---
title: "ClickHouse MCP Server: review, setup & limitations · Undominated.ai"
canonical: https://undominated.ai/mcp-servers/clickhouse/
description: "Queries ClickHouse and inspects table metadata, with read-only SQL by default and separate authentication for network transports."
---

# ClickHouse MCP Server: review, setup & limitations · Undominated.ai

> Queries ClickHouse and inspects table metadata, with read-only SQL by default and separate authentication for network transports.

[← Explore all mcp servers](/mcp-servers/)

ANALYTICS DATABASES / ClickHouse

# ClickHouse MCP Server

Queries ClickHouse and inspects table metadata, with read-only SQL by default and separate authentication for network transports.

 See setup guidance ↓Original source ↗

SOURCE REVIEW

 Reviewed 2026-09-21
 Evidence 4 linked sources
 Publisher ClickHouse
 Licence Apache-2.0 ↗
 Revision 801284456397
 Read what was—and wasn’t—checked ↓

“An MCP server for ClickHouse.”

 ClickHouse · upstream description ↗ Our analysis follows below.

01 / THE REASONING

## Why this made the selection.

 - Documents query permissions and HTTP caller authentication separately.
- Supports metadata inspection and parameterized SQL, useful for targeted analytics investigations.

### A good fit for

 - Inspecting schemas and running analytical queries under a limited ClickHouse user.

### Weigh up before choosing

 - Read-only queries can still consume substantial database resources.
- The server-side write guard is described as best effort; ClickHouse grants remain the permission boundary.

02 / THE REVIEW RECORD

## What we actually inspected.

Source review has boundaries. A clear record is more useful than a “safe” badge.

### Material inspected

 - README.md (installation, capability and permission sections)
- pyproject.toml
- mcp_clickhouse/mcp_server.py
- LICENSE

### Our findings

 - CLICKHOUSE_ALLOW_WRITE_ACCESS defaults false; writes require an explicit change.
- HTTP/SSE requires a configured bearer token or FastMCP auth provider unless authentication is deliberately disabled; the health probe is unauthenticated.

### Not established by this review

 - Server startup, authentication and tool execution were not tested.

The review applies to the material and revision named here. A newer upstream release can change its behavior.

03 / PUT IT TO WORK

## Connect a server deliberately.

Upstream setup instructions ↗

DOCUMENTED COMMAND

 uv run --with mcp-clickhouse --python 3.12 mcp-clickhouse Copy command ↗

Copying does not execute this command. It may retrieve a newer version than the reviewed source.

 - Follow the documented client launch configuration and configure the ClickHouse connection.
- Keep write access disabled for exploration; configure caller authentication before using a network transport.

### Before you start

 - Python/uv or the documented container runtime and a ClickHouse user.

### Compatibility

An MCP client supporting stdio, SSE, Streamable HTTP. · Python/uv or the documented container runtime and a ClickHouse user.

### Implementation

Python

### Transports

stdio · SSE · Streamable HTTP

### Analytics databases / Configured credentials or local access

 - Reads ClickHouse schemas and query results; explicit write mode can alter data or schema.
- Optional chDB functionality can access local data separately from the remote database.

### Cost model

ClickHouse hosting and query compute determine cost.

04 / FOLLOW THE EVIDENCE

## The source trail.

Our notes are separate from the original resource. Check upstream before adopting a new version.

 - Official documentation ↗ Checked 2026-09-21 https://github.com/ClickHouse/mcp-clickhouse/blob/80128445639794fa161179608c4862e9829a65fd/README.md Supports: summary, whySelected, bestFor, limitations, install, access, compatibility, transports
- Applicable source licence ↗ Checked 2026-09-21 https://github.com/ClickHouse/mcp-clickhouse/blob/80128445639794fa161179608c4862e9829a65fd/LICENSE Supports: license
- Implementation: pyproject.toml ↗ Checked 2026-09-21 https://github.com/ClickHouse/mcp-clickhouse/blob/80128445639794fa161179608c4862e9829a65fd/pyproject.toml Supports: language, review
- Implementation: mcp_clickhouse/mcp_server.py ↗ Checked 2026-09-21 https://github.com/ClickHouse/mcp-clickhouse/blob/80128445639794fa161179608c4862e9829a65fd/mcp_clickhouse/mcp_server.py Supports: language, review

KEEP COMPARING

## Other approaches to consider.

Related by category or shared topics. These are alternatives to inspect, not a measured quality order.

 [### AWS IaC MCP Server ↗ Validates CloudFormation templates, retrieves CDK/CloudFormation guidance and investigates failed deployments.](/mcp-servers/aws-iac/)[### Chroma MCP ↗ Lets an assistant create and query Chroma collections through local, HTTP or Chroma Cloud database connections.](/mcp-servers/chroma/)[### dbt MCP Server ↗ Connects dbt project metadata and lineage with optional local CLI, Semantic Layer and Platform operations.](/mcp-servers/dbt/)

 [AI Tools ↗](/tools/)[Skills ↗](/skills/)[Agents ↗](/agents/)[MCP Servers ↗](/mcp-servers/)
