Skip to content
actionlake
Start free

Connectors · Observability & analytics

Google BigQuery MCP server for AI agents.

actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to Google BigQuery and the rest of your systems. It gives your agent 134 Google BigQuery actions (66 reads, 44 writes and 24 that wait for a confirmation) and holds the Google BigQuery credential itself, so the agent never does. It paces Google BigQuery's limits and records every call. Connecting Google BigQuery needs your own Google BigQuery OAuth app and one admin sign-in.

Connect Google BigQuery freeSee pricing

100,000 free calls a month. No card, no email, no password: press Start a free sandbox on the next screen and keep the key it shows you.

claude mcp add --transport http actionlake https://connect.actionlake.com/mcp
the first call
do {  action: "google_bigquery.projects.list",  intent: "check which Google BigQuery account is connected"}# Projects: list · read · 1 upstream call
  • 134actionsin one signed file
  • 66readlook, never change
  • 44writecreate and update
  • 24wait for a yesdestructive, confirmed first

Set up

Connect Google BigQuery in three steps.

  1. 01

    Add actionlake to Claude.

    Run the command once in Claude Code; its consent page can create a free sandbox: 100,000 calls a month, no card. In the Claude app, add a custom connector with the URL below and paste your key on the consent page. Creating a new sandbox from the Claude app is not yet tested by us.

    https://connect.actionlake.com/mcp
  2. 02

    Connect Google BigQuery.

    OAuth 2.0, authorization code with PKCE. You register your own OAuth app with Google BigQuery, an admin signs in once, and you paste the app's client id and secret and the tokens from that sign-in; actionlake renews the access token from then on. actionlake stores what you paste write-only, sealed under your tenant's key.

    Before the first sign-in, Google BigQuery needs an OAuth app of your own, registered in its developer portal. actionlake has no Google BigQuery app to lend you.

    • access_tokenAccess token from one sign-in with your own OAuth app
    • refresh_tokenRefresh token from the same sign-in; the access token is renewed from it
    • client_idGoogle OAuth client id (Google Cloud console > APIs & Services > Credentials; enable the BigQuery API, and the Data…
    • client_secretGoogle OAuth client secret
    account { op: "connections.requirements", service: "google_bigquery" }
# free: exactly what to paste, and what is found for you
  3. 03

    Make the first call. Check before a write.

    Ask in plain words; Claude picks the action. Before anything changes, check validates the call and prices it, for free.

    do    { action: "google_bigquery.projects.list" }# projects: list · read · 1 upstream callcheck { action: "google_bigquery.datasets.insert", input: { … } }# free: inputs, permissions, cost and confirmations, nothing sent

Google BigQuery at a glance

Sign-in
OAuth 2.0, authorization code with PKCE. You register your own OAuth app with Google BigQuery, an admin signs in once, and you paste the app's client id and secret and the tokens from that sign-in; actionlake renews the access token from then on.
You provide
access_tokenrefresh_tokenclient_idclient_secret
Limits we pace to
100 calls a second, per connection
API covered
BigQuery API v2 (Discovery revision 20260922); BigQuery Data Transfer API v1 (Discovery revision 20261017); BigQuery Reservation API v1 (Discovery revision 20261017); BigQuery Connection API v1 (Discovery revision 20260922); BigQuery Data Policy API v1 (Discovery revision 20261014)bigquery.googleapis.com
Built from
Google BigQuery's documentation, read 22 October 2026googleapis.comgoogleapis.comgoogleapis.com
Service file
Version 1.0.0, signed (Ed25519)
Cached reads
65 of 66 reads, for up to 5 minutes; a write clears what it affects
Shared entities
2, so ask can line Google BigQuery up with your other systems
Data
Processed and stored in Sydney, Australia. No US region yet

Capabilities

What an AI agent can do in Google BigQuery.

134 actions in 21 areas. Every one carries its effect, so a key can allow the reads and hold the rest.

