SnowConvert AI CLI - Data validation¶

Tip

This page covers SnowConvert AI CLI (scai) commands for cloud data validation. For validation levels, architecture, and platform guidance using the Snowflake AIM Agent for Data Warehouses, see Data validation. For workflow and Worker field definitions, see Data validation configuration reference.

Use this page when you run data validation from the CLI instead of Snowflake CoCo. Cloud validation commands mirror migration under scai data validate …, sharing scai data worker … and scai data orchestrator ….

Supported source platforms: SQL Server, Amazon Redshift, Teradata, Oracle, and PostgreSQL. Platform-specific prerequisites (for example Teradata HASH_MD5 for L3), Worker TOML, and data type mappings live on the AIM pages:

Prerequisites¶

Before you run CLI commands, make sure the following are in place:

SCAI CLI installed and a SnowConvert AI project on disk.
Snowflake connection with a role that can create and administer SNOWCONVERT_AI.
Source connection on Workers (same TOML as data migration). Generate with scai data worker generate-config.
Target tables in Snowflake already loaded. Don’t alter migrated data between migration and validation.
Validation runtime on Workers that execute validation tasks (the CLI installs this when you use scai data validate start or --start-worker paths locally).

Install orchestrator and workers¶

Reuse the same Orchestrator and Workers as data migration when possible:

scai data worker generate-config .scai/settings/DataExchangeWorkerConfig.toml

# SPCS (first time):
scai data orchestrator setup --compute-pool MY_COMPUTE_POOL --connection my-snowflake
scai data worker setup --compute-pool MY_COMPUTE_POOL --connection my-snowflake

# Local Worker:
scai data worker start --local .scai/settings/DataExchangeWorkerConfig.toml

Worker TOML reference: Data validation configuration reference (same format as migration).

Run a validation workflow¶

Generate validation config¶

scai data validate generate-config
scai data validate generate-config --where "source.schema = 'public'"
scai data validate generate-config -o .scai/config/data-validation-config.yaml
scai data validate generate-config --affinity my-team
scai data validate generate-config --help

Default output: .scai/config/data-validation-config.yaml. The source_platform field must match your project dialect. Full workflow schema: Data validation configuration reference.

Submit and monitor workflows¶

# Local all-in-one (generates config if missing, starts local orchestrator + worker, watches):
scai data validate start --connection my-snowflake
scai data validate start --config .scai/config/data-validation-config.yaml --connection my-snowflake

# Submit only:
scai data validate create-workflow --config my-validation.yaml --connection my-snowflake
scai data validate create-workflow --config my-validation.yaml --connection my-snowflake --watch

# Status and listing:
scai data validate status DATA_VALIDATION_WORKFLOW_xx_yy_zz --watch
scai data validate list --status running --limit 20

Migrate data first with Data migration (CLI) if targets aren’t loaded yet.

Command reference¶

Command	Purpose
`scai data validate generate-config`	Generate validation workflow YAML from the project
`scai data validate generate-config -o <path>`	Write config to a custom path
`scai data validate start`	Local all-in-one validation run
`scai data validate create-workflow --config <path>`	Submit a validation workflow
`scai data validate status <WORKFLOW_NAME>`	Show or watch workflow status
`scai data validate list`	List validation workflows
`scai data validate pause <WORKFLOW_NAME>`	Pause a running workflow
`scai data validate resume <WORKFLOW_NAME>`	Resume a paused workflow
`scai data validate cancel <WORKFLOW_NAME>`	Cancel a workflow
`scai data worker generate-config <path>`	Generate shared Worker TOML
`scai data worker start --local <config.toml>`	Run a local Worker
`scai data orchestrator setup --compute-pool <POOL>`	Deploy Orchestrator on SPCS
`scai data orchestrator stop`	Suspend Orchestrator SPCS service
`scai connection test -l <dialect> -s <profile> --json`	Verify source connectivity

Monitor progress in SNOWCONVERT_AI.DATA_VALIDATION (TABLE_PROGRESS, TABLE_PROGRESS_DETAIL, DATA_VALIDATION_ERROR, DATA_VALIDATION_WARNING) and the DATA_VALIDATION_DASHBOARD Streamlit app. See Data validation for outcome categories and query tagging.