BigQuery
Run BigQuery SQL, stream rows in, and manage datasets, tables and jobs.
About
BigQuery ships in the w6w first-party pack. It declares 18 actions, 2 health checks, and the host runs its code in a sandbox that never sees the credential.
- App id
io.w6w.bigquery- Version
- 0.1.1
- Author
- w6w
- Licence
- MIT
- Categories
- Data Warehousing & ETL · Databases
Overview
Run SQL against BigQuery, page through the results, and stream rows straight into a table without going through a query at all. Jobs can be started for query, load, extract or copy work, and tracked, cancelled or listed by state — with the label to watch being “done,” which BigQuery uses to mean finished, not succeeded.
The app absorbs a few of BigQuery’s rougher edges rather than passing them on: query results come back as positional cells rather than named fields, so rows are decoded into a plain array alongside the raw response; a dry run prices a query without executing it; and streaming inserts, which BigQuery can accept with a success response while silently rejecting individual rows, are checked row by row and can carry a stable per-row id so a retry doesn’t duplicate data.
Datasets and tables are manageable end to end — create, list, update schema, delete — alongside the list of projects a connection can query in. Good for workflows that run reporting or ELT jobs against BigQuery, stream events into a table for later analysis, or need job status back before moving on to the next step.
Build with BigQuery
Three routes to the same 18 actions. The Workflow tab is generated from BigQuery's own manifest and carries its real ids, so it is copy-pasteable; the Code and CLI examples are the same call for any action on any app, so every app-specific value in them is a blank you fill in.
A workflow step names the app and the action, and the editor fills in the
connection when you pick one. This is the Step shape from the
workflow spec, carrying BigQuery's real ids.
{
"manifestVersion": "2",
"name": "bigquery-example",
"steps": [
{
"id": "dataset-create",
"uses": {
"app": "io.w6w.bigquery",
"action": "dataset-create",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"datasetId": "<datasetId>"
}
}
]
}Here are some of the things you can do
-
Create a dataset
performdataset-create -
Get a dataset
readdataset-get -
List datasets
readdataset-list -
Get a job
readjob-get -
Start a job
performjob-insert
+13 more actions available
Every app-specific value here is a blank you have to fill in. An
app action is reached through the connection that authenticates it, so the
address is a connection id, not the app id — and connections belong to your account,
so a public page cannot know yours. Create one for BigQuery, then fill in
the three blanks: conn_YOUR_CONNECTION_ID, the action key, and the
parameters that action declares. The call itself is real — the shape is transcribed
from the studio's own snippet builder, which prints the same kind of blanks — but
nothing in it is specific to BigQuery. The Workflow tab is where this app's
real ids are.
npm install @w6w/sdkyarn add @w6w/sdkpnpm add @w6w/sdkdeno add npm:@w6w/sdkimport { W6wClient, isActionRun } from "@w6w/sdk";
// Reads W6W_BASE_URL and W6W_TOKEN from the environment when omitted.
const client = new W6wClient();
const envelope = await client.run({
urn: "conn_YOUR_CONNECTION_ID",
action: "dataset-create",
payload: {
// projectId: "<value>",
datasetId: "<value>",
// location: "<value>",
// friendlyName: "<value>",
// description: "<value>",
// defaultTableExpirationMs: "<value>",
// labels: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value);npm install -g @w6w/cliw6w run conn_YOUR_CONNECTION_ID --action dataset-create --payload '{"datasetId":"<value>"}'Give an AI agent BigQuery — without giving it BigQuery's credentials. One MCP endpoint exposes every app, function and workflow the caller is entitled to, as tools it can discover and run. Access is granted per team while we onboard.
One tool call{
"name": "w6w_invoke",
"arguments": {
"ref": "app:io.w6w.bigquery#dataset-create",
"input": {
"datasetId": "<datasetId>"
}
}
}
Every tool names its target with a single ref. The
app: form above doesn't name a connection at all — the
host resolves which of the caller's BigQuery connections to sign
with, and refuses rather than guesses when the answer is ambiguous.
What the agent gets
Credentials it can't read
The token is attached host-side, at the moment of the call. It is never a tool argument, never in the model's context, and never in a transcript — so a prompt injection has nothing to exfiltrate.
A tool surface scoped to the caller
Tools are derived per end user from what that person has actually connected and is entitled to — not one shared bot identity carrying the union of everyone's access.
A durable workflow in one call
Multi-step work runs on the workflow engine and returns a run handle the agent can poll — retries, branching and state survive the conversation that started them.
Health-aware discovery
BigQuery's declared health checks are on the surface too, so an agent can tell "the vendor is down" from "your credential expired" before it burns a retry on either.
The MCP surface is part of the hosted platform. BigQuery itself is MIT, and the runtime that executes it is source-available (FSL).
Health checks
BigQuery declares its own checks, so its health is a property of the app rather than something the host guesses at.
BigQuery platform status
Open incidents for Google BigQuery on the Google Cloud status dashboard — not the Workspace or Ads one. Unauthenticated and unsigned.