Skip to main content
Glama
README.md
# vaiz-mcp

[![smithery badge](https://smithery.ai/badge/vaiz/vaiz)](https://smithery.ai/servers/vaiz/vaiz)

MCP (Model Context Protocol) client for Vaiz — connect Cursor/Claude to your Vaiz workspace.

## Installation

```bash
npm install -g vaiz-mcp
```

Or use directly via npx:

```bash
npx vaiz-mcp
```

## Configuration

### Environment Variables

| Variable | Required | Description |
|----------|----------|-------------|
| `VAIZ_API_TOKEN` | Yes | Your Vaiz API key (Bearer token) |
| `VAIZ_SPACE_ID` | No | Your Vaiz Space ID |
| `VAIZ_API_URL` | No | MCP API URL (default: `https://api.vaiz.com/mcp`) |
| `VAIZ_DEBUG` | No | Set to `true` for debug output to stderr |

### Cursor Configuration

Create or edit `~/.cursor/mcp.json`:

```json
{
  "mcpServers": {
    "vaiz": {
      "command": "npx",
      "args": ["vaiz-mcp@latest"],
      "env": {
        "VAIZ_API_TOKEN": "your-api-key",
        "VAIZ_SPACE_ID": "your-space-id"
      }
    }
  }
}
```

Or after global installation (`npm install -g vaiz-mcp`):

```json
{
  "mcpServers": {
    "vaiz": {
      "command": "vaiz-mcp@latest",
      "env": {
        "VAIZ_API_TOKEN": "your-api-key",
        "VAIZ_SPACE_ID": "your-space-id"
      }
    }
  }
}
```

### Claude Desktop Configuration

Add to `claude_desktop_config.json`:

```json
{
  "mcpServers": {
    "vaiz": {
      "command": "npx",
      "args": ["vaiz-mcp@latest"],
      "env": {
        "VAIZ_API_TOKEN": "your-api-key",
        "VAIZ_SPACE_ID": "your-space-id"
      }
    }
  }
}
```

## Usage

After configuration, Cursor/Claude will automatically connect to your Vaiz workspace and gain access to:

- 🔍 Search tasks, projects, users
- 📋 Task management (create, edit, comments)
- 📊 View boards and projects
- 👥 Team member information
- 📝 Work with documents and milestones

## Debugging

To enable debug output, add the `VAIZ_DEBUG` variable:

```json
{
  "mcpServers": {
    "vaiz": {
      "command": "npx",
      "args": ["vaiz-mcp@latest"],
      "env": {
        "VAIZ_API_TOKEN": "your-api-key",
        "VAIZ_DEBUG": "true"
      }
    }
  }
}
```

Debug messages will be output to stderr.

## Programmatic Usage

You can also use the library programmatically:

```typescript
import { createVaizMCPClient } from 'vaiz-mcp';

const client = createVaizMCPClient({
  apiKey: 'your-api-key',
  spaceId: 'your-space-id',
});

// Initialize connection
const initResult = await client.initialize();

// Get list of tools
const tools = await client.listTools();

// Call a tool
const result = await client.callTool('search', { 
  query: 'important task',
  entityType: 'task' 
});
```

## Development

```bash
# Clone the repository
git clone https://github.com/vaiz/vaiz-mcp.git
cd vaiz-mcp

# Install dependencies
npm install

# Build the project
npm run build

# Run in development mode
npm run dev
```

## Platforms

[![LobeHub](https://lobehub.com/badge/mcp/vaiz-vaiz-mcp)](https://lobehub.com/mcp/vaiz-vaiz-mcp)

## License

MIT

TDQS

B3.2/5.0

Scored across 33 tools

Disambiguation5/5

Each tool targets a distinct entity or action (e.g., create_task vs edit_task vs get_task), and descriptions clarify boundaries like edit_document_content vs edit_task. No obvious overlap between tools.

Naming Consistency4/5

Most tools follow a consistent verb_noun pattern (create_, get_, list_, edit_), but 'current_user' deviates from verb_noun, and some singular/plural mismatches (get_task vs get_tasks) reduce consistency slightly.

Tool Count4/5

33 tools cover a broad project management domain (tasks, milestones, documents, boards, etc.) and is on the higher side but still reasonable for a comprehensive MCP server. Not excessive, but could be trimmed.

Completeness2/5

Missing critical lifecycle operations: no delete tool for any entity, no update for milestones or documents, no create document tool. The set covers reading and creation but lacks key CRUD operations, leading to potential agent dead-ends.

Maintenance

ActivitySlowing
ResponsivenessUnresponsive