Ask your AI assistant about Teams presence (MCP)

You can now ask an AI assistant who on your team is online, when, and for how long. Presify is a Microsoft Teams presence tracker, and it is an MCP server. It supports the Model Context Protocol (MCP), the open standard that lets AI assistants like Claude and ChatGPT use tools from apps you choose to connect. Microsoft Copilot can use MCP servers too, through Copilot Studio.

Connect Presify once, and questions like these get answered live from the presence history in your workspace:

• How many hours a day is my team actually online?
• How has online time changed since last month?
• Show me Maya's timeline for yesterday.
• Does anyone's presence look gamed this week?

Instead of guessing, the assistant calls Presify's read-only tools (presence summaries, timelines, coverage, anomalies, connection health), gets real numbers back, and explains them in plain language.

What is MCP?

MCP (Model Context Protocol) is a standard way for AI assistants to use external tools. Instead of pasting screenshots or exporting CSVs into a chat, you connect an app once. The assistant can then request exactly the data it needs, when you ask for it. The assistant sees what you could already see in the app, nothing more.

What the connector can and cannot do

The Presify connector is deliberately narrow:

• Read-only. Every tool reads; none of them change anything. Your assistant cannot edit settings, add people, or delete data.
• Role-scoped. The connection acts as you. An owner sees the workspace; a manager with the Monitor role sees only their assigned people. The assistant inherits exactly those limits.
• Revocable. Every connection is listed in Settings, and revoking one takes effect immediately.
• Available on every plan, including trials. The presence flags it can report depend on your plan, the same as in the app: Standard includes the Idle suppression flag, and all six run on Pro and Scale.

One thing worth saying plainly: when you ask an assistant a question, the answer (your team's presence data) flows to that assistant's provider under their terms. Sending it is your call to make, per conversation, the same as pasting a report into a chat would be. Presify never sends data to an AI provider on its own.

Why is an assistant a better way to read presence data?

An assistant answers the question you actually have, not only the questions a dashboard predicted. You ask in the words you would use with a colleague, and it can follow up: from a team average, down to one person's day, to the flag behind it.

Following up matters because the green dot shows one moment, not the whole story. The person who is always green may have a day of unbroken Available with no idle time. The person who always looks yellow may be steadily online in the evening. Asking follow-ups is how you get from the glance to the pattern.

Presence data stays what it should be: directional, and a starting point for a conversation.

How do you connect an AI assistant?

You add Presify as a connector in Claude, ChatGPT, or Copilot, and it takes about a minute. Point the connector at your Presify workspace, sign in with your normal Microsoft account, review what the connection can see, and approve it. There are no API keys to create for the standard flow. The in-app guide covers command-line setups that use a personal access token instead.

If your team runs on Microsoft Teams, this is a practical place to start with an AI assistant at work: real questions, real numbers, no exports. Presify itself takes one read-only Microsoft sign-in and about five minutes to set up, with nothing to install on anyone's laptop. See how Presify works or start with one Microsoft sign-in.