Wire Airtable and BigQuery into OpenCode and it can manage your warehouse and keep your base in step with it. BigQuery is the analytical store, so the agent creates datasets, tables, and routines, and cancels or cleans up jobs. Airtable is the lightweight system of record your team works in, so the same run documents each change as records, comments, and tables. Provisioning in the warehouse and tracking it in the base happen together instead of drifting apart.
Pick the operations you want and each becomes a tool the agent can call directly while it works.
List all accessible bases
Get the schema (tables and fields) of a base
Create a new table in a base
Update a table's name or description
Create a new field in a table
Update a field's name or description
List records from a table with filtering, sorting, and pagination
Get a single record by ID
Create one or more records (batch create up to 10)
Update one or more records (batch update up to 10)
Update a single record by ID
Delete one or more records (batch delete up to 10)
Delete a single record by ID
Create or update records based on field matching (upsert)
List comments on a record
Create a comment on a record
Update a comment
Delete a comment
List all webhooks for a base
Create a webhook for a base
Delete a webhook
List payloads for a webhook
Refresh a webhook to extend its expiration
Enable notifications for a webhook
Run a SQL query and return results inline (synchronous, bounded by timeout).
Page through the results of a (possibly async) query job.
Start a query / load / extract / copy job (async).
Fetch a job's status and statistics (poll until DONE).
List jobs in a project (filter by state, time range).
Request cancellation of a running job.
Delete a job's metadata.
List datasets in the project.
Fetch dataset metadata.
Create a new dataset.
Full replace of dataset metadata (PUT).
Partial update of dataset metadata (PATCH).
Delete a dataset.
List tables/views in a dataset.
Fetch table metadata/schema.
Create a table or view with a schema.
Partial update of a table (add columns, set expiration, etc.).
Full replace of table metadata (PUT).
Delete a table or view.
Streaming insert of JSON rows into a table (real-time append).
Read rows from a table directly (paginated, no SQL).
List stored procedures / UDFs / TVFs in a dataset.
Fetch a routine definition.
Create a stored procedure / UDF / TVF.
List BigQuery ML models in a dataset.
Fetch BQML model metadata.
Delete a BQML model.
Return the BigQuery service account for the project (KMS/transfer grants).
List projects the caller can access (for project picker UIs).
Restore a dataset within its time-travel window (datasets.undelete).
Replace a routine's full definition (routines.update, PUT).
Delete a routine (routines.delete).
Update mutable BQML model metadata (models.patch).
Get the IAM policy for a table (tables.getIamPolicy).
Set the IAM policy for a table (tables.setIamPolicy).
Test the caller's permissions on a table (tables.testIamPermissions).
List row access policies on a table (rowAccessPolicies.list).
Get a single row access policy (rowAccessPolicies.get).
OpenCode is an open, model-agnostic command-line coding agent. In NoClick it runs hosted and gains your apps as tools, so it can act on real systems rather than just edit files. Wire an integration into the agent and OpenCode can use that app’s operations directly while it works. It is a flexible choice when you want an open agent with broad model support and real tool access.
NoClick runs OpenCode for you with Airtable and Google BigQuery wired in as tools. Connect your account and run.