Where to set it up
- Settings → Developers, in app.intake.es
- Generate the MCP server credential
- Choose what the connected client can read and do
Connect your models to Intake conversations over MCP
Model Context Protocol is the standard a model uses to talk to external tools. Intake runs its own MCP server: any compatible client can read conversations, search the knowledge base and trigger actions with the credentials and permissions you define.
What it solves
Your conversations, available to your own agents
What support learns (what people ask, how it was solved, what breaks most) rarely leaves the support tool. With the MCP server, your models and agents can work with it like any other source, without exporting anything or building a copy.
What it solves
A standard protocol instead of another API
Every custom integration is a contract someone has to maintain whenever it changes. MCP is an open protocol your tools already understand, so connecting Intake is close to plug and play. And what you learn carries over to the next system you connect.
What it solves
You decide how far it reaches
Opening support conversations to a model is a decision you need to be able to limit. The server offers what you enable and nothing else, and you can shut it off from your side without depending on anyone. You can start with public knowledge and expand later.
What it does
What the connection gives the agent and what it does with it
What counts in an integration is what it lets the agent do. This is the data MCP brings and what the agent does with it on its own.
What it offers
- Read conversations and their history
- Search the knowledge base
- Trigger actions and hand the conversation to a person
How MCP helps the AI agent
Three concrete things that change your team’s day.
Your agent, with what Intake already knows
If you already have your own agent, you don’t need to migrate it: it connects over MCP and works with the conversations and knowledge you already have in Intake.
Clear permissions
The credential defines what the connected client can read and run. You set the scope when you create it.
An open standard
Because it’s MCP, any compatible client works. You don’t have to write an adapter for each tool you want to connect.
How it works
Three steps, and only the first needs a MCP admin. You can change the rest later without reconnecting anything.
Authorize the connection
Open the integrations panel, pick MCP and authorize access with the account your team already uses. You need admin rights in MCP, so if you don’t have them, this is the only step you’ll have to ask someone for.
Choose what the agent sees
Select the channels, events and account fields it can access. Anything you don’t select stays hidden from it, and you can change it anytime without reconnecting.
The agent answers with that data
From the first conversation, the agent checks that data before replying, resolves what it can close on its own and hands the rest to a person, with the case already prepared.
Frequently asked questions
How long does it take to connect MCP?
A few minutes. You authorize access and choose what it can see; nothing needs rewriting and no development work is needed to start.
Can I revoke access later?
Yes, from either side: by revoking the connection in Intake or removing the permission in MCP. The agent stops reading that data right away and keeps answering with everything else.
Does using MCP cost extra?
No. Every integration is included in the plan, from €49 a month. There are no add-ons to buy and no per-call charges.
What happens if MCP stops responding?
The agent answers with what it does have and, if the question depended on that data, hands the conversation to a person and explains why. It never makes up data it couldn’t look up.
What is MCP?
Model Context Protocol is an open standard that lets a language model use external tools. Instead of writing one integration per model, you offer an MCP server and any compatible client can use it.
Does Intake offer a server or connect to third-party ones?
It offers its own: your agent or MCP client connects to Intake. The credential and its scope are generated in Settings → Developers.
Can I limit what the connected client sees?
Yes. When you create the credential you choose what it can read and which actions it can trigger. Anything you don’t enable stays out of reach.
