Vaiz MCP

What Vaiz MCP does

Vaiz MCP MCP server connects an MCP-compatible assistant to a Vaiz workspace. It exposes workspace information and task operations so an assistant can search across Vaiz entities, inspect project and board information, and work with documents and milestones.

The integration is listed in the Workplace & Productivity category and is suited to workflows where an agent needs to find or update work managed in Vaiz. Supported assistant clients documented by the project are Cursor and Claude Desktop.

How it works

Vaiz MCP MCP server runs as an npm package and communicates with Vaiz through the MCP API. The server authenticates with a Vaiz API key supplied as a bearer token. VAIZ_SPACE_ID can identify the Vaiz space to use, while VAIZ_API_URL can override the default MCP endpoint, https://api.vaiz.com/mcp.

Once configured, the client starts the server and discovers its available tools through MCP. The project also exposes a programmatic client that can initialize a connection, list tools, and call a named tool with arguments. The README shows a search call using the search tool with a query and an entity type.

Setup and configuration

Install the package globally with npm or run it directly with npx:

npm install -g vaiz-mcp
npx vaiz-mcp

Set VAIZ_API_TOKEN to a Vaiz API key. Optionally set VAIZ_SPACE_ID to select a space and VAIZ_API_URL to use a different MCP API URL. Set VAIZ_DEBUG=true to write diagnostic output to stderr.

For Cursor, add an mcpServers.vaiz entry to ~/.cursor/mcp.json. For Claude Desktop, add the equivalent entry to claude_desktop_config.json. Both examples start the package with npx and pass the credentials through the env object. The package can also be built and run from a cloned repository with npm install, npm run build, and npm run dev.

Tools and capabilities

Vaiz MCP MCP server provides access to these documented areas:

• Search for tasks, projects, and users.
• Create and edit tasks.
• Add task comments.
• View boards and projects.
• Retrieve team member information.
• Work with documents and milestones.

The exact tool list can be inspected through the programmatic client’s listTools() method after initialization. Tool calls can then be made with callTool() and a tool name plus input arguments.

Limitations and notes

The provided material does not specify individual tool schemas beyond the search example, nor does it document permissions, pagination, supported task fields, or error handling. A Vaiz API token is required; the space ID, API URL, and debug flag are optional according to the configuration table. Debug output is sent to stderr rather than returned as workspace data.

The README documents Cursor and Claude Desktop configurations only. It does not confirm compatibility with other MCP clients. The package is released under the MIT license.

Getting started with this Vaiz MCP MCP server

Always refer to the official documentation for the most accurate and up-to-date information.