snowflake-cortex-agent-mcp-server

Snowflake-Labs/snowflake-cortex-agent-mcp-server

3.2

If you are the rightful owner of snowflake-cortex-agent-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

The Snowflake Cortex Model Context Protocol (MCP) Server is an experimental project designed to integrate Snowflake Cortex insights directly into development environments like Visual Studio Code.

The Snowflake Cortex Model Context Protocol (MCP) Server is an innovative solution that facilitates seamless integration with Snowflake Cortex APIs, allowing developers to access valuable insights directly within their preferred development environments, such as Visual Studio Code. This experimental project leverages the Model Context Protocol (MCP) to provide a streamlined and efficient way to interact with Snowflake Cortex, enhancing the development workflow by embedding data insights and analytics capabilities. The server is designed to be used for development and evaluation purposes, offering a unique opportunity to explore the potential of Snowflake Cortex in a controlled environment. By utilizing Docker for deployment, the server ensures a consistent and isolated environment, making it easier to manage dependencies and configurations. The project is not officially supported by Snowflake and is intended for those who wish to experiment with integrating advanced data insights into their development processes.

Features

  • Seamless integration with Snowflake Cortex APIs for enhanced data insights.
  • Support for Visual Studio Code and other MCP-compatible clients.
  • Docker-based deployment for easy setup and consistent environments.
  • Programmatic access token for secure authentication with Snowflake Cortex APIs.
  • Experimental project for development and evaluation purposes.

Usages

docker with VSCode

{
  "mcpServers": {
    "snowflake-cortex-agent": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "--name",
        "snowflake-cortex-agent",
        "-e",
        "SNOWFLAKE_ACCOUNT_URL",
        "-e",
        "SNOWFLAKE_USER",
        "-e",
        "SNOWFLAKE_PASSWORD",
        "-e",
        "SEMANTIC_MODEL_FILE",
        "-e",
        "CORTEX_SEARCH_SERVICE",
        "-e",
        "SNOWFLAKE_WAREHOUSE",
        "-e",
        "CORTEX_AGENT_LLM_MODEL",
        "ghcr.io/kameshsampath/snowflake-cortex-mcp/server:latest"
      ],
      "env": {
        "SNOWFLAKE_ACCOUNT_URL": "https://${input:snowflake_account}.snowflakecomputing.com",
        "SNOWFLAKE_USER": "${input:snowflake_user}",
        "SNOWFLAKE_PASSWORD": "${input:snowflake_password}",
        "SEMANTIC_MODEL_FILE": "${input:semantic_model_file}",
        "CORTEX_SEARCH_SERVICE": "${input:cortex_search_service}",
        "SNOWFLAKE_WAREHOUSE": "${input:snowflake_warehouse}",
        "CORTEX_AGENT_LLM_MODEL": "${input:cortex_agent_llm_model}"
      }
    }
  }
}

docker with Claude Desktop

{
  "mcpServers": {
    "snowflake-cortex-agent": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "--name",
        "snowflake-cortex-agent",
        "-e",
        "SNOWFLAKE_ACCOUNT_URL",
        "-e",
        "SNOWFLAKE_USER",
        "-e",
        "SNOWFLAKE_PASSWORD",
        "-e",
        "SEMANTIC_MODEL_FILE",
        "-e",
        "CORTEX_SEARCH_SERVICE",
        "-e",
        "SNOWFLAKE_WAREHOUSE",
        "-e",
        "CORTEX_AGENT_LLM_MODEL",
        "ghcr.io/kameshsampath/snowflake-cortex-mcp/server:latest"
      ],
      "env": {
        "SNOWFLAKE_ACCOUNT_URL": "https://${input:snowflake_account}.snowflakecomputing.com",
        "SNOWFLAKE_USER": "${input:snowflake_user}",
        "SNOWFLAKE_PASSWORD": "${input:snowflake_password}",
        "SEMANTIC_MODEL_FILE": "${input:semantic_model_file}",
        "CORTEX_SEARCH_SERVICE": "${input:cortex_search_service}",
        "SNOWFLAKE_WAREHOUSE": "${input:snowflake_warehouse}",
        "CORTEX_AGENT_LLM_MODEL": "${input:cortex_agent_llm_model}"
      }
    }
  }
}