Google Cloud

BigQuery

Included as a governed analytical warehouse with a Google-managed MCP endpoint that can expose BigQuery metadata and query execution to agents, including an explicitly separable read-only SQL path and a denyable unrestricted SQL tool.

Automation, data & MCP bridgesSystem of orchestrationAgent-accessible SaaS
Representative source ↗Product site ↗Suggest a correction
An illustration of access surfaces surrounding a governed system.
System roleSystem of orchestration
Access maturityAgent-accessible SaaS
MCP supportOfficial MCP server
Reference updatedJul 10, 2026
BYO-UI reviewedJul 22, 2026

Concrete capability record

Data, retrieval, actions, identity, and operating limits

A field-by-field summary of what the reviewed first-party references actually support. Publisher update dates and BYO-UI review dates are shown separately below.

01 · Records and state owned

Projects, datasets, tables, schemas and metadata, BigQuery jobs, SQL query results and statistics, and catalog resources available to the authenticated principal.

02 · Data you can retrieve

List BigQuery resources, retrieve metadata, and run SQL through the preferred read-only execution tool. The server also exposes tool discovery without authentication.

03 · Actions and write paths

The managed `execute_sql` tool can run any BigQuery SQL, including INSERT, UPDATE, DELETE, CREATE, and AI/ML statements. Google recommends the read-only tool for retrieval, and administrators can deny calls to `execute_sql` with an IAM deny policy.

04 · Authentication and permissions

The managed server uses OAuth 2.0 and Google Cloud IAM; API keys are not supported. Typical query access requires MCP Tool User, BigQuery Job User, and BigQuery Data Viewer, with additional resource permissions for the requested task.

05 · Monitoring, approval, and recovery

IAM roles and deny policies govern tool calls and data access. `execute_sql` supports dry runs, identifies the billing project, labels jobs `goog-mcp-server:true`, and returns bytes processed or billed and DML-row statistics for review and monitoring.

06 · Limits and caveats

Managed query tools do not support Google Drive external tables. The default execution timeout is three minutes and results are limited to 3,000 rows. The read-only tool rejects DML, DDL, and Python UDFs; query execution consumes BigQuery resources and can incur charges.

First-party MCP tool catalog

6 named MCP tools · BigQuery core managed MCP server

The exact tools published for the managed bigquery.googleapis.com/mcp endpoint; Data Transfer Service and Migration are separate server variants below. Reviewed Jul 10, 2026. Invocation requires Google OAuth and IAM permissions. Query jobs bill the project supplied in the request. execute_sql_readonly accepts SELECT only; execute_sql can run DML, DDL, and AI or ML statements and therefore requires explicit confirmation.

First-party exact listComplete list on cited pagebigquery.googleapis.com/mcp
Dataset and table metadata4
list_dataset_idsreadget_dataset_inforeadlist_table_idsreadget_table_inforead
SQL execution2
execute_sql_readonlyreadexecute_sqlconsequential write
Catalog evidence: BigQuery MCP Reference: bigquery.googleapis.com

First-party MCP tool catalog

13 named MCP tools · BigQuery Data Transfer Service managed MCP server

The exact tools published for the separate bigquerydatatransfer.googleapis.com/mcp endpoint. Reviewed Jul 10, 2026. This is a service-specific MCP endpoint, not an extension of the core endpoint's live catalog. Creating, updating, or deleting transfer configurations and starting or deleting runs can change scheduled data movement, incur costs, and affect downstream datasets.

First-party exact listComplete list on cited pagebigquerydatatransfer.googleapis.com/mcp
13 shown
Data-source discovery3
list_data_sourcesreadget_data_sourcereadcheck_valid_credsread
Transfer configurations5
create_transfer_configconsequential writeupdate_transfer_configconsequential writedelete_transfer_configconsequential writeget_transfer_configreadlist_transfer_configsread
Transfer runs and logs5
start_manual_transfer_runsconsequential writelist_transfer_runsreadget_transfer_runreaddelete_transfer_runconsequential writelist_transfer_logsread
Catalog evidence: BigQuery Data Transfer Service MCP Reference

First-party MCP tool catalog

5 named MCP tools · BigQuery Migration managed MCP server

The exact tools published for the separate bigquerymigration.googleapis.com/mcp endpoint. Reviewed Jul 10, 2026. These tools produce or retrieve translations and DDL suggestions; they do not execute the suggested DDL against a BigQuery dataset. Google OAuth, IAM, quotas, and migration-service availability still apply.

First-party exact listComplete list on cited pagebigquerymigration.googleapis.com/mcp
Query translation3
translate_querydraftget_translationreadexplain_translationread
DDL suggestions2
generate_ddl_suggestiondraftfetch_ddl_suggestionread
Catalog evidence: BigQuery Migration MCP Reference

First-party connector catalog

32 named connectors · BigQuery Data Transfer Service supported sources

The named inbound data sources on the first-party BigQuery Data Transfer Service overview, with Preview labels retained in item descriptions. Reviewed Jul 10, 2026. This connector catalog describes managed inbound transfers into BigQuery and is separate from all MCP tool catalogs. The service does not transfer data out of BigQuery. Availability, authentication, regions, schedules, pricing, backfill behavior, and data-delivery SLO coverage vary by source; Preview labels are time-sensitive.