  • 66readlook, never change
  • 44writecreate and update
  • 24destructivewait for a confirmation

  1. Transfer configs

    14

    4 read · 2 write · 8 destructive

    Transfer configs: create · Transfer configs: create in location · Transfer configs: list · Transfer configs: list in location

  2. Assignments

    10

    5 read · 4 write · 1 destructive

    Assignments: create · Assignments: list · Assignments: search · Assignments: search all

  3. Data policies

    9

    4 read · 4 write · 1 destructive

    Data policies: create · Data policies: list · Data policies: test iam permissions · Data policies: get

  4. Data sources

    9

    6 read · 3 write · 0 destructive

    Data sources: check valid creds · Data sources: check valid creds in location · Data sources: list · Data sources: list in location

  5. Reservations

    9

    4 read · 4 write · 1 destructive

    Reservations: create · Reservations: list · Reservations: test iam permissions · Reservations: failover reservation

  6. Tables

    9

    4 read · 4 write · 1 destructive

    Tables: list · Tables: insert · Tables: get · Tables: patch

  7. Connections

    8

    4 read · 3 write · 1 destructive

    Connections: create · Connections: list · Connections: test iam permissions · Connections: get

  8. Routines

    8

    4 read · 3 write · 1 destructive

    Routines: insert · Routines: list · Routines: test iam permissions · Routines: get

13 more areas · 58 actions
  • Row access policies 8
  • Capacity commitments 7
  • Datasets 7
  • Jobs 7
  • Transfer runs 6
  • Reservation groups 5
  • Models 4
  • Transfer resources 4
  • Bi reservation 2
  • Projects 2
  • Tabledata 2
  • Transfer locations 2
  • Transfer logs 2

Top actions

The Google BigQuery API, shaped for agents.

The ids your agent types, what each does, and the one name it shares with other systems. 15 of Google BigQuery's 15 action names are shared with at least one other connector, so an agent learns them once.

ActionWhat it doesEffectSame name in
google_bigquery.datasets.listDatasets: listLists all datasets in the specified project to which the user has been granted the READER dataset role.readanalytics.dataset.list28 connectors
google_bigquery.tables.listTables: listLists all tables in the specified dataset.readcontent.table.list21 connectors
google_bigquery.datasets.insertDatasets: insertCreates a new empty dataset.writeanalytics.dataset.create21 connectors
google_bigquery.datasets.deleteDatasets: deleteDeletes the dataset specified by the datasetId value.destructiveanalytics.dataset.delete23 connectors
google_bigquery.tables.insertTables: insertCreates a new, empty table in the dataset.writecontent.table.create16 connectors
google_bigquery.jobs.get_query_resultsJobs: get query resultsRPC to get the results of a query job.readanalytics.query_result.get6 connectors
google_bigquery.jobs.listJobs: listLists all jobs that you started in the specified project.readinfra.job.list12 connectors
google_bigquery.jobs.insertJobs: insertStarts a new asynchronous job.writeinfra.job.create10 connectors
google_bigquery.projects.get_service_accountProjects: get service accountRPC to get the service account for a project used for interactions with Google Cloud KMS.readinfra.service_account.get4 connectors
google_bigquery.jobs.queryJobs: queryRuns a BigQuery SQL query synchronously and returns query results if the query completes within a specified timeout.writethis connector
google_bigquery.models.listModels: listLists all models in the specified dataset.readthis connector
google_bigquery.projects.listProjects: listRPC to list projects to which the user has been granted any project role.readthis connector

Try it

Prompts to paste into Claude.

Each one names the Google BigQuery actions it will call, so you know what it reads and what it changes before you send it.

  1. Readread

    Projects: list in Google BigQuery and tell me what stands out.

    callsgoogle_bigquery.projects.list

  2. Writewrite

    Datasets: insert in Google BigQuery. Show me exactly what you will send before you save it.

    callsgoogle_bigquery.datasets.insert

Governance for Google BigQuery

Destructive actions wait for a yes.

