# Your warehouse. Answers in plain words.

Source: https://corespeed.io/integrations/bigquery

Connect BigQuery to Claude Code, Codex, Cursor or any MCP agent through CoreSpeed. Your agent lists your datasets and tables, reads a table's schema and sample rows, and runs the SQL that answers your question. It signs in with your own Google account and sees only the projects that account can. Running a query is a separate tool call.

## Set up an agent

Give your agent this line: `set up https://corespeed.io/SKILL.md (read it raw, e.g. with curl)`

It fetches the skill at https://corespeed.io/SKILL.md, adds the CoreSpeed MCP server, signs you in (or up), and keeps the skill for next time. Connect BigQuery after setup. Access follows the connected account and its permissions. Create an account: https://app.corespeed.io/sign-up

## Ask the question. Get the number and the SQL

Your agent finds the table, checks its schema, and runs one query. You get the answer with the SQL behind it.

You ask your agent: "Find our daily orders table in BigQuery. Check its columns, then show revenue by region for the last full week, with the SQL you ran."

It calls CoreSpeed:

- `bigquery__list_tables`: ok, 388 ms
- `bigquery__get_table`: ok, 251 ms
- `bigquery__query`: ok, 1840 ms

It reads from: BigQuery.

### Weekly revenue: Last week, by region.

Revenue rose in two regions and fell in one. The SQL is under the numbers.

**By region**

- West: $182,400, up 6%.
- East: $151,900, up 2%.
- Central: $98,300, down 4%.

**Query**

- SUM(revenue) from analytics.orders_daily by region, Nov 2 to Nov 8.

_Illustrative example · Fictional data · One SELECT, no data changed_

All connectors: https://corespeed.io/docs/connectors

## From a question to the right query

Requests you can make once BigQuery is connected. The data is fictional.

### Find the right table

analytics · tables · Datasets and tables

- Dataset: analytics
- Best match: orders_daily
- From the projects you can access

### Read before you query

orders_daily · schema · Columns and types

- Columns: day, region, orders, revenue
- Partition: day
- No query run yet

### Hold a heavy query

A query waits for a person · Approvals

- Table: analytics.events_raw
- Policy: Ask before querying events_raw
- Ask

Full example: https://corespeed.io/solutions/agent-approvals (Markdown: https://corespeed.io/solutions/agent-approvals.md)

## Keep your agent. Bring your work

One CoreSpeed account, across compatible agents.

1. **Set up CoreSpeed**: Give your agent the setup line. It adds CoreSpeed and signs you in.
2. **Connect BigQuery**: Authorize the account and the content you need.
3. **Start with one task**: Ask for a summary, a brief, or a draft. Review the answer.

Setup guides: Claude Code https://corespeed.io/docs/mcp#claude-code · Codex https://corespeed.io/docs/mcp#codex · Cursor https://corespeed.io/docs/mcp#cursor

## Questions

### Can the agent change my BigQuery data?

Only if your Google account can. Inserting rows or deleting a table are separate tool calls, and your approval policy can hold them for a person.

### Who pays for the queries?

BigQuery bills each query to your own Google Cloud project, as for any client. The catalog page lists what CoreSpeed charges per call.

### Do I have to switch agents?

No. CoreSpeed adds tools to the agent you already use: Claude Code, Codex, Cursor, or another MCP client. Each one signs in once.

### Can my agent access every account in our organization?

No. Access follows your CoreSpeed identity, each connection's visibility, and the app's own permissions. Personal connections stay private.

### Does this work without connecting an account?

No. Connect the app's account first; your CoreSpeed workspace shows what each app needs. The examples on this page use fictional data. All connectors: https://corespeed.io/docs/connectors

Every tool and price: https://corespeed.io/catalog/bigquery

## Related pages

- [Approval gates for AI agent tool actions](https://corespeed.io/solutions/agent-approvals): Set approval policies for writes through CoreSpeed. Review held tool actions while your team keeps using its preferred agents.
- [PostHog for Claude Code, Codex, and Cursor](https://corespeed.io/integrations/posthog): Read PostHog insights and dashboards from your agent. Get the numbers behind a question, and mark releases on your charts.
- [Snowflake for Claude Code, Codex, and Cursor](https://corespeed.io/integrations/snowflake): Query Snowflake from your agent under your own token and role. It finds the table, reads the columns, and runs the SQL you asked for.

## Sources

- Connectors: https://corespeed.io/docs/connectors
- BigQuery REST API reference: https://cloud.google.com/bigquery/docs/reference/rest
- Run a query: https://cloud.google.com/bigquery/docs/running-queries

Reviewed 2026-11-10.

## Links

- Product overview: https://corespeed.io/ (Markdown: https://corespeed.io/index.md)
- All integrations: https://corespeed.io/integrations (Markdown: https://corespeed.io/integrations.md)
- Documentation: https://corespeed.io/docs
- Pricing: https://corespeed.io/pricing (Markdown: https://corespeed.io/pricing.md)