Sisense MCP Server Quick Start Guide

Prerequisites

• Bun 1.0.0 or later (recommended) or Node.js 18.0.0 or later

• Access to a Sisense instance

• A Sisense API token

Clone and Install

If you have not done so already, clone the repository:

git clone git@gitlab.rnd.sisense.com:SisenseTeam/sisense-mcp-server.git
cd sisense-mcp-server
# Install dependencies
bun install
# or
npm install

Configure Environment

Create an .env file in the project root:

SISENSE_URL=https://your-instance.sisense.com
SISENSE_TOKEN=your-api-token-here
PORT=3000

Get Your API Token:

1. Log in to your Sisense instance.

2. Navigate to your profile settings.

3. Generate an API token.

4. Copy it to the .env file.

Start the Server

Enter the following command:

# Development mode (with hot reload)
bun run dev

# Or production mode
bun run build
bun run start

The following appears:

Sisense MCP Server (SSE) running on http://localhost:3000
MCP endpoint: http://localhost:3000/mcp
Health check: http://localhost:3000/health

Verify Installation

Test the endpoint by entering the following command:

curl http://localhost:3000/health

The following response should appear:

{
  "status": "ok",
  "activeSessions": 0
}

Configure Your MCP Client

For Cursor IDE:

1. Open Cursor settings and navigate to MCP settings.

2. Add the server configuration:
   

   Copy

   {
     "mcpServers": {
       "sisense": {
         "command": "bun",
         "args": ["run", "src/sse-server.ts"],
         "env": {
           "SISENSE_URL": "https://your-instance.sisense.com",
           "SISENSE_TOKEN": "your-api-token-here"
         }
       }
     }
   }

3. Restart Cursor.

For Claude Desktop:

1. Edit claude_desktop_config.json (location varies by OS).

2. Add the server configuration (same as above).

3. Restart Claude Desktop.

Test the Tools

Once connected, run the following commands in your MCP client to ensure it is working properly:

List Data Sources:

sisense: get data sources

Get Fields for a Data Source:

sisense: get fields for "Sample ECommerce"

Create a Chart:

sisense: build chart for "Sample ECommerce" showing total revenue by month

Common First Steps

• Explore Available Data Sources - start by listing all available data sources:

  Copy

  sisense: list data sources

• Examine a Data Source - pick a data source and see what fields are available:

  Copy

  sisense: get fields for "StarWars_Analytics"

• Create Your First Chart:

  Copy

  sisense: build chart for "StarWars_Analytics" showing top box office revenue by movie

• View the Chart:

  Sisense will render the chart will be and provide a URL of the screenshot will be provided. You can:

  • Open the URL in a browser

  • View it in your MCP client (if supported)

  • Access it via: http://localhost:3000/screenshots/[filename]

Troubleshooting

Server will not start:

Problem: "SISENSE_URL and SISENSE_TOKEN environment variables are required"

Solution:

• Check that .env file exists in project root

• Verify both variables are set

• Restart the server

Cannot Connect to Sisense:

Problem: "Failed to initialize Sisense clients"

Solution:

• Verify SISENSE_URL is correct and accessible

• Check that SISENSE_TOKEN is valid

• Test connectivity: curl https://your-instance.sisense.com/api/v1/...

Tools Not Appearing

Problem: MCP client does not show Sisense tools

Solution:

• Restart your MCP client

• Check server logs for errors

• Verify MCP client configuration is correct

• Ensure server is running: curl http://localhost:3000/health

Chart Creation Fails

Problem: "Failed to create chart"

Solution:

• Verify data source name is exact (case-sensitive)

• Check that data source has required fields

• Try a simpler prompt first

• Check server logs for detailed error

Next Steps

• Read the full readme for detailed documentation

• Check API for complete API reference

• Explore example prompts and use cases

• Review the code structure in src/ directory

Example Workflow

The following is an example of a complete workflow:

# 1. Start the server
bun run dev
# 2. In your MCP client, list data sources
sisense: get data sources
# 3. Pick a data source and explore fields
sisense: get fields for "Sample ECommerce"
# 4. Create a chart
sisense: build chart for "Sample ECommerce" showing top 10 products by sales
# 5. Create another chart with context
sisense: build chart for "Sample ECommerce" showing revenue trend over time
# 6. View charts at http://localhost:3000/screenshots/