GoodData MCP Server

Give Claude Code direct access to your GoodData analytics platform.

A Model Context Protocol (MCP) server that enables AI assistants like Claude to interact with GoodData, allowing natural language queries of your business intelligence data. Also includes a standalone CLI for direct terminal access.

Why Use This?

• AI-Powered Analytics: Ask Claude questions about your GoodData dashboards, metrics, and insights in natural language
• Read-Only by Design: All operations are strictly read-only, so there's no risk of modifying your data
• Zero Configuration: Works with Claude Code out of the box (just add your credentials)
• Export Capabilities: Export dashboards to PDF and visualizations to CSV/Excel
• CLI Included: Use the gooddata command directly from your terminal

Features

Feature	Description
List Resources	Workspaces, dashboards, insights, metrics, datasets, users, groups
Query Data	Retrieve data from any insight/visualization
Export Reports	Export dashboards to PDF, visualizations to CSV/XLSX
Explore Data Models	Get the logical data model (LDM) for documentation
User Management	List users, groups, and group memberships

Quick Start

1. Install

# Clone the repository
git clone https://github.com/aalexmrt/gooddata-mcp-server.git
cd gooddata-mcp-server

# Create virtual environment
python3 -m venv .venv
source .venv/bin/activate

# Install with MCP support
pip install -e ".[mcp]"

2. Configure

cp .env.example .env

Edit .env with your GoodData credentials:

GOODDATA_HOST=https://your-org.cloud.gooddata.com
GOODDATA_TOKEN=your-api-token
GOODDATA_WORKSPACE=your-default-workspace-id

Getting an API Token: Go to your GoodData profile settings and create a Personal Access Token.

3. Add to Claude Code

claude mcp add gooddata \
  -s user \
  -e GOODDATA_HOST="https://your-org.cloud.gooddata.com" \
  -e GOODDATA_TOKEN="your-api-token" \
  -e GOODDATA_WORKSPACE="your-workspace-id" \
  -- /path/to/gooddata-mcp-server/.venv/bin/python -m gooddata_cli.mcp_server

4. Start Using

In Claude Code, you can now ask questions like:

• "List all dashboards in my GoodData workspace"
• "What metrics are available?"
• "Show me the data from the Revenue Overview insight"
• "Export the Sales Dashboard to PDF"

---

MCP Tools Reference

Tool	Description
list_workspaces	List all available workspaces
list_insights	List all insights (visualizations) in a workspace
list_dashboards	List all dashboards in a workspace
list_metrics	List all metrics in a workspace
list_datasets	List all datasets in a workspace
get_dashboard_insights	Get all insights contained in a dashboard
get_dashboard_filters	Get all filters configured on a dashboard
get_insight_metadata	Get detailed metadata for an insight
get_insight_data	Get data from an insight
get_logical_data_model	Get the workspace's logical data model
list_users	List all users in the organization
list_user_groups	List all user groups
get_user_group_members	Get members of a specific group
export_dashboard_pdf	Export a dashboard to PDF
export_visualization_csv	Export a visualization to CSV
export_visualization_xlsx	Export a visualization to Excel

MCP Server Management

# Check server status
claude mcp list

# Remove the server
claude mcp remove gooddata

---

CLI Usage

The package also includes a standalone CLI for direct terminal access.

List Resources

# List all workspaces
gooddata list workspaces

# List insights in a workspace
gooddata list insights -w <workspace_id>

# List dashboards
gooddata list dashboards -w <workspace_id>

# List metrics
gooddata list metrics -w <workspace_id>

# List datasets
gooddata list datasets -w <workspace_id>

# JSON output
gooddata list workspaces --json

Query Data

# Get data from a specific insight
gooddata insight <insight_id> -w <workspace_id>

Export

# Export dashboard to PDF
gooddata export pdf <dashboard_id> -w <workspace_id>

# Export visualization to CSV
gooddata export csv <visualization_id> -w <workspace_id>

# Export visualization to Excel
gooddata export xlsx <visualization_id> -w <workspace_id>

# Custom output path
gooddata export pdf <dashboard_id> -o ./my-report.pdf

---

Python API

You can also use the package programmatically:

from gooddata_cli.query import list_workspaces, list_insights
from gooddata_cli.export import export_dashboard_pdf, export_visualization_tabular

# List workspaces
workspaces = list_workspaces()

# List insights
insights = list_insights(workspace_id="my_workspace")

# Export dashboard
path = export_dashboard_pdf("dashboard_id", workspace_id="my_workspace")

---

Project Structure

gooddata-mcp-server/
├── .env.example           # Environment template
├── pyproject.toml         # Package configuration
├── README.md              # This file
├── LICENSE                # MIT License
├── CONTRIBUTING.md        # Contribution guidelines
└── src/gooddata_cli/
    ├── __init__.py        # Package exports
    ├── sdk.py             # SDK initialization
    ├── query.py           # Query operations
    ├── export.py          # Export operations
    ├── cli.py             # CLI entry point
    └── mcp_server.py      # MCP server for Claude Code

---

Security Note

All operations are read-only. This MCP server cannot create, update, or delete any data in GoodData. Operations are limited to:

• Listing resources (workspaces, insights, dashboards, metrics, datasets, users, groups)
• Querying existing data
• Exporting reports to local files

---

Dependencies

• gooddata-sdk - Official GoodData Python SDK
• gooddata-pandas - Pandas integration for GoodData
• mcp - Model Context Protocol SDK
• click - CLI framework
• rich - Terminal formatting
• python-dotenv - Environment variable loading

---

Contributing

Contributions are welcome! Please see CONTRIBUTING.md for guidelines.

---

License

MIT License - see LICENSE for details.

---

Related Projects

• GoodData Python SDK - Official GoodData SDK
• Model Context Protocol - MCP specification
• Claude Code - AI coding assistant