24 Google BigQuery actions are destructive. These are all of them:

  • Datasets: deletegoogle_bigquery.datasets.delete
  • Tables: deletegoogle_bigquery.tables.delete
  • Jobs: deletegoogle_bigquery.jobs.delete
  • Models: deletegoogle_bigquery.models.delete
  • Routines: deletegoogle_bigquery.routines.delete
  • Jobs: cancelgoogle_bigquery.jobs.cancel
  • Transfer configs: creategoogle_bigquery.transfer_configs.create
  • Transfer configs: schedule runsgoogle_bigquery.transfer_configs.schedule_runs
  • Transfer configs: start manual runsgoogle_bigquery.transfer_configs.start_manual_runs
  • Transfer configs: create in locationgoogle_bigquery.transfer_configs.create_in_location
  • Transfer configs: schedule runs in locationgoogle_bigquery.transfer_configs.schedule_runs_in_location
  • Transfer configs: start manual runs in locationgoogle_bigquery.transfer_configs.start_manual_runs_in_location
  • Assignments: deletegoogle_bigquery.assignments.delete
  • Connections: deletegoogle_bigquery.connections.delete
  • Reservations: deletegoogle_bigquery.reservations.delete
  • Data policies: deletegoogle_bigquery.data_policies.delete
  • Transfer runs: deletegoogle_bigquery.transfer_runs.delete
  • Transfer configs: deletegoogle_bigquery.transfer_configs.delete
  • Reservation groups: deletegoogle_bigquery.reservation_groups.delete
  • Row access policies: deletegoogle_bigquery.row_access_policies.delete
  • Capacity commitments: deletegoogle_bigquery.capacity_commitments.delete
  • Row access policies: batch deletegoogle_bigquery.row_access_policies.batch_delete
  • Transfer runs: delete in locationgoogle_bigquery.transfer_runs.delete_in_location
  • Transfer configs: delete in locationgoogle_bigquery.transfer_configs.delete_in_location

A destructive action runs only with confirm: true. Under the human policy the agent cannot confirm for itself: the run stops until a named person approves, in the console or from a signed link in an email. The cap on a key is refused before the call is sent, never after.

dogoogle_bigquery.datasets.deleteconfirm_policy: human

Datasets: delete?

intent: the user asked me to datasets: delete

run suspended · waiting for a named person · key google-bigquery-agent cannot confirm itself
In the console, or from an emailed signed link.
a key for this connector
account {  op: "keys.create", name: "google-bigquery-agent",  allowed_actions: ["google_bigquery.*"], cap_calls: 2000,  confirm_policy: "human_for_writes",  approval_rules: [    { match: "google_bigquery.datasets.delete", approver: "[email protected]" },    { match: "google_bigquery.tables.delete", approver: "[email protected]" },    { match: "google_bigquery.jobs.delete", approver: "[email protected]" }  ]}
Illustrative. Scoped to Google BigQuery only, capped at 2,000 upstream calls, every write confirmed, and the destructive ones approved by a named person.

Cost

What Google BigQuery costs on actionlake.

Every Google BigQuery action is one upstream call, the only thing actionlake bills. Searching, check, cost, account and any call Google BigQuery refused with 401, 403, 429 or a 5xx are free. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.

  • 1upstream call per actionall 134 Google BigQuery actions
  • 28actions pageeach page of a long list is one call
  • 65reads can be cacheda cache hit costs nothing
  • 100,000calls a month, freepaid plans from US$49 a month for 250,000

See pricing

Action index

Every Google BigQuery action.

134 actions by area: the id your agent calls, what it does, and its effect.

