Skip to main content
Feedback

Snowflake MCP connector

Snowflake is a cloud data platform for storing, querying, and analyzing large volumes of structured and semi-structured data. The Snowflake MCP connector allows AI agents to execute SQL statements, poll for execution results, and manage running queries. It also supports exploring account resources by listing and fetching details on databases, schemas, tables, and views.

Authentication type

  • OAuth 2.0 Authorization Code - Requires a one-time user login to authorize the agent to access the service on their behalf.

Uses

Use the Snowflake MCP connector to perform the following actions:

  • Run SQL queries against Snowflake databases directly from a workflow
  • Track long-running query execution status without manual polling
  • Discover databases, schemas, tables, and views across a Snowflake account
  • Inspect warehouse configurations to monitor compute resource allocation
  • Build and manage Apache Iceberg tables for open lakehouse architectures
  • Recover accidentally dropped Iceberg tables within the retention window
  • Convert externally cataloged Iceberg tables to Snowflake-managed catalogs
  • Tag Iceberg tables to support governance and data classification

Example prompts

Use the following example prompts to invoke Snowflake MCP connector tools from your AI assistant or Boomi Connect workflow:

  • Show me all open invoices from last month in Snowflake.
  • Run a query in Snowflake to find our top 10 customers by revenue.
  • List all the databases available in our Snowflake account.
  • Check if my Snowflake query has finished running yet.
  • Cancel that long-running Snowflake query I just started.
  • What tables exist in the sales schema in Snowflake?
  • Show me the structure of the customer table in Snowflake.
  • How much compute capacity does our Snowflake warehouse have?
  • Create a new Iceberg table in Snowflake for our shipment records.
  • Bring back the Iceberg table I accidentally deleted in Snowflake.

Snowflake MCP connector tools

The Snowflake MCP connector provides the following tools. Each tool maps to a specific action you can invoke from your AI agent or automation.

ToolDescription
SubmitStatementSubmits a SQL statement for execution.
GetStatementStatusChecks the status of the execution of a statement
CancelStatementCancels the execution of a statement.
listDatabasesList databases.
fetchDatabaseFetches a database.
listSchemasList schemas.
fetchSchemaFetches a schema.
listTablesList tables
fetchTableFetch a table.
listViewsList views
fetchViewFetch a view
listWarehousesList warehouse
fetchWarehouseDescribe warehouse
listIcebergTablesList iceberg tables
createSnowflakeManagedIcebergTableCreate a snowflake managed iceberg table
createSnowflakeManagedIcebergTableAsSelectCreate a snowflake managed iceberg table as select
createUnmanagedIcebergTableFromAWSGlueCatalogCreate an unmanaged iceberg table from AWS Glue catalog
createUnmanagedIcebergTableFromDeltaCreate an unmanaged iceberg table from Delta
createUnmanagedIcebergTableFromIcebergFilesCreate an unmanaged iceberg table from Iceberg files
createUnmanagedIcebergTableFromIcebergRestCreate an unmanaged iceberg table from Iceberg REST
fetchIcebergTableDescribe iceberg table
dropIcebergTableDrop iceberg table
resumeReclusterIcebergTableResume recluster of an iceberg table
suspendReclusterIcebergTableSuspend recluster of an iceberg table
refreshIcebergTableRefreshes the metadata for an Apache Iceberg table that uses an external Iceberg catalog
convertToManagedIcebergTableConverts an Apache Iceberg table that uses an external Iceberg catalog into a table that uses Snowflake as the catalog
undropIcebergTableUndrop iceberg table
cloneSnowflakeManagedIcebergTableClone a snowflake managed iceberg table
createSnowflakeManagedIcebergTableLikeCreate iceberg table like
setIcebergTableTagsSet tags on an Iceberg table.
unsetIcebergTableTagsUnset tags from an Iceberg table.
getIcebergTableTagsGet the tag assignments for an Iceberg table.
On this Page