mcp-registry/mcp-server-bigquery

    ==================
      
       /// MCP ///
      /// MCP ///
        
    ==================
        
    [server:online]
    [protocol:ready]

mcp-server-bigquery

by LucasHild

A Model Context Protocol (MCP) server that lets LLMs inspect BigQuery schemas and run SQL queries.

113
29
Specialized
GitHub

01

execute-query

Executes a SQL query using BigQuery dialect

02

list-tables

Lists all tables in the BigQuery database

03

describe-table

Describes the schema of a specific table

Installation

1. Clone the repository
git clone https://github.com/LucasHild/mcp-server-bigquery.git cd mcp-server-bigquery
2. Create a virtual environment (optional but recommended)
python -m venv .venv source .venv/bin/activate # Windows: .venv\Scripts\activate
3. Install Python dependencies
pip install -r requirements.txt
4. Provide Google BigQuery credentials
• Create a GCP service-account with BigQuery read/write permissions.
• Download its JSON key and export its path:
export GOOGLE_APPLICATION_CREDENTIALS=/absolute/path/key.json
5. Configure the MCP server
• Copy the sample configuration and edit it:
• Set values for project_id, default_dataset, allowed_tables, host, and port.
cp config.sample.yaml config.yaml
6. Run database migrations (if any)
python scripts/init_db.py # only if the repo supplies a migration script
7. Start the server
python -m mcp_server_bigquery --config config.yaml # or uvicorn mcp_server_bigquery.app:app --host 0.0.0.0 --port 8080
8. Verify
curl http://localhost:8080/healthz

Documentation

License: MIT License
Updated 7/30/2025