Transfer configs14
  • google_bigquery.transfer_configs.createTransfer configs: createdestructive
  • google_bigquery.transfer_configs.create_in_locationTransfer configs: create in locationdestructive
  • google_bigquery.transfer_configs.deleteTransfer configs: deletedestructive
  • google_bigquery.transfer_configs.delete_in_locationTransfer configs: delete in locationdestructive
  • google_bigquery.transfer_configs.getTransfer configs: getread
  • google_bigquery.transfer_configs.get_in_locationTransfer configs: get in locationread
  • google_bigquery.transfer_configs.listTransfer configs: listread
  • google_bigquery.transfer_configs.list_in_locationTransfer configs: list in locationread
  • google_bigquery.transfer_configs.patchTransfer configs: patchwrite
  • google_bigquery.transfer_configs.patch_in_locationTransfer configs: patch in locationwrite
  • google_bigquery.transfer_configs.schedule_runsTransfer configs: schedule runsdestructive
  • google_bigquery.transfer_configs.schedule_runs_in_locationTransfer configs: schedule runs in locationdestructive
  • google_bigquery.transfer_configs.start_manual_runsTransfer configs: start manual runsdestructive
  • google_bigquery.transfer_configs.start_manual_runs_in_locationTransfer configs: start manual runs in locationdestructive
Assignments10
  • google_bigquery.assignments.createAssignments: createwrite
  • google_bigquery.assignments.deleteAssignments: deletedestructive
  • google_bigquery.assignments.get_iam_policyAssignments: get iam policyread
  • google_bigquery.assignments.listAssignments: listread
  • google_bigquery.assignments.moveAssignments: movewrite
  • google_bigquery.assignments.patchAssignments: patchwrite
  • google_bigquery.assignments.searchAssignments: searchread
  • google_bigquery.assignments.search_allAssignments: search allread
  • google_bigquery.assignments.set_iam_policyAssignments: set iam policywrite
  • google_bigquery.assignments.test_iam_permissionsAssignments: test iam permissionsread
Data policies9
  • google_bigquery.data_policies.createData policies: createwrite
  • google_bigquery.data_policies.deleteData policies: deletedestructive
  • google_bigquery.data_policies.getData policies: getread
  • google_bigquery.data_policies.get_iam_policyData policies: get iam policyread
  • google_bigquery.data_policies.listData policies: listread
  • google_bigquery.data_policies.patchData policies: patchwrite
  • google_bigquery.data_policies.renameData policies: renamewrite
  • google_bigquery.data_policies.set_iam_policyData policies: set iam policywrite
  • google_bigquery.data_policies.test_iam_permissionsData policies: test iam permissionsread
Data sources9
  • google_bigquery.data_sources.check_valid_credsData sources: check valid credsread
  • google_bigquery.data_sources.check_valid_creds_in_locationData sources: check valid creds in locationread
  • google_bigquery.data_sources.enrollData sources: enrollwrite
  • google_bigquery.data_sources.enroll_in_locationData sources: enroll in locationwrite
  • google_bigquery.data_sources.getData sources: getread
  • google_bigquery.data_sources.get_in_locationData sources: get in locationread
  • google_bigquery.data_sources.listData sources: listread
  • google_bigquery.data_sources.list_in_locationData sources: list in locationread
  • google_bigquery.data_sources.unenroll_in_locationData sources: unenroll in locationwrite
Reservations9
  • google_bigquery.reservations.createReservations: createwrite
  • google_bigquery.reservations.deleteReservations: deletedestructive
  • google_bigquery.reservations.failover_reservationReservations: failover reservationwrite
  • google_bigquery.reservations.getReservations: getread
  • google_bigquery.reservations.get_iam_policyReservations: get iam policyread
  • google_bigquery.reservations.listReservations: listread
  • google_bigquery.reservations.patchReservations: patchwrite
  • google_bigquery.reservations.set_iam_policyReservations: set iam policywrite
  • google_bigquery.reservations.test_iam_permissionsReservations: test iam permissionsread
Tables9
  • google_bigquery.tables.deleteTables: deletedestructive
  • google_bigquery.tables.getTables: getread
  • google_bigquery.tables.get_iam_policyTables: get iam policyread
  • google_bigquery.tables.insertTables: insertwrite
  • google_bigquery.tables.listTables: listread
  • google_bigquery.tables.patchTables: patchwrite
  • google_bigquery.tables.set_iam_policyTables: set iam policywrite
  • google_bigquery.tables.test_iam_permissionsTables: test iam permissionsread
  • google_bigquery.tables.updateTables: updatewrite
Connections8
  • google_bigquery.connections.createConnections: createwrite
  • google_bigquery.connections.deleteConnections: deletedestructive
  • google_bigquery.connections.getConnections: getread
  • google_bigquery.connections.get_iam_policyConnections: get iam policyread
  • google_bigquery.connections.listConnections: listread
