LeashLeash
MCP
data

BigQuery

Run read-only SQL against Google BigQuery datasets from inside your Leash app.

Requires Growth plan or above.

Connect BigQuery

Create a service account JSON key in Google Cloud Console → IAM → Service Accounts (grant BigQuery Data Viewer + BigQuery Job User), then paste the JSON on the BigQuery card at /dashboard/connections.

Example

Run a SQL query against a BigQuery public dataset and return rows.

app/api/top-names/route.ts
export async function GET() {
  const res = await fetch('https://leash.build/api/integrations/bigquery/query', {
    method: 'POST',
    headers: {
      'x-api-key': process.env.LEASH_API_KEY!,
      'content-type': 'application/json',
    },
    body: JSON.stringify({
      sql: `
        SELECT name, SUM(number) AS total
        FROM \`bigquery-public-data.usa_names.usa_1910_2013\`
        WHERE state = 'CA' AND year >= 2000
        GROUP BY name
        ORDER BY total DESC
        LIMIT 5
      `,
    }),
  })
  const { data } = await res.json()


  // data =>
  // {
  //   schema: [
  //     { name: 'name',  type: 'STRING'  },
  //     { name: 'total', type: 'INTEGER' }
  //   ],
  //   rows: [
  //     { name: 'Daniel',   total: 42117 },
  //     { name: 'Anthony',  total: 41008 },
  //     ...
  //   ],
  //   totalRows: 5,
  //   totalBytesProcessed: 7340032
  // }


  return Response.json(data)
}

Available tools

  • query — run a read-only SQL statement and return rows.

Common gotchas

  • Read-only by design. INSERT, UPDATE, CREATE TABLE are rejected — service account scopes prevent writes.
  • Fully-qualify tables. Always use `project.dataset.table`; default dataset is not set.
  • You pay per byte scanned. Filter on partition columns (_PARTITIONDATE) and avoid SELECT * on large tables.

See the BigQuery SQL reference for the underlying provider docs.

Other languages: see SDK overview.