List every workspace the authenticated user has access to, including their permission on each.

Create a new workspace, owned by the authenticated user. Use this if list_my_workspaces returns none.

List either the data specs (parsing + mapping rule sets, resource "specs") or the data processing jobs (executions of a spec, resource "jobs") defined in a workspace. Paginated (default 25/page, max 100, newest first); pass the returned nextCursor to fetch more. Not a table listing — for the live set of Iceberg tables in a workspace, use submit_query with SHOW TABLES instead.

Poll the status of either a data spec's analysis (pass specId, reaches "ready"/"failed") or a data-load job (pass jobId, reaches "complete"/"failed"). Pass exactly one of specId or jobId.

Permanently delete a data spec and its associated configuration.

Run a SQL query against the Iceberg tables loaded into a workspace. Run SHOW TABLES to list the tables that actually exist. Reference tables by name only (no catalog/schema-qualified references).

Create, list, test, or delete a workspace connection to an external data source. For sftp, create generates a keypair and returns the public key to install in the remote server's authorized_keys. For aws_s3, create generates an ExternalId and trust policy for an IAM role the customer must create. A connection must pass test before it can be used in a trigger.

Create, list, update, delete, or fire a workspace job trigger. Types: "sftp"/"aws_s3" (pulls files from a connection into a spec on a schedule), "spec_success" (fires a spec when a different spec's job completes), "schedule" (fires a spec on a plain cadence, UTC). Use action "run-history" to monitor a fired trigger's outcome.

First step of setting up a new data integration: creates a data spec. sourceType "file" (default) returns presigned upload URL(s) for a sample file; upload, then call finish_data_source_onboarding. sourceType "tables" derives/aggregates data already loaded into workspace tables via an AI-generated SQL query instead, and requires autoRefresh (how the spec stays up to date: spec_success, schedule, or none).

Call after uploading the file(s) returned by onboard_data_source — kicks off AI analysis and waits until the spec reaches "ready" or "failed". Poll with get_status instead of re-calling this if it times out.

Change an existing data spec's configuration. Runs synchronously when no replacement files are given (saves changes, re-runs analysis by default). If a replacement file name is given, returns upload URL(s) instead; upload, then call finish_data_spec_update. Only pass fields you want to change.

Call after uploading the file(s) returned by update_data_spec — kicks off AI analysis and waits until the spec reaches "ready" or "failed". Poll with get_status instead of re-calling this if it times out.

First step of processing new data files through an already-configured data spec: creates a job and returns upload URL(s). Upload, then call finish_data_job. Only for files beyond the initial sample.

Call after uploading the file(s) returned by run_data_job — starts processing and waits until the job completes or fails. Poll with get_status instead of re-calling this if it times out.

End-to-end setup for "pull files from this SFTP server / S3 bucket on a schedule": reuses a matching connection if one exists, otherwise creates one; tests it; then creates a trigger feeding an already-analyzed spec on the given frequency. Pass hostname for sftp, or roleArn (+ s3Bucket) for aws_s3.

Escape hatch for DPF capabilities with no dedicated tool. Every DPF endpoint is POST with a JSON body of { action, ...fields }, authenticated with your OAuth session. Billing mutations (purchase-credits, modify-subscription, manage-payment, create-customer) are not available via MCP — use the DPF web UI.