  • google_bigquery.connections.patchConnections: patchwrite
  • google_bigquery.connections.set_iam_policyConnections: set iam policywrite
  • google_bigquery.connections.test_iam_permissionsConnections: test iam permissionsread
Routines8
  • google_bigquery.routines.deleteRoutines: deletedestructive
  • google_bigquery.routines.getRoutines: getread
  • google_bigquery.routines.get_iam_policyRoutines: get iam policyread
  • google_bigquery.routines.insertRoutines: insertwrite
  • google_bigquery.routines.listRoutines: listread
  • google_bigquery.routines.set_iam_policyRoutines: set iam policywrite
  • google_bigquery.routines.test_iam_permissionsRoutines: test iam permissionsread
  • google_bigquery.routines.updateRoutines: updatewrite
Row access policies8
  • google_bigquery.row_access_policies.batch_deleteRow access policies: batch deletedestructive
  • google_bigquery.row_access_policies.deleteRow access policies: deletedestructive
  • google_bigquery.row_access_policies.getRow access policies: getread
  • google_bigquery.row_access_policies.get_iam_policyRow access policies: get iam policyread
  • google_bigquery.row_access_policies.insertRow access policies: insertwrite
  • google_bigquery.row_access_policies.listRow access policies: listread
  • google_bigquery.row_access_policies.test_iam_permissionsRow access policies: test iam permissionsread
  • google_bigquery.row_access_policies.updateRow access policies: updatewrite
Capacity commitments7
  • google_bigquery.capacity_commitments.createCapacity commitments: createwrite
  • google_bigquery.capacity_commitments.deleteCapacity commitments: deletedestructive
  • google_bigquery.capacity_commitments.getCapacity commitments: getread
  • google_bigquery.capacity_commitments.listCapacity commitments: listread
  • google_bigquery.capacity_commitments.mergeCapacity commitments: mergewrite
  • google_bigquery.capacity_commitments.patchCapacity commitments: patchwrite
  • google_bigquery.capacity_commitments.splitCapacity commitments: splitwrite
Datasets7
  • google_bigquery.datasets.deleteDatasets: deletedestructive
  • google_bigquery.datasets.getDatasets: getread
  • google_bigquery.datasets.insertDatasets: insertwrite
  • google_bigquery.datasets.listDatasets: listread
  • google_bigquery.datasets.patchDatasets: patchwrite
  • google_bigquery.datasets.undeleteDatasets: undeletewrite
  • google_bigquery.datasets.updateDatasets: updatewrite
Jobs7
  • google_bigquery.jobs.cancelJobs: canceldestructive
  • google_bigquery.jobs.deleteJobs: deletedestructive
  • google_bigquery.jobs.getJobs: getread
  • google_bigquery.jobs.get_query_resultsJobs: get query resultsread
  • google_bigquery.jobs.insertJobs: insertwrite
  • google_bigquery.jobs.listJobs: listread
  • google_bigquery.jobs.queryJobs: querywrite
Transfer runs6
  • google_bigquery.transfer_runs.deleteTransfer runs: deletedestructive
  • google_bigquery.transfer_runs.delete_in_locationTransfer runs: delete in locationdestructive
  • google_bigquery.transfer_runs.getTransfer runs: getread
  • google_bigquery.transfer_runs.get_in_locationTransfer runs: get in locationread
  • google_bigquery.transfer_runs.listTransfer runs: listread
  • google_bigquery.transfer_runs.list_in_locationTransfer runs: list in locationread
Reservation groups5
  • google_bigquery.reservation_groups.createReservation groups: createwrite
  • google_bigquery.reservation_groups.deleteReservation groups: deletedestructive
  • google_bigquery.reservation_groups.getReservation groups: getread
  • google_bigquery.reservation_groups.listReservation groups: listread
  • google_bigquery.reservation_groups.patchReservation groups: patchwrite
Models4
  • google_bigquery.models.deleteModels: deletedestructive
  • google_bigquery.models.getModels: getread
  • google_bigquery.models.listModels: listread
  • google_bigquery.models.patchModels: patchwrite
Transfer resources4
  • google_bigquery.transfer_resources.getTransfer resources: getread
  • google_bigquery.transfer_resources.get_in_locationTransfer resources: get in locationread
  • google_bigquery.transfer_resources.listTransfer resources: listread
  • google_bigquery.transfer_resources.list_in_locationTransfer resources: list in locationread
Bi reservation2
  • google_bigquery.bi_reservation.getBi reservation: getread
  • google_bigquery.bi_reservation.updateBi reservation: updatewrite
Projects2
  • google_bigquery.projects.get_service_accountProjects: get service accountread
  • google_bigquery.projects.listProjects: listread
Tabledata2
  • google_bigquery.tabledata.insert_allTabledata: insert allwrite
  • google_bigquery.tabledata.listTabledata: listread
Transfer locations2
  • google_bigquery.transfer_locations.getTransfer locations: getread
  • google_bigquery.transfer_locations.listTransfer locations: listread
Transfer logs2
  • google_bigquery.transfer_logs.listTransfer logs: listread
  • google_bigquery.transfer_logs.list_in_locationTransfer logs: list in locationread

