Inbox Monster publishes a connector that Claude can use to read your account. Instead of opening a report, exporting it and pasting it into a chat, you ask a question for Claude to answer.
The connector uses the Model Context Protocol (MCP), an open standard for connecting assistants to outside systems. Everything Claude does through the MCP is done on the specific team you selected, and only within the access you approve.
What the connector can reach
The connector covers three areas of the platform.
Area | What it covers |
Deliverability | Inbox placement, blocklists, spamtraps, reputation, DMARC and Monster Score |
Creative Rendering | Renderings and engagement |
Administration | Account administration, such as managing domains, IPs and users |
Everything is scoped to a single team. If your account covers several brands, you choose one when you connect, and Claude sees that team only.
Connecting Claude
In Claude, open Settings > Connectors, then find Inbox Monster MCP.
The connector, as it appears in Claude, with the list of tools it provides.
Click Connect.
Before you connect, the connector shows a single Connect button.
Inbox Monster opens its own page and asks which team to use. Log in first if you are not already signed in.
Choose the team Claude should work with, then click Allow.
Choose your team and click Allow. Claude confirms the connection, and the connector is ready to use.
You are never asked to paste an API key. Approving creates a token for the team you picked, and nothing else.
ℹ The access level comes from the assistant Read and write access is part of the request Claude makes, not something you pick on the consent screen. The team is your choice; the access level is the connector’s. |
Controlling what Claude may do
Once connected, Claude lists every tool the connector provides and lets you decide when each one may run. Open the connector in Settings > Connectors to see them.
Every tool has its own control, so you can let some run freely and hold others for approval.
Tools are grouped by what they do:
Read-only tools fetch information and change nothing: campaign results, rendering previews, engagement figures, blocklist and spamtrap hits.
Interactive tools act on your account, such as creating a rendering or running a Custom QA check.
Set the interactive ones to ask before running if you would rather approve each action. You will only ever see the tools your account already gives you access to.
✦ Claude inherits your access, not more The connection cannot reach anything the team you picked cannot reach, and it cannot do anything your own permissions would not allow. It is a faster way to ask the same questions, not a way around your permissions. |
Disconnecting
Open the connector in Settings > Connectors and click Disconnect. Claude stops reaching your account straight away. Earlier chats that used the connector stay readable, but Claude cannot fetch anything new from them.
To connect a different team, disconnect and reconnect, then choose the other team.
Frequently asked questions
Does Claude see my whole account?
No. It sees the one team you chose when you connected, and only what that team can see.
Can I connect more than one team at a time?
No. The connection is for a single team. Reconnect to switch to a different one.
Do I need an API key?
No. You log in, choose a team and approve. A token is created for you.
Why does Claude say the connector is not verified?
Inbox Monster publishes it as a community connector, so Claude shows a notice explaining that it has not been verified by Anthropic. The connector is built and maintained by Inbox Monster.
Is this the same as the AI features inside Inbox Monster?
No. AI analysis, AI summaries and Custom QA run inside the Inbox Monster platform. The connector lets you access your Inbox Monster data from Claude.
Claude has stopped returning answers. What happened?
The connection was most likely disconnected or replaced. Reconnect and approve access again.