Altinity MCP
Production-ready MCP server designed to empower AI agents and LLMs to interact seamlessly with ClickHouse. It exposes your ClickHouse database as a set of standardized tools and resources that adhere to the MCP protocol, making it easy for agents built on OpenAI, Claude, or other platforms to query, explore, and analyse your data.
README
Altinity MCP Server
A Model Context Protocol (MCP) server that provides tools for interacting with ClickHouse® databases. This server enables AI assistants and other MCP clients to query, analyze, and interact with ClickHouse® databases through a standardized protocol.
Features
- Multiple Transport Options: Support for STDIO, HTTP, and Server-Sent Events (SSE) transports
- JWE Authentication: Optional JWE-based authentication with encryption for secure database access
- TLS Support: Full TLS encryption support for both ClickHouse® connections and MCP server endpoints
- Comprehensive Tools: Built-in tools for listing tables, describing schemas, and executing queries
- Dynamic Tools: Automatically generate MCP tools from ClickHouse® views (see Dynamic Tools Documentation)
- Resource Templates: Dynamic resource discovery for database schemas and table information
- Query Prompts: AI-assisted query building and optimization prompts
- Configuration Management: Flexible configuration via files, environment variables, or CLI flags
- Hot Reload: Dynamic configuration reloading without server restart
Table of Contents
- Quick Start
- Integration Guide
- Installation & Deployment
- Configuration
- Available Tools
- Available Resources
- Available Prompts
- OpenAI GPTs Integration
- JWE Authentication
- TLS Configuration
- Testing
- Development
- CLI Reference
- Contributing
- License
- Support
Quick Start
Using STDIO Transport (Default)
# Basic usage with default settings
./altinity-mcp --clickhouse-host localhost --clickhouse-port 8123
# With custom database and credentials
./altinity-mcp \
--clickhouse-host clickhouse.example.com \
--clickhouse-port 9000 \
--clickhouse-protocol tcp \
--clickhouse-database analytics \
--clickhouse-username reader \
--clickhouse-password secret123 \
--clickhouse-limit 5000
Using HTTP Transport with OpenAPI
./altinity-mcp \
--transport http \
--address 0.0.0.0 \
--port 8080 \
--clickhouse-host localhost \
--openapi http
Using SSE Transport with JWE Authentication and OpenAPI
./altinity-mcp \
--transport sse \
--port 8080 \
--allow-jwe-auth \
--jwe-secret-key "your-jwe-encryption-secret" \
--jwt-secret-key "your-jwt-signing-secret" \
--clickhouse-host localhost \
--openapi http
Integration Guide
For detailed instructions on integrating Altinity MCP with various AI tools and platforms, see our Integration Guide.
Installation & Deployment
Using Docker
docker run -it --rm ghcr.io/altinity/altinity-mcp:latest --clickhouse-host clickhouse
Kubernetes with Helm
From OCI helm registry (recommended)
# Install from OCI registry
helm install altinity-mcp oci://ghcr.io/altinity/altinity-mcp/helm/altinity-mcp \
--set config.clickhouse.host=clickhouse.example.com \
--set config.clickhouse.database=default \
--set config.clickhouse.limit=5000
From local helm chart
git clone https://github.com/Altinity/altinity-mcp
cd altinity-mcp
helm install altinity-mcp ./helm/altinity-mcp \
--set config.clickhouse.host=clickhouse-service \
--set config.clickhouse.database=analytics \
--set config.server.transport=http \
--set config.server.port=8080
For detailed Helm chart configuration options, see Helm Chart README.
Docker Compose
version: '3.8'
services:
altinity-mcp:
build: .
ports:
- "8080:8080"
environment:
- CLICKHOUSE_HOST=clickhouse
- MCP_TRANSPORT=http
- MCP_PORT=8080
depends_on:
- clickhouse
entrypoint: ["/bin/sh", "-c", "/bin/altinity-mcp"]
clickhouse:
image: clickhouse/clickhouse-server:latest
ports:
- "8123:8123"
From Source
git clone https://github.com/altinity/altinity-mcp.git
cd altinity-mcp
go build -o altinity-mcp ./cmd/altinity-mcp
Configuration
Configuration File
Create a YAML or JSON configuration file:
# config.yaml
clickhouse:
host: "localhost"
port: 8123
database: "default"
username: "default"
password: ""
protocol: "http"
read_only: false
max_execution_time: 600
tls:
enabled: false
ca_cert: ""
client_cert: ""
client_key: ""
insecure_skip_verify: false
server:
transport: "stdio"
address: "0.0.0.0"
port: 8080
tls:
enabled: false
cert_file: ""
key_file: ""
ca_cert: ""
jwt:
enabled: false
secret_key: ""
openapi:
enabled: false
tls: false
dynamic_tools:
- regexp: "mydb\\..*"
prefix: "db_"
- name: "get_user_data"
regexp: "users\\.user_info_view"
logging:
level: "info"
Note: For detailed information about dynamic tools configuration, see the Dynamic Tools Documentation.
Use the configuration file:
./altinity-mcp --config config.yaml
Environment Variables
All configuration options can be set via environment variables:
export CLICKHOUSE_HOST=localhost
export CLICKHOUSE_PORT=8123
export CLICKHOUSE_DATABASE=analytics
export CLICKHOUSE_LIMIT=5000
export MCP_TRANSPORT=http
export MCP_PORT=8080
export LOG_LEVEL=debug
./altinity-mcp
Available Tools
execute_query
Executes SQL queries against ClickHouse® with optional result limiting.
Parameters:
query(required): The SQL query to executelimit(optional): Maximum number of rows to return (default: server configured limit, max: 10,000)
Available Resources
clickhouse://schema
Provides complete schema information for the ClickHouse® database in JSON format.
clickhouse://table/{database}/{table}
Provides detailed information about a specific table including schema, sample data, and statistics.
Available Prompts
No prompts currently available
OpenAI GPTs Integration
The Altinity MCP Server supports seamless integration with OpenAI GPTs through its OpenAPI-compatible endpoints. These endpoints enable GPT assistants to perform ClickHouse® database operations directly.
Authentication
- With JWE: Add the JWE token to either:
- Path parameter:
/{jwe_token}/openapi/...(now required) - Authorization header:
Bearer {token}(alternative) x-altinity-mcp-keyheader (alternative)
- Path parameter:
- Without JWE: Use server-configured credentials (no auth needed in requests)
Available Actions
1. Execute SQL Query
Path: /openapi/execute_query
Parameters:
jwe_token(path param): JWE authentication tokenquery(query param): SQL query to execute (required)limit(query param): Maximum rows to return (optional, default 1000, max 10000)
Example OpenAPI Path:
GET /{jwe_token}/openapi/execute_query?query=SELECT%20*%20FROM%20table&limit=500
Configuration Example for GPTs
{
"openapi": "3.1.0",
"info": {
"title": "ClickHouse® SQL Interface",
"version": "1.0.0"
},
"servers": [
{"url": "https://your-server:8080/{token}"}
],
"paths": {
"/{jwe_token}/openapi/execute_query": {
"get": {
"operationId": "execute_query",
"parameters": [
{
"name": "jwe_token",
"in": "path",
"required": true,
"schema": {"type": "