mcp-bigquery-server
Author: ergut
Description: A Model Context Protocol (MCP) server that provides secure, read-only access to Google BigQuery so LLMs can run natural-language queries against your data.
Stars: 101
Forks: 26
License: MIT License
Category: Open Source
Overview
Installation
1. Prerequisites
• Node.js ≥ 18
• A Google Cloud project that has BigQuery enabled
• A BigQuery service-account JSON key with at least bigquery.jobs.create and bigquery.tables.getData permissions on the datasets you wish to expose.
2. Acquire credentials
export GOOGLE_APPLICATION_CREDENTIALS=/absolute/path/to/key.json3. Install the server
git clone https://github.com/ergut/mcp-bigquery-server.git
cd mcp-bigquery-server
npm install --production
(or install globally from NPM)
npm install -g mcp-bigquery-server4. Configure allowed datasets and server options
Create a config file (config.json):
{
"projectId": "your-gcp-project",
"allowedDatasets": ["project.dataset1", "project.dataset2"],
"port": 8080,
"auth": {
"mode": "bearer",
"tokens": ["<YOUR_API_TOKEN>"]
}
}5. Start the MCP server
npx mcp-bigquery-server --config ./config.json
# or, if installed globally
mcp-bigquery-server --config ./config.json6. Verify
curl http://localhost:8080/health
→ {"status":"ok"}
The server is now ready to service MCP requests from LLM tooling.License: MIT License
Updated 7/15/2025