Run WeSpeak
by talking to your AI
WeSpeak MCP adds your account's tools to Claude or ChatGPT web: check metrics, search conversations or edit the assistant's knowledge without opening the panel — with your role's permissions.
Your WeSpeak account, one question away.
WeSpeak MCP is WeSpeak's official connector for Claude and ChatGPT web. MCP is the standard protocol AIs use to connect to external tools: your AI gets the same tools you already use in the panel, and runs them as your user.
Connect once and it's ready.
Enable the feature
Your advisor turns it on for the account and it appears in the panel's Settings.
Connect your account once
In Settings, follow the steps in the Claude or ChatGPT tab and authorize access with your WeSpeak user. The Claude tab also includes a video.
Your AI gains the tools
Your AI now has your WeSpeak account's tools, with your permissions and nothing more.
Ask in plain language
"Find yesterday's conversations that went unanswered" — and the answer comes back with your account's data.
Check, organise and adjust, without opening the panel.
Check and diagnose
Search conversations and read messages, look up a booking, search the help guides and report a poor assistant reply.
Manage the team's work
Plus: assign, close, reopen or snooze conversations, move leads between funnel stages and check metrics.
Adjust the assistant
Plus: view and edit the assistant's knowledge (validated and reversible), send test messages, and pause or reactivate it.
Combine with your other tools
Since it lives alongside your AI's other connectors, you can ask for things the panel can't do: for example, cross your metrics with your spreadsheets.
Multi-property
Each account has its own connection. They all coexist in your AI app and each one sees only its own data.
AI usage on your own account
Queries count toward the usage available in your AI provider’s plan, not your WeSpeak plan. Your advisor confirms how activation works in your case.
What changes when you ask instead of searching.
Answers in seconds
The figure that used to mean opening the panel and filtering through three screens now takes one question.
Less repetitive work
Review pending items, move leads or adjust knowledge without repeating the same clicks every day.
Control without risk
Every action respects your role and is logged: nobody does more than they already could.
No technical setup
It connects with your usual WeSpeak user: no new keys, no API tokens and no development involved.
Your usual permissions. No more, no less.
The connection doesn't create a parallel access: it runs tools as your user and respects exactly the role you have in the panel.
WeSpeak can't see your AI chats
Your conversation with the AI is governed by its provider’s privacy settings and policies. WeSpeak only receives calls to its own tools, logged under your name.
Revocable in one click
Disconnect it from your AI's settings whenever you want. Each person connects their own user — no shared access.
It never messages guests
It exists to run your account, not as a service channel: the connector never writes to your guests.
The connection works with Claude and ChatGPT web. Tools that use AI internally may take a few minutes per run.
For the people who live outside the panel.
Owners and management
Ask for the month's status or the detail behind a metric without opening the platform or waiting for a report.
Operations supervisors
Review unanswered conversations, reorder the funnel and spot weak assistant replies on the spot.
Teams already working with AI
If your operation already lives in Claude or ChatGPT web, WeSpeak joins as one more tool and combines with what you have connected.
What hotels usually ask us about WeSpeak MCP.
No. It connects with your usual WeSpeak user: you sign in once from your AI and that's it.
No. The connector is for running your account — checking, organising, adjusting — but it never sends messages to guests. It is not a service channel.
Most tools only read. What can be modified: pausing or reactivating the assistant, editing its knowledge (validated, backed up and reversible), changing a conversation's status and moving a lead between stages.
Yes, it works with Claude and ChatGPT web. Settings includes the steps for each application. For other AI applications, check compatibility with your advisor.
No, they coexist: the same tools remain in the panel and, additionally, in your AI. Use each wherever it suits you best — and what you do from your AI counts the same and is logged under your name.
Yes. Each person connects their own AI app with their own WeSpeak user and sees only their role's tools. If you manage several properties, add one connection per account and they all coexist in the same AI app.
Remove the WeSpeak connector from your AI app's settings and your user's access is cut instantly. To disable the feature for the whole account, ask your advisor.
The MCP connection is enabled by your WeSpeak advisor. Queries count toward the usage available in your AI provider’s plan, not your WeSpeak plan. Prices for all modules are in the Extras section of the platform: app.wespeak.pro/extras.
No. WeSpeak only sees its own tools being run, logged under your name. The rest of your conversation is governed by your AI provider’s privacy settings and policies.
Your platform answers
when you ask it.
No new keys, with your role's permissions and revocable whenever you want — your WeSpeak account, inside the AI you already use.