Questions

The Google BigQuery MCP server, answered.

Is there a Google BigQuery MCP server for Claude?

Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 134 Google BigQuery actions across transfer configs, assignments, data policies and data sources, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by Google BigQuery.

What do I need to connect Google BigQuery?

Your own Google BigQuery OAuth app, registered with Google BigQuery (its client id and client secret), and the access and refresh tokens from one admin sign-in. actionlake has no Google BigQuery app of its own, so this is not a one-step consent. actionlake renews the access token as it expires. The free call account { op: "connections.requirements", service: "google_bigquery" } lists exactly what to paste.

Can an AI agent delete or change things in Google BigQuery?

Only if its key allows it, and only with an explicit confirmation. 24 Google BigQuery actions are destructive, including datasets: delete, tables: delete and jobs: delete. Each runs only with confirm: true. Under the human confirm policy, or an approval rule on the key, the run waits until a named person approves it.

Will an AI agent use up my Google BigQuery API limit?

actionlake paces every call to the limits in the Google BigQuery service file: 100 calls a second per connection. Repeated reads are answered from cache (65 Google BigQuery actions can be), which never reaches Google BigQuery.

What does a Google BigQuery action cost?

One upstream call, the only unit actionlake bills. A list that pages bills one call per page (28 Google BigQuery actions page). Cache hits, searches, dry runs and calls Google BigQuery refused with 401, 403, 429 or a 5xx are free. The Free plan includes 100,000 calls a month with no card. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.

Which Google BigQuery API does it cover?

The BigQuery API v2 (Discovery revision 20260922); BigQuery Data Transfer API v1 (Discovery revision 20261017); BigQuery Reservation API v1 (Discovery revision 20261017); BigQuery Connection API v1 (Discovery revision 20260922); BigQuery Data Policy API v1 (Discovery revision 20261014) at bigquery.googleapis.com. It is built from Google BigQuery's documentation as read on 22 October 2026. 134 actions in 21 areas, each one in the signed service file (version 1.0.0).

Does actionlake keep my Google BigQuery data, and where?

actionlake processes and stores data in Sydney, Australia; there is no US region yet. Reads may be cached for up to 5 minutes, and a write clears the reads it affects. Each run's request and response is kept, sealed, for your plan's trace window: 1 day on Free, 3 days on Starter, 7 days on Pro, 14 days on Team, 30 days on Business and 30 days on Enterprise. Your Google BigQuery credential is sealed under your tenant's key, never logged and never returned. Your traffic is not used to train models.

Which AI clients can use it?

Claude Code, tested by us end to end, and the Claude app as a custom connector, tested by us with an existing key. Other MCP clients that support remote servers over Streamable HTTP (MCP revisions 2025-06-18 and 2025-11-25) and OAuth dynamic client registration, or a bearer header, should connect the same way; we have not tested them. Or plain REST, under the same action names.

More connectors like Google BigQuery

Shares the most action names with

Joins with, on shared entities

Browse

actionlake is independent. It is not made or endorsed by Google BigQuery; the name is the vendor's own. Counts are from the Google BigQuery service file, version 1.0.0.

Give your agent Google BigQuery today.

Connect Google BigQuery freeSee pricing

Start free - 100,000 calls/month, no card. No signup. Hard caps on every key. Cached answers are free.