Installation and setup
To use the MCP tools with your AI assistant, you'll need to install the Snowplow CLI and configure your chosen MCP client. This guide walks you through both steps.
1. Install Snowplow CLI
# Using Homebrew
brew install snowplow/taps/snowplow-cli
# Or download binary directly
curl -L -o snowplow-cli https://github.com/snowplow/snowplow-cli/releases/latest/download/snowplow-cli_linux_x86_64
chmod u+x snowplow-cli
If you have node.js
set up then no need to install, you can run via npx
.
2. Configure your MCP client
The Snowplow CLI MCP server can be used with any MCP-compatible client. Below are configuration examples for popular clients. For a complete list of supported clients and their configurations, see the MCP reference.
Claude desktop
Config location:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json
- Windows:
%APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"snowplow-cli": {
"command": "snowplow-cli",
"args": ["mcp"]
}
}
}
VS code
Add to .vscode/mcp.json
in your workspace:
{
"servers": {
"snowplow-cli": {
"type": "stdio",
"command": "snowplow-cli",
"args": ["mcp"]
}
}
}
Cursor
Add to .cursor/mcp.json
in your workspace:
{
"mcpServers": {
"snowplow-cli": {
"command": "snowplow-cli",
"args": ["mcp", "--base-directory", "."]
}
}
}
Using with npx
If using via npx
, use this configuration format instead:
{
"mcpServers": {
"snowplow-cli": {
"command": "npx",
"args": ["-y", "@snowplow/snowplow-cli", "mcp"]
}
}
}
For VS Code, adjust the type
field accordingly:
{
"servers": {
"snowplow-cli": {
"type": "stdio",
"command": "npx",
"args": ["-y", "@snowplow/snowplow-cli", "mcp"]
}
}
}
3. Start using MCP
After configuring your chosen client, start a new conversation or session. The Snowplow CLI MCP tools should be available for use.