Google BigQuery

Google BigQuery MCP Connector for Claude

D

Empower your AI agent to query massive datasets via BigQuery — execute Standard SQL, track active jobs, and inspect table schemas natively.

7 tools Official Updated Jun 28, 2026 Official Vinkius Partner

Connect your Google BigQuery data warehouse to any AI agent and empower it to act as a fractional data analyst. Traverse structured schemas, audit data pipelines, and execute complex aggregations over petabytes of data purely through conversational prompts.

What you can do

  • Execute Queries — Prompt natively structural Data Analytics requests and allow the LLM to write, run, and summarize exact Standard SQL instantly
  • Discover Schemas — Inspect deep table column mappings, discovering strict clustering logic and native partitioning limits
  • Audit Workloads — Paginate recent cluster jobs, identify heavily delayed computations globally, and read bytes explicitly processed by runs
  • Dataset Topologies — Traverse nested datasets logically mapping GCP access properties recursively
  • Performance Troubleshooting — Read exact job error traces directly confirming syntax failures natively

How it works

  1. Subscribe to this server
  2. Enter your GCP Project ID and an active OAuth/Service Account Token
  3. Start querying terabytes of rows securely from Claude, Cursor, or your preferred agent workspace

Stop switching into the GCP Console for quick data validations. Check database constraints and summarize recent daily logs all from your chat.

Who is this for?

  • Data Engineers — troubleshoot failing scheduled queries and explore undocumented columns securely on-the-fly
  • Marketing Analysts — request customer cohorts using conversational logic that natively translates to optimized SQL
  • Backend Developers — rapidly confirm if application background pipelines successfully inserted the necessary rows without breaking flow
sqldata-warehousebig-datacloud-computingdata-pipelinesquery-optimization

7 tools expose this connector's capabilities to your AI agent.

list_datasets

List all explicit Datasets in the active GCP project

get_dataset

Get exact details of a specific BigQuery dataset

list_tables

List explicit Tables natively contained within a Dataset

get_table

Get explicit metadata and schema details of a pure BigQuery Table

execute_query

Run an explicit BigQuery Standard SQL command

list_jobs

List recent explicit BigQuery runtime Jobs securely

get_job

Get complete details of a specific BigQuery Job run

See how to talk to your AI agent using Google BigQuery.

Get the table schema for `users_prod` in the `analytics` dataset.

Schema fetched. `users_prod` contains 12 columns, notably `user_id` (STRING, required), `signup_timestamp` (TIMESTAMP, partitioned), and `plan_tier` (STRING). Would you like to check some sample rows?

Find out the top 3 countries with the most signups this month in the `users` table.

Executing SQL... The query completed successfully processing 12MB. The top 3 countries are: 1) United States (12,400), 2) Brazil (8,900), 3) Japan (4,150).

Did the overnight cron job compute successfully or did it fail?

Listing recent jobs. Job `cron_transform_01` (ID: 11a0ab3-xxx) failed globally at 03:00AM. Root cause trace indicates native `Unrecognized name: user_account_id` syntax failure halting workflow explicitly mapped.

Yes! The agent will typically use `list_tables` and `get_table` to study the columns first. Then, realizing constraints, it will natively invoke `execute_query` running an optimized Standard SQL string to fetch exactly what you asked for.

Related Connectors