Data · Google Cloud

Set up BigQuery

Google-hosted endpoint to explore BigQuery datasets and run SQL. Cloud SQL and Spanner have their own.

Publisher: Google Cloud Official
Hosted endpoint: https://bigquery.googleapis.com/mcp
Transports: streamable-http
Authentication: Google OAuth with the BigQuery scope; IAM decides what the agent may do.
What it can touch: Read and write, governed by IAM.
Status: Active

Install

The vendor hosts this server, so you only add its URL and nothing runs on your machine. If it needs authentication, the row above says how; OAuth servers prompt you on first connection.

Claude Code

shell

claude mcp add --transport http --scope project google-bigquery https://bigquery.googleapis.com/mcp

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "google-bigquery": {
      "url": "https://bigquery.googleapis.com/mcp"
    }
  }
}

Codex

~/.codex/config.toml

[mcp_servers.google-bigquery]
url = "https://bigquery.googleapis.com/mcp"

Config for other clients →

Other servers in data

Airtable

Query, create and update records in Airtable bases, limited to the bases you authorise.

Google

MCP Toolbox for Databases

One server with prebuilt tool sets for Postgres, MySQL, BigQuery, AlloyDB, Spanner and other databases.

Supabase

Manage Supabase projects from an agent: run SQL, apply migrations, read logs and deploy Edge Functions.

Neon

Create Neon Postgres projects and branches, run SQL and do branch-based migrations. Meant for development use.

MongoDB

Query and administer MongoDB databases and Atlas clusters, using a connection string or an Atlas service account.

Redis

Lets an agent work with Redis data structures, search and vector indexes in plain language.

Sources

docs.cloud.google.com/bigquery/docs/use-bigquery-mcp
docs.cloud.google.com/mcp/supported-products