Materialize MCP Server
About
Materialize MCP Server transforms your Materialize indexed views into fully-typed, callable tools via the Model Context Protocol (MCP). It lets you define stable, versioned, and secure data tools simply by creating SQL views and indexing them—no additional code required.
Details
- Author
- MaterializeInc
- Downloads
- 115
- Categories
- Other
Jump to
- Stable: define once, used repeatedly with consistent business logic.
- Typed: input and output schemas are automatically derived from the index.
- Observable: usage is logged per tool, making cost and performance explicit.
- Secure: only callable if a view and index are created.
- No extra code: tools are defined entirely through SQL views and indexes.
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
Materialize MCP ServerCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Clone the repository and run uv run materialize-mcp-server (or uv run materialize-mcp). Configure the server using command-line arguments or environment variables, such as --mz-dsn for the Materialize connection string and --transport for the communication protocol (stdio or sse).
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"materialize mcp server": {
"materialize-mcp-server": {
"command": "uv",
"args": [
"run",
"materialize-mcp-server"
]
}
}
}
}McpServers
{
"materialize-mcp-server": {
"command": "uv",
"args": [
"run",
"materialize-mcp-server"
]
}
}Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
