> ## Documentation Index > Fetch the complete documentation index at: https://developers.qomon.com/llms.txt > Use this file to discover all available pages before exploring further. # MCP Server > Connect Claude or ChatGPT to your Qomon space and query live data in real time. ## Overview The Qomon MCP server lets you connect your Qomon space directly to AI assistants like Claude and ChatGPT. Unlike documentation files, MCP gives your assistant **live access** to your data — contacts, actions, users, fundraising — without copy-pasting anything. ## Choose this when * You want an assistant to read live data from a specific Qomon space. * You need answers based on current records, not just public docs. * You are already past the point where documentation context alone is enough. The MCP server authenticates via an API key tied to a specific Qomon space. Each space requires its own key and its own connector. *** ## Create a Qomon account Before connecting Qomon to your AI assistant, you need a Qomon account and access to a workspace. If you don't have one yet, here's how to get started in 2 minutes. 1. Go to [**qomon.com**](http://qomon.com) 2. Click **Login** — then **Create a new account** if you don't have one yet 3. Once logged in, select or create the **workspace** you want to connect *** ## Prerequisites Before connecting, generate an API key for the Qomon space you want to query: 1. Go to your Qomon space → **Settings → Integrations & API** 2. Under **Generate API Keys**, click **+ Add** 3. Name your key (e.g. `mcp-claude`) and copy it — you'll need it during setup *** ## Connect to Claude Available on the **Team plan**. Once added at the organization level, all team members have access automatically — no individual setup required. Go to [claude.ai](https://claude.ai) → **Settings** → **Connectors** For Jason (5) For Jason (15) In the Name field, enter `qomon-mcp`. In the "Remote MCP server URL" field, paste `https://mcp.qomon.ai/mcp` and click **Add**. For Jason (17) Scroll to find **qomon-mcp** and click **Connect** For Jason (16) An OAuth popup opens — paste your **Qomon API key**, select your region, and click **Connect** For Jason (10) Open a new chat — You're all set! Try asking Claude to search your contacts, list actions, or browse fundraising data For Jason (14) Browser cookie issue. Retry in a **private/incognito window**. Always open a **new conversation** — existing chats won't pick up new connectors. Each space needs its own connector and API key. Contact your Claude Team admin to add a new one. *** ## Connect to ChatGPT In **beta**. Requires a **Plus or Pro** subscription with Developer Mode enabled. Go to [chatgpt.com](https://chatgpt.com) (Web version) → **Settings** → **Apps** In **Settings → Connectors**, click **Create** Go to [**chatgpt.com**](http://chatgpt.com) (web version) → **Settings** → **Apps** → **Advanced settings** → toggle **Developer mode** on, then click **Create app** For Jason (19) ```text theme={"system"} https://mcp.qomon.ai/mcp ``` A "New App" form opens. Fill in: **Name** → `Qomon`, **Connection** → paste `https://mcp.qomon.ai/mcp`, **Authentication** → leave OAuth. Check "I understand and want to continue" and click **Create** For Jason (20) A Qomon OAuth popup opens — select your **region**, paste your **Qomon API key**, and click **Connect** For Jason (10) 1 New chat → click **+** → **More** → **Developer Mode** → **Add sources** → enable Qomon Your Qomon app now appears under **Enabled apps** in Settings → Apps For Jason (21) Open a new chat. You're all set! Try asking ChatGPT to search your contacts, list actions, or browse fundraising data For Jason (22) If you only need public documentation context, use [`/llms.txt`](/llms.txt) or [`/llms-full.txt`](/llms-full.txt) instead. *** ## Connect to your own model The Qomon MCP server follows the open **Model Context Protocol (MCP)** standard — which means it works with any MCP-compatible AI assistant or environment, not just Claude or ChatGPT. Cursor, Windsurf, Continue, or any custom LLM setup can connect to it the same way. Use the following MCP server URL: ```text theme={"system"} https://mcp.qomon.ai/mcp ``` If you want to give your model context about the Qomon API (endpoints, structure, guides) rather than live workspace data, check out the [LLM-Ready Documentation](https://developers.qomon.com/pages/v1/build-with-llms) instead. *** ## Best use cases Once Qomon is connected to your AI assistant, here are the prompts that work best. On ChatGPT Plus/Pro, MCP connectors are **read-only**. Write operations require a Business or Enterprise plan. | Profile / Use case | What you can do | Example prompt | | :------------------- | :---------------------------------------------------------------- | :---------------------------------------------------------------------------------------- | | **Field organizer** | Search and segment contacts by tag, location, or support level | *"Find all contacts tagged 'volunteer' in London with an email address"* | | **Campaign manager** | Analyze your actions — participation, status, history | *"List my last 10 canvassing actions and their participation rates"* | | **Fundraising team** | Browse donations and transactions in real time | *"Show me all donations above \$500 from the last 30 days"* | | **Operations** | Query your team members and org structure | *"Who are the active users in my organization?"* | | **Data entry** | Create or update contacts directly from a conversation | *"Create a new contact: Jane Smith, [jane@example.com](mailto:jane@example.com), London"* | | **Reporting** | Cross-reference contacts, actions and donations for a quick brief | *"How many contacts attended an action in Birmingham this year?"* |