Google Cloud · Cloud service
Google BigQuery with OrchestrAI
Catalog exported 2026-09-02
Query BigQuery and create datasets and tables from chat, with SQL results returned in the conversation.
OrchestrAI exposes 3 BigQuery operations: 1 is low-risk (read-only or low-impact), and 2 create or modify resources and run only after you confirm the plan.
What teams use it for
Data teams use OrchestrAI to ask a question in plain language, have it turned into SQL, and get the BigQuery result back without opening the query editor. Creating a dataset and its first tables for a new pipeline is the other frequent use, since the schema can be described in a sentence. Nothing in the toolset deletes a dataset or table and there is no load-job operation, so ingesting files and cleaning up old tables still goes through the console or the bq CLI.
Every BigQuery operation, with its risk level
| Operation | What it does | Risk | Step-level approval |
|---|---|---|---|
BigQuery Query |
Execute a SQL query on Google BigQuery | Low risk | No |
Create BigQuery Dataset |
Create a Google BigQuery dataset | Creates resources | No |
Create BigQuery Table |
Create a BigQuery table | Creates resources | No |
Risk tiers come from the catalog: low is read-only or low-impact, medium creates resources and is reversible, high modifies existing resources, destructive may lose data. Every plan that creates or changes resources is shown with its cost estimate and waits for your confirmation. Operations marked with a step-level approval pause again on their own step. Destructive operations require a typed risk phrase.
Prompts that work
- Create a BigQuery dataset named analytics_staging in the US multi-region
- Add a table called events to analytics_staging with columns event_id STRING, user_id STRING, and ts TIMESTAMP
- Run a query that counts rows per day in analytics_prod.events for the last 7 days
Before anything runs
Every mutation shows its plan, cost estimate, and blast radius, then waits for your confirmation. Destructive operations require a typed risk phrase. Credentials are minted per run through OIDC federation and discarded afterward; nothing you create here is invisible later, because every resource lands in the desired-state ledger where drift is detected and can be converged. Details on the security page.
Frequently asked questions
- Is running a BigQuery query through OrchestrAI safe?
- Query execution is classified low risk because it only reads data. Creating datasets and tables is medium risk because it adds new resources, and those are recorded in the ledger so drift can be detected later.
- Can OrchestrAI load data into BigQuery?
- Not yet, because the current operations are create dataset, create table, and execute SQL, so loading from Cloud Storage or streaming inserts must be set up separately.
- Which BigQuery operations need an extra approval step?
- None of the BigQuery operations currently carries a step-level approval gate; they are read-only or run after plan confirmation like any other change.
Other Google Cloud services
Related integrations
Try it on your own account
Connect your cloud read-only and see your resources, drift, and costs before anything runs. $5 minimum to start. Unused credits refunded in your first 14 days.
Unused credits refunded in your first 14 days.