Snowflake Labs MCP Server
Snowflake Cortex query and object management capabilities for AI workflows.
Data last scanned yesterday · Reviewed 2 months ago
Overview
This repository provides a community-developed MCP server designed to interface with Snowflake Cortex AI services. It enables AI agents to perform SQL orchestration, manage Snowflake objects, and interact with structured and unstructured data through Cortex Search, Analyst, and Agent tools. The project utilizes a configuration-based approach to grant specific SQL statement permissions and enable services like semantic view querying. It is implemented in Python and intended for users building RAG applications or automated data analysis tasks within the Snowflake ecosystem.
Our verdict
This server is deprecated and carries a high risk level with a security score of 30/100. The lack of any authentication method means anyone with access to the configuration file can execute tools, and the project explicitly confirms it is no longer maintained. Users should strictly avoid this version and migrate to the official, supported Snowflake MCP Server.
- Zero direct dependencies reduce supply-chain surface area.
- Licensed under the permissive Apache-2.0 license.
- Project is officially deprecated and no longer maintained.
- No authentication method implemented for tool access.
- Lacks read-only mode for SQL execution and object management.
Tools
| Tool | Description | Risk |
|---|---|---|
| create_object | Creates a new Snowflake object based on the specified type. | medium |
| drop_object | Drops an existing Snowflake object. | high |
| create_or_alter_object | Creates or alters an existing Snowflake object. | medium |
| describe_object | Describes the structure of a specified Snowflake object. | low |
| list_objects | Lists Snowflake objects of a specific type. | low |
| execute_sql | Executes generic SQL statements based on configured permissions. | high |
| list_semantic_views | Lists available Snowflake Semantic Views. | low |
| describe_semantic_view | Describes a specified Snowflake Semantic View. | low |
| list_semantic_view_metrics_and_dimensions | Lists the metrics and dimensions for a Semantic View. | low |
| query_semantic_view | Queries a Snowflake Semantic View. | low |
Compatibility
| Client | Local | Docker | Remote | Read-only |
|---|---|---|---|---|
| ChatGPT | ||||
| Claude Desktop | ||||
| Cursor | ||||
| VS Code | ||||
| Windsurf |
Frequently asked questions
›Is this Snowflake MCP server still supported?
No, this specific repository is deprecated and no longer maintained. Users should migrate to the official Snowflake MCP Server provided by Snowflake.
›How do I configure the server's available tools?
You provide a configuration file, typically named configuration.yaml, which contains sections for Cortex services, other service categories, and SQL execution permissions.
›Can I enable specific tool groups in the configuration?
Yes, you can enable object management, query execution, and semantic view querying by setting them to True within the other_services section of your configuration file.
›How does the server control what SQL statements can be executed?
The sql_statement_permissions section allows you to explicitly set individual SQL command types, such as Select, Create, or Drop, to either True or False to govern permitted operations.
›How are connections to Snowflake handled?
The server utilizes the Snowflake Python Connector for all authentication and connection processes.
Alternatives
Changelog
Badge
Maintain this server? Add the live badge to your README.