How to connect to Vaiz MCP in Claude and Cursor

MCP allows AI tools like Claude and Cursor to securely access your Vaiz workspace: tasks, projects, documents, and more, directly inside the editor or chat.

There are two ways to connect:

• OAuth connector (recommended). Connect Claude to Vaiz in a couple of clicks. No API tokens, no configuration files, nothing to install.
• Local MCP server with a Personal Access Token. For Cursor and other tools that run MCP servers locally.

Connect Vaiz to Claude with OAuth

This is the fastest way to connect. Claude talks to the Vaiz MCP server directly over the internet, and you authorize access with your Vaiz account, the same way you would sign in with Google. The setup takes about two minutes.

You'll need the Claude Desktop app and a Vaiz account.

Step 1: Open settings in Claude

Open Claude Desktop and click Customize in the left sidebar.

Step 2: Add a custom connector

Go to the Connectors section, click Add in the top right corner, and select Add custom connector.

Step 3: Enter the Vaiz server details

In the dialog, fill in two fields and click Add:

• Name: Vaiz
• URL: https://api.vaiz.com/mcp

Step 4: Connect your account

Vaiz now appears in your connectors list. Open it and click Connect. Claude will open the Vaiz authorization page in your browser.

Step 5: Authorize Claude in Vaiz

The authorization page shows which space Claude is requesting access to and exactly what it will be able to do: read and write tasks, projects, and documents, and read members and notifications.

Review the permissions and click Authorize.

Step 6: Return to Claude

Once authorized, click Open Claude in the browser prompt to return to the desktop app. The connection is now active, and you can ask Claude about your tasks right away.

Step 7: Fine-tune tool permissions (optional)

By default, Claude asks for your approval before using Vaiz tools. You can adjust this per tool in the connector settings: set read-only tools like search to Always allow, keep write actions like creating tasks on Needs approval, or block specific tools entirely.

Note: Claude is now connected to Vaiz. The same connector works in Claude Desktop, on claude.ai, and in the Claude mobile apps.

Connect Cursor with a Personal Access Token

Use this method if you work in Cursor, or if your setup requires running the Vaiz MCP server locally with token-based access.

Prerequisites: install Node.js

Before proceeding with the setup, you must have Node.js installed on your computer. Both Claude Code and Cursor rely on the npx command to fetch and run the Vaiz MCP server.

If you haven't installed it yet, you can download it from the official Node.js website. Once installed, you can verify it by opening your terminal and typing npm -v or npx -v.

Connect Vaiz to Cursor

Step 1: Open MCP settings in Cursor

In Cursor, open Settings, then Tools & MCP. Scroll to Installed MCP Servers and click “Add Custom MCP”.

Step 2: Create or Edit your MCP configuration File

Cursor stores MCP server configurations in the file: ~/.cursor/mcp.json.

Add the Vaiz MCP configuration:

Use the example below to quickly add the Vaiz MCP configuration — just copy and paste it.

{
  "mcpServers": {
    "Vaiz": {
      "command": "npx",
      "args": [
        "vaiz-mcp@latest"
      ],
      "env": {
        "VAIZ_API_TOKEN": "your_personal_access_token",
        "VAIZ_SPACE_ID": "your_space_id (optional)"
      }
    }
  }
}

Step 3: Enable the Vaiz MCP Server

After saving the config file, return to Tools & MCP. You’ll now see Vaiz listed as an installed server — toggle it on.

Step 4: Start a new chat and use Vaiz Tools

Open a new chat and send a request that relies on your Vaiz data. For example: “Let’s see how to solve task DEMO-32”.

Cursor will call the Vaiz MCP tools and display structured task information.

Step 5: Continue your workflow using Vaiz Data

Cursor can now generate follow-up tasks, code, plans, and deliverables using your workspace context.

Note: Your Vaiz MCP integration is ready.

Conclusion

Once MCP is connected, Vaiz becomes directly accessible from tools like Claude and Cursor. You can ask questions about your tasks, get summaries, and work with real project data without leaving your AI workspace. After you set up your Personal Access Token and complete the MCP configuration, the connection works securely in the background and is ready whenever you need it.