MCP Hydrolix
MCP server that lets AI clients run SQL and inspect schemas on a Hydrolix cluster
- Pricing
- Open source
- Platforms
- CLI, API, self-hosted
- Runs
- Locally
- Version
- 0.3.9
- Listing updated
- today
Similar MCP servers
About MCP Hydrolix
MCP Hydrolix is an open-source MCP server for databases. It runs locally on your machine and is started by your AI client.
The README says this server connects an MCP-compatible assistant such as Claude Desktop or Claude Code to a Hydrolix cluster, so the model can list databases and tables, read table schemas, and execute SQL queries against live data. Connection details are supplied through environment variables covering the cluster URL and either a service account token or a username and password.
The server runs over stdio by default and can also be served over HTTP or SSE, where per-request Bearer tokens are supported. A health endpoint reports whether it can reach the query head. Each query call takes a purpose field that is recorded with the query as a comment, which helps teams audit what was run.
It suits data engineers and analysts who already store logs or event data in Hydrolix and want to explore it conversationally. The README recommends using a read-only service account and including explicit time ranges in prompts to keep queries efficient.
Install
Run in a terminal
claude mcp add mcp-hydrolix --env HYDROLIX_URL='<HYDROLIX_URL>' --env HYDROLIX_TOKEN='<HYDROLIX_TOKEN>' --env HYDROLIX_PASSWORD='<HYDROLIX_PASSWORD>' -- uvx mcp-hydrolixAdd to claude_desktop_config.json
{
"mcpServers": {
"mcp-hydrolix": {
"command": "uvx",
"args": [
"mcp-hydrolix"
],
"env": {
"HYDROLIX_URL": "<HYDROLIX_URL>",
"HYDROLIX_TOKEN": "<HYDROLIX_TOKEN>",
"HYDROLIX_PASSWORD": "<HYDROLIX_PASSWORD>"
}
}
}
}Add to ~/.cursor/mcp.json
{
"mcpServers": {
"mcp-hydrolix": {
"command": "uvx",
"args": [
"mcp-hydrolix"
],
"env": {
"HYDROLIX_URL": "<HYDROLIX_URL>",
"HYDROLIX_TOKEN": "<HYDROLIX_TOKEN>",
"HYDROLIX_PASSWORD": "<HYDROLIX_PASSWORD>"
}
}
}
}Add to .vscode/mcp.json
{
"servers": {
"mcp-hydrolix": {
"type": "stdio",
"command": "uvx",
"args": [
"mcp-hydrolix"
],
"env": {
"HYDROLIX_URL": "<HYDROLIX_URL>",
"HYDROLIX_TOKEN": "<HYDROLIX_TOKEN>",
"HYDROLIX_PASSWORD": "<HYDROLIX_PASSWORD>"
}
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"mcp-hydrolix": {
"command": "uvx",
"args": [
"mcp-hydrolix"
],
"env": {
"HYDROLIX_URL": "<HYDROLIX_URL>",
"HYDROLIX_TOKEN": "<HYDROLIX_TOKEN>",
"HYDROLIX_PASSWORD": "<HYDROLIX_PASSWORD>"
}
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"mcp-hydrolix": {
"command": "uvx",
"args": [
"mcp-hydrolix"
],
"env": {
"HYDROLIX_URL": "<HYDROLIX_URL>",
"HYDROLIX_TOKEN": "<HYDROLIX_TOKEN>",
"HYDROLIX_PASSWORD": "<HYDROLIX_PASSWORD>"
}
}
}
}Using several servers? Build one config file with the MCP config generator.
Configuration
HYDROLIX_URLThe URL of your Hydrolix cluster, e.g. https://mycluster.hydrolix.liveRequiredHYDROLIX_TOKENService account token (alternative to username/password)HYDROLIX_USERHydrolix username (use with HYDROLIX_PASSWORD)HYDROLIX_PASSWORDHydrolix password (use with HYDROLIX_USER)
Tools this server provides (4)
Taken from the project documentation; the running server may differ.
- run_select_query
- Execute SQL queries on your Hydrolix cluster.
- list_databases
- List all databases on your Hydrolix cluster.
- list_tables
- List all tables in a database.
- get_table_info
- Get table metadata such as schema.
What it does
- SQL query execution
- Runs SQL against a Hydrolix cluster with an optional cap on result cell count.
- Schema discovery
- Lists databases and tables and returns table metadata such as schema.
- Per-query purpose logging
- Requires a reason for each query, recorded alongside it as a query comment.
- Flexible authentication
- Accepts a service account token or username and password through environment variables.
- HTTP and SSE transports
- Can run as a remote server with per-request credentials and a health check endpoint.
Good for
- Ask an assistant to explore which databases and tables exist on a cluster
- Query recent event data by time range from a chat interface
- Check a table's schema before writing analytics queries
- Let analysts investigate Hydrolix data without writing SQL by hand
Pricing
Open source
Open-source MCP server; connects to a Hydrolix cluster you already run.
MCP Hydrolix FAQ
How do I install it?
The README recommends running it with uv, which downloads mcp-hydrolix on demand. Alternatively, install it with pip on Python 3.13 or later.
Which credentials does it need?
It needs HYDROLIX_URL plus either HYDROLIX_TOKEN or both HYDROLIX_USER and HYDROLIX_PASSWORD for stdio use.
Can it run without local credentials?
When using HTTP or SSE transport, credentials can be supplied per request as a Bearer token or a token query parameter.
Is MCP Hydrolix free?
Yes. MCP Hydrolix is open source, so you can use and self-host it for free.
How do I add MCP Hydrolix to Claude Code?
Run this command in your terminal: claude mcp add mcp-hydrolix --env HYDROLIX_URL='<HYDROLIX_URL>' --env HYDROLIX_TOKEN='<HYDROLIX_TOKEN>' --env HYDROLIX_PASSWORD='<HYDROLIX_PASSWORD>' -- uvx mcp-hydrolix
What are the best alternatives to MCP Hydrolix?
The closest alternatives to MCP Hydrolix are Mysql, Mongodb MCP Server, MCP Server Couchbase, MCP Toolbox for Databases and HelixDB.
Where this comes from
- Description: written from the project's README and its MCP Registry entry.
- Install steps: from the official MCP Registry entry.