Answers plain-English questions over Snowflake, BigQuery and Databricks data products and exposed API endpoints via Alkemi.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add alkemi --env BEARER_TOKEN='<BEARER_TOKEN>' --env MCP_NAME='<MCP_NAME>' --env PRODUCT_ID='<PRODUCT_ID>' -- npx @alkemiai/alkemi-mcp~/.codex/config.toml
[mcp_servers.alkemi]
command = "npx"
args = ["@alkemiai/alkemi-mcp"]
env = { BEARER_TOKEN = "<BEARER_TOKEN>", MCP_NAME = "<MCP_NAME>", PRODUCT_ID = "<PRODUCT_ID>" }
# codex mcp add alkemi --env BEARER_TOKEN='<BEARER_TOKEN>' --env MCP_NAME='<MCP_NAME>' --env PRODUCT_ID='<PRODUCT_ID>' -- npx @alkemiai/alkemi-mcp~/.cursor/mcp.json
{
"mcpServers": {
"alkemi": {
"command": "npx",
"args": [
"@alkemiai/alkemi-mcp"
],
"env": {
"BEARER_TOKEN": "<BEARER_TOKEN>",
"MCP_NAME": "<MCP_NAME>",
"PRODUCT_ID": "<PRODUCT_ID>"
}
}
}
}.vscode/mcp.json
{
"servers": {
"alkemi": {
"type": "stdio",
"command": "npx",
"args": [
"@alkemiai/alkemi-mcp"
],
"env": {
"BEARER_TOKEN": "<BEARER_TOKEN>",
"MCP_NAME": "<MCP_NAME>",
"PRODUCT_ID": "<PRODUCT_ID>"
}
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"alkemi": {
"command": "npx",
"args": [
"@alkemiai/alkemi-mcp"
],
"env": {
"BEARER_TOKEN": "<BEARER_TOKEN>",
"MCP_NAME": "<MCP_NAME>",
"PRODUCT_ID": "<PRODUCT_ID>"
}
}
}
}claude_desktop_config.json
{
"mcpServers": {
"alkemi": {
"command": "npx",
"args": [
"@alkemiai/alkemi-mcp"
],
"env": {
"BEARER_TOKEN": "<BEARER_TOKEN>",
"MCP_NAME": "<MCP_NAME>",
"PRODUCT_ID": "<PRODUCT_ID>"
}
}
}
}.trae/mcp.json
{
"mcpServers": {
"alkemi": {
"command": "npx",
"args": [
"@alkemiai/alkemi-mcp"
],
"env": {
"BEARER_TOKEN": "<BEARER_TOKEN>",
"MCP_NAME": "<MCP_NAME>",
"PRODUCT_ID": "<PRODUCT_ID>"
}
}
}
}Environment variables
BEARER_TOKENMCP_NAMEPRODUCT_ID
Replace the placeholders with your own credentials and keep them out of version control.