MCP 服务器

BigQuery Validator

Enables validation and dry-run analysis of BigQuery SQL queries without execution. Provides cost estimates, schema previews, and syntax validation for BigQuery queries.

README

mcp-bigquery

PyPI - Downloads

The mcp-bigquery package provides a minimal MCP server for BigQuery SQL validation and dry-run analysis. This server provides exactly two tools for validating and analyzing BigQuery SQL queries without executing them.

** IMPORTANT: This server does NOT execute queries. All operations are dry-run only. Cost estimates are approximations based on bytes processed.**

Features

SQL Validation: Check BigQuery SQL syntax without running queries
Dry-Run Analysis: Get cost estimates, referenced tables, and schema preview
Parameter Support: Validate parameterized queries
Cost Estimation: Calculate USD estimates based on bytes processed

Quick Start

Prerequisites

Python 3.10+
Google Cloud SDK with BigQuery API enabled
Application Default Credentials configured

Installation

From PyPI (Recommended)

# Install from PyPI
pip install mcp-bigquery

# Or with uv
uv pip install mcp-bigquery

From Source

# Clone the repository
git clone https://github.com/caron14/mcp-bigquery.git
cd mcp-bigquery

# Install with uv (recommended)
uv pip install -e .

# Or install with pip
pip install -e .

Authentication

Set up Application Default Credentials:

gcloud auth application-default login

Or use a service account key:

export GOOGLE_APPLICATION_CREDENTIALS=/path/to/service-account-key.json

Configuration

Environment Variables

Variable	Description	Default
`BQ_PROJECT`	GCP project ID	From ADC
`BQ_LOCATION`	BigQuery location (e.g., US, EU, asia-northeast1)	None
`SAFE_PRICE_PER_TIB`	Default price per TiB for cost estimation	5.0

Claude Code Integration

Add to your Claude Code configuration:

{
  "mcpServers": {
    "mcp-bigquery": {
      "command": "mcp-bigquery",
      "env": {
        "BQ_PROJECT": "your-gcp-project",
        "BQ_LOCATION": "asia-northeast1",
        "SAFE_PRICE_PER_TIB": "5.0"
      }
    }
  }
}

Or if installed from source:

{
  "mcpServers": {
    "mcp-bigquery": {
      "command": "python",
      "args": ["-m", "mcp_bigquery"],
      "env": {
        "BQ_PROJECT": "your-gcp-project",
        "BQ_LOCATION": "asia-northeast1",
        "SAFE_PRICE_PER_TIB": "5.0"
      }
    }
  }
}

Tools

bq_validate_sql

Validate BigQuery SQL syntax without executing the query.

Input:

{
  "sql": "SELECT * FROM dataset.table WHERE id = @id",
  "params": {"id": "123"}  // Optional
}

Success Response:

{
  "isValid": true
}

Error Response:

{
  "isValid": false,
  "error": {
    "code": "INVALID_SQL",
    "message": "Syntax error at [3:15]",
    "location": {
      "line": 3,
      "column": 15
    },
    "details": [...]  // Optional
  }
}

bq_dry_run_sql

Perform a dry-run to get cost estimates and metadata without executing the query.

Input:

{
  "sql": "SELECT * FROM dataset.table",
  "params": {"id": "123"},  // Optional
  "pricePerTiB": 6.0  // Optional, overrides default
}

Success Response:

{
  "totalBytesProcessed": 1073741824,
  "usdEstimate": 0.005,
  "referencedTables": [
    {
      "project": "my-project",
      "dataset": "my_dataset",
      "table": "my_table"
    }
  ],
  "schemaPreview": [
    {
      "name": "id",
      "type": "STRING",
      "mode": "NULLABLE"
    },
    {
      "name": "created_at",
      "type": "TIMESTAMP",
      "mode": "REQUIRED"
    }
  ]
}

Error Response:

{
  "error": {
    "code": "INVALID_SQL",
    "message": "Table not found: dataset.table",
    "details": [...]  // Optional
  }
}

Examples

Validate a Simple Query

# Tool: bq_validate_sql
{
  "sql": "SELECT 1"
}
# Returns: {"isValid": true}

Validate with Parameters

# Tool: bq_validate_sql
{
  "sql": "SELECT * FROM users WHERE name = @name AND age > @age",
  "params": {
    "name": "Alice",
    "age": 25
  }
}

Get Cost Estimate

# Tool: bq_dry_run_sql
{
  "sql": "SELECT * FROM `bigquery-public-data.samples.shakespeare`",
  "pricePerTiB": 5.0
}
# Returns bytes processed, USD estimate, and schema

Analyze Complex Query

# Tool: bq_dry_run_sql
{
  "sql": """
    WITH user_stats AS (
      SELECT user_id, COUNT(*) as order_count
      FROM orders
      GROUP BY user_id
    )
    SELECT * FROM user_stats WHERE order_count > 10
  """
}

Testing

Run tests with pytest:

# Run all tests (requires BigQuery credentials)
pytest tests/

# Run only tests that don't require credentials
pytest tests/test_min.py::TestWithoutCredentials

Development

# Install development dependencies
uv pip install -e ".[dev]"

# Run the server locally
python -m mcp_bigquery

# Or using the console script
mcp-bigquery

Limitations

No Query Execution: This server only performs dry-runs and validation
Cost Estimates: USD estimates are approximations based on bytes processed
Parameter Types: Initial implementation treats all parameters as STRING type
Cache Disabled: Queries always run with use_query_cache=False for accurate estimates