First-party exact listVerified partial listBigQuery Data Transfer Service
32 shown
SaaS platforms3
Salesforceread/writeSalesforce Marketing Cloudread/writeServiceNowread/write
Marketing platforms4
Facebook Adsread/writeHubSpotread/writeKlaviyoread/writeMailchimpread/write
Payment and commerce platforms3
PayPalread/writeStriperead/writeShopifyread/write
Databases and data warehouses8
Amazon Redshiftread/writeApache Hive Metastoreread/writeMicrosoft SQL Serverread/writeMySQLread/writeOracleread/writePostgreSQLread/writeSnowflakeread/writeTeradataread/write
Cloud storage3
Cloud Storageread/writeAmazon Simple Storage Service (Amazon S3)read/writeAzure Blob Storageread/write
Google services11
Campaign Managerread/writeComparison Shopping Service (CSS) Centerread/writeDisplay & Video 360read/writeGoogle Adsread/writeGoogle Ad Managerread/writeGoogle Analytics 4read/writeGoogle Merchant Centerread/writeSearch Ads 360read/writeGoogle Playread/writeYouTube Channelread/writeYouTube Content Ownerread/write
Catalog evidence: BigQuery Data Transfer Service: Supported data sources

First-party access

Documented surfaces

BigQuery SQL ↗BigQuery provides native SQL query execution, including through its documented MCP surface. Google Cloud IAM, OAuth, dataset permissions, location, quotas, compute cost, and sensitive-data controls govern what can be queried and whether a query may write data.Source: Use the BigQuery MCP serverAPI ↗Developer accessDeveloper docs ↗SDKs and platform guidanceProduct ↗Native platform experience

Agent access

MCP support

Official MCP serverGoogle-managed remote MCPCurrent managed service

Support: Official MCP server

Read scope: Use resource and metadata tools plus the read-only SQL tool for dataset inspection and query results; each query remains bounded by the caller's IAM access and the selected billing project.

Write scope: 8 of 24 reviewed named tools have a write, draft, or mixed action label, including execute_sql, create_transfer_config, update_transfer_config, delete_transfer_config, start_manual_transfer_runs, and others. Read each catalog boundary before enabling them.

Authentication: OAuth 2.0 access tokens for Google Cloud identities at the managed endpoint; API-key authentication is unavailable.

Approval boundary: Default analytical workflows to the read-only SQL tool. Before unrestricted `execute_sql`, show the billed project, statement, target datasets or tables, dry-run bytes, and whether the query contains DML, DDL, AI/ML, or UDF work; require confirmation for state changes or material cost.

Protocol does not erase product boundaries.

Confirm the current tool catalog, plan, region, scopes, rate limits, terms, and write behavior before implementation.

Editorial assessment

Access-maturity dimensions

A comparative architecture lens—not a quality score, market ranking, or buying recommendation. Scale: 1–5.

Data access4
Action access4
Event access1
Identity4
Governance4
Agent access4
UI extensibility3
Portability4
Observability4
Documentation4

Proposed customer-shaped experiences

What customers could create on top.

The output could be an export, report, graph, artifact, application, agent, workflow, or downstream feed. These proposals are derived from documented access—not claims that Google Cloud ships them.

Single-platform patterns

  • Warehouse-backed campaign and customer-data explorer
  • Marcom control plane
  • Data-quality exception console
  • Approved capability registry
Use in the brief builder →

Multi-platform compositions

  • BigQuery + systems of record: marcom control plane
  • BigQuery + systems of engagement + interface host or builder: cross-system decision workspace
Explore composition recipes →

Evidence and dates

First-party references, with publisher and review dates separated

“Publisher updated” is shown only when the page exposes an update date. “BYO-UI reviewed” records when this research checked the reference. A missing publisher date is reported as missing—not replaced with the review date.

8 recorded sources
MCPRepresentative source
https://docs.cloud.google.com/bigquery/docs/use-bigquery-mcp ↗

Official Google Cloud docs confirm remote server enabled with BigQuery API and governed by IAM/OAuth; query execution can incur cost and access sensitive data.

Publisher updated
Jul 10, 2026
BYO-UI reviewed
Jul 12, 2026

Targeted first-party MCP/agent-role review; broader API inventory retained

APISource inventory
https://cloud.google.com/bigquery/docs/reference/rest ↗

First-party API reference or API overview recorded for this platform.

Publisher updated
May 30, 2026
BYO-UI reviewed
Jul 10, 2026

Inventory source: structurally normalized; content was not individually reopened in this pass.

Developer docsSource inventory
https://cloud.google.com/bigquery/docs ↗

First-party developer documentation or platform overview.

Publisher updated
Not stated by publisher
BYO-UI reviewed
Jul 10, 2026

Inventory source: structurally normalized; content was not individually reopened in this pass.

HomepageSource inventory
https://cloud.google.com/bigquery ↗

First-party product homepage used to confirm product identity and current positioning.

Publisher updated
Not stated by publisher
BYO-UI reviewed
Jul 10, 2026

Inventory source: structurally normalized; content was not individually reopened in this pass.

MCP tool catalogTargeted review
BigQuery Migration MCP Reference ↗

Named catalog evidence used on this profile.

Publisher updated
Mar 25, 2026
BYO-UI reviewed
Jul 10, 2026

Targeted first-party catalog review

Verified fact

Tied to cited first-party evidence reviewed for this profile.

Source inventory

Official links recorded for deeper research but not necessarily reopened endpoint by endpoint.

Editorial assessment

System role, maturity interpretation, and architectural boundary.

Proposed design

Interface patterns and compositions—not vendor product claims.