Snowflake
Lets an agent query Snowflake data, run SQL, manage objects, and use Cortex AI services. Summary written by AI Index.
The registry’s description: MCP Server for Snowflake from Snowflake Labs
Add to your client
Add the Snowflake MCP server to Claude Code
Run the command
claude mcp add snowflake -- uvx snowflake-labs-mcp --service-config-fileFill in the values ↓
Check it’s connected
Add the Snowflake MCP server to Cursor
Open mcp.json· Where the file is:
~/.cursor/mcp.json,.cursor/mcp.jsonPaste the server entry
{ "mcpServers": { "snowflake": { "command": "uvx", "args": [ "snowflake-labs-mcp", "--service-config-file" ] } } }Fill in the values ↓
Check it’s connected
Add the Snowflake MCP server to VS Code
Open mcp.json· Where the file is:
.vscode/mcp.jsonPaste the server entry
{ "servers": { "snowflake": { "command": "uvx", "args": [ "snowflake-labs-mcp", "--service-config-file" ] } } }Fill in the values ↓
Check it’s connected
Add the Snowflake MCP server to Codex
Run the command
codex mcp add snowflake -- uvx snowflake-labs-mcp --service-config-fileFill in the values ↓
Check it’s connected
Add the Snowflake MCP server to Devin Desktop
Open mcp_config.json· Where the file is:
~/.config/devin/mcp_config.json,%APPDATA%\devin\mcp_config.jsonPaste the server entry
{ "mcpServers": { "snowflake": { "command": "uvx", "args": [ "snowflake-labs-mcp", "--service-config-file" ] } } }Fill in the values ↓
Check it’s connected
Add the Snowflake MCP server to Gemini CLI
Open settings.json· Where the file is:
~/.gemini/settings.json,.gemini/settings.jsonPaste the server entry
{ "mcpServers": { "snowflake": { "command": "uvx", "args": [ "snowflake-labs-mcp", "--service-config-file" ] } } }Fill in the values ↓
Check it’s connected
Add the Snowflake MCP server to GitHub Copilot
Run the command
copilot mcp add snowflake -- uvx snowflake-labs-mcp --service-config-fileFill in the values ↓
Check it’s connected
Add the Snowflake MCP server to Cline
Open cline_mcp_settings.json· Where the file is:
~/.cline/data/settings/cline_mcp_settings.jsonPaste the server entry
{ "mcpServers": { "snowflake": { "command": "uvx", "args": [ "snowflake-labs-mcp", "--service-config-file" ] } } }Fill in the values ↓
Check it’s connected
Add the Snowflake MCP server to Kiro
Open mcp.json· Where the file is:
.kiro/settings/mcp.json,~/.kiro/settings/mcp.jsonPaste the server entry
{ "mcpServers": { "snowflake": { "command": "uvx", "args": [ "snowflake-labs-mcp", "--service-config-file" ] } } }Fill in the values ↓
Check it’s connected
Add the Snowflake MCP server to Zed
Open settings.json
Paste the server entry
{ "context_servers": { "snowflake": { "command": "uvx", "args": [ "snowflake-labs-mcp", "--service-config-file" ], "env": {} } } }Fill in the values ↓
Check it’s connected
Add the Snowflake MCP server to OpenCode
Open opencode.json· Where the file is:
~/.config/opencode/opencode.json,opencode.jsonPaste the server entry
{ "$schema": "https://opencode.ai/config.json", "mcp": { "snowflake": { "type": "local", "command": [ "uvx", "snowflake-labs-mcp", "--service-config-file" ], "enabled": true } } }Fill in the values ↓
Check it’s connected
Add the Snowflake MCP server to JetBrains AI Assistant
Open JetBrains AI Assistant’s MCP config
- Open Settings | Tools | AI Assistant | Model Context Protocol (MCP).
- Choose Add.
Paste the server entry
{ "mcpServers": { "snowflake": { "command": "uvx", "args": [ "snowflake-labs-mcp", "--service-config-file" ] } } }Fill in the values ↓
Check it’s connected
If the file already has servers, add the entry inside the existing object. To check it’s connected, restart or reload the client and look for snowflake in its list of MCP servers. A server that needs sign-in opens it on first use.
What you fill in
Required values appear as <NAME> placeholders in the snippets; add optional ones the same way. Descriptions are the server’s registry entry’s.
<SNOWFLAKE_ACCOUNT>Environment variable OptionalAccount identifier (e.g. xy12345.us-east-1)
<SNOWFLAKE_HOST>Environment variable OptionalSnowflake host URL
<SNOWFLAKE_USER>Environment variable OptionalUsername for authentication
<SNOWFLAKE_PASSWORD>Environment variable Optional · SecretPassword or programmatic access token
<SNOWFLAKE_ROLE>Environment variable OptionalRole to use for connection
<SNOWFLAKE_WAREHOUSE>Environment variable OptionalWarehouse to use for queries
<SNOWFLAKE_PASSCODE>Environment variable OptionalMFA passcode for authentication
<SNOWFLAKE_PRIVATE_KEY>Environment variable Optional · SecretPrivate key for key pair authentication
<SNOWFLAKE_PRIVATE_KEY_FILE>Environment variable OptionalPath to private key file
<SNOWFLAKE_PRIVATE_KEY_FILE_PWD>Environment variable OptionalPassword for encrypted private key
<SNOWFLAKE_MCP_HOST>Environment variable OptionalHost address to bind the server to (default: 0.0.0.0)
<SNOWFLAKE_MCP_PORT>Environment variable OptionalPort number for the server to listen on (default: 9000)
<SNOWFLAKE_MCP_ENDPOINT>Environment variable OptionalEndpoint path for the MCP server (default: /mcp)
Client compatibility
By transport: whether the client’s docs list a transport this server offers. It doesn’t cover sign-in or other requirements.
- Claude Code Transport supported
- Cursor Transport supported
- VS Code Client docs don’t list transports
- Codex Transport supported
- Devin Desktop Transport supported
- Gemini CLI Transport supported
- GitHub Copilot Client docs don’t list transports
- Cline Transport supported
- Kiro Transport supported
- Zed Client docs don’t list transports
- OpenCode Client docs don’t list transports
- JetBrains AI Assistant Transport supported
- Claude Client docs don’t list transports
- ChatGPT No shared transport
Sources
- Official MCP Registry registry.modelcontextprotocol.io · checked Nov 4, 2026
- GitHub organization Snowflake-Labs (verified domain) github.com · checked Nov 4, 2026
This data comes from the Official MCP Registry.