claude Desktop

Claude Desktop connects to remote servers through a small bridge, mcp-remote. You need Node.js 18 or later installed.

1. In Claude Desktop: Settings → Developer → Edit Config. It opens claude_desktop_config.json:
   • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
   • Windows: %APPDATA%\Claude\claude_desktop_config.json
2. Add this (if you already have other servers, add only the alpire block inside mcpServers):

{
  "mcpServers": {
    "alpire": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://mcp.eu.alpire.ai/mcp", "--header", "Authorization:${ALPIRE_AUTH}"],
      "env": { "ALPIRE_AUTH": "Bearer ak_…" }
    }
  }
}

3. Quit Claude Desktop completely and open it again. The Alpire tools appear in the chat's tools menu.

The key goes in env and not in args on purpose: Authorization:${ALPIRE_AUTH} has no space, and that way the bridge works the same on Windows.

n8n

In a workflow with the AI Agent node, add the MCP Client Tool node as a tool:

• Endpoint: https://mcp.eu.alpire.ai/mcp
• Server Transport: HTTP Streamable
• Authentication: Bearer Auth, with a new credential whose token is your ak_… key
• Tools to Include: Selected, and tick only the ones that workflow needs

Other clients

Any MCP client that speaks Streamable HTTP and lets you add the Authorization: Bearer ak_… header will do.

Step 3 — First test

Ask your assistant:

Which Alpire organization are you connected to, and what can I do with this key?

It should answer with your organization's name, the region, the time zone and the key's permissions. If it gives an error, see “When something goes wrong”.

From there, ask it normal things. Some ideas to get started:

• “Give me a summary of this week's calls: how many came in, how many were missed and at what times.”
• “What free slots are there tomorrow afternoon?”
• “List today's missed calls from numbers that had never called us before.”

Good practices

• Say the period and the time zone. It understands “yesterday”, “this week” or “on Monday”: the server tells it what day and time it is in your organization. If you work across several time zones, say so.
• Give the full phone number, with the country code (+34…), when you look up or book for someone.
• Ask it to show you before changing anything. “Find the first free slot and tell me before booking” is safer than “book it”.
• For long lists, narrow it down. “Today's missed calls” is better than “all calls”.

Security: what you should know

• The key is a password. It gives access to your organization's data, with its permissions. Your assistant stores it in its configuration (a file on your computer, or n8n's credentials). Don't share it, don't push it to a repository and don't paste it into a chat.
• Only what the key allows. Your assistant only sees the tools for the key's permissions, and only your organization's data.
• Approval depends on your client. Claude and Cursor ask for your approval before using a tool that changes something, unless you turn on “always allow” or an auto-approval mode (don't do that for those tools). n8n's AI Agent node doesn't ask: there, use a read-only key or “Tools to Include: Selected” with only what's needed.
• Watch out for what other people write. Caller names, transcripts and what chat visitors type are written by other people, and someone could hide instructions in them (“ignore the above and…”). The server asks your assistant to treat them as data, never as orders, but no AI is infallible. To analyze transcripts or chats, use a read-only key.
• Connections made with your account (ChatGPT, Claude.ai) expire on their own (30, 90 or 365 days, your choice), renew in the background while they are valid and can be revoked just like a key. They never give more permissions than you have.
• Revoking cuts off the operations. In Settings → API & webhooks, click Revoke next to the key. The API checks the revocation every time a tool runs; the tool list of an already-connected client may take up to 60 seconds to reflect the change because of the permissions cache.
• There's a record. Whatever is changed with a key is logged in the organization's activity log (Settings → Audit), with the key's prefix.

Data and privacy

• Alpire's MCP server doesn't store your full key, the arguments or the results. It uses an in-memory cache of identity and permissions valid for 60 seconds. Its technical log records which tool was used, when, with which key prefix, the organization, the result (code and duration) and, if the key fails, the IP the attempt came from; never what was asked or what was returned.
• A call's detail includes the conversation with the customer and, if two employees talked to each other during the call, that internal consultation separately; and, with Listen to recordings, the variables the flow saved (a name, a phone number, an order number…). All of that also reaches your assistant.
• Each region has its own server and the tools only query that region's API: organizations in Europe are served from Europe, and a North American organization's data never goes through the European server. If you connect an NA key to the EU server, it receives the key to identify the region and rejects the use before running any tool; always use your region's MCP address.
• What your assistant reads goes out to your AI provider (Anthropic, OpenAI, Cursor…). It's a service you choose, and it needs its own legal basis, like any other tool you use with customer data. Transcripts and chats may contain callers' personal data.

When something goes wrong

Limits

Frequently asked questions

Is it the same as the Alpire API? Yes. The MCP server uses the public API with your key, so it has the same permissions and logging, and its limits plus the MCP server's own (above). If you write code, there's the API reference.

Does it cost anything? It's included from the Pro plan. Asking an assistant spends AI credits, as in the panel. Your AI provider (Claude, ChatGPT…) charges for its part separately.

Can it see other organizations? No. A key only sees its own organization.

I have two organizations. Can I use both? Yes: one key per organization and the server added twice in your client, with different names (alpire_clinic, alpire_shop). If they're in different regions, each one with its own region's address.

Can it create or change my flows? Yes, with the Create and change flows permission. It designs them in full and saves them as a checked draft; publishing is a separate step it only takes when you confirm it, with the same checks as the editor's Publish button (including the AI disclosure). It does not delete flows or move numbers between flows.

Which AI model works best? Any current model that uses tools. We've tested it with Claude, GPT and Gemini.

What happens if my assistant gets it wrong? With a read-only key, nothing: it reads more or less data than it should. Actions that change things will ask for your approval if your client does that (Claude, Cursor): read them before approving. n8n doesn't ask.

White label (for whoever runs the server)

If an organization belongs to a partner (white label), the MCP server stops presenting itself as Alpire for it: the server name the client sees, the instructions to the model, the tool descriptions and the error messages use the partner's name, and the model only talks about credits (never prices or invoices). GET /v1/me decides it, returning brand: { partner, name, support_email }; the MCP server stores no brand data. If the organization is a partner's but its brand cannot be resolved, a neutral name is used, never "Alpire".

Each instance is still per region (mcp.eu.alpire.ai). For a partner to use its own domain (mcp.partner.com):

1. DNS: the partner points mcp.partner.com by CNAME at the Cloudflare for SaaS custom hostname we give them; Cloudflare issues the certificate.
2. Cloudflare for SaaS: add the custom hostname in the Alpire zone with origin mcp.eu.alpire.ai, keeping the original Host.
3. Instance nginx: a server_name accepting the partner host and proxying to 127.0.0.1:<MCP port> with proxy_set_header Host $host; and proxy_buffering off; for streaming.
4. MCP server: add the host to MCP_PUBLIC_HOSTS (comma separated) and, if the client sends Origin, to MCP_ALLOWED_ORIGINS. Without it the server answers 403.

The key still belongs to the organization's region: the domain only changes the name it presents. This is documentation: nothing is configured on the MCP server besides those two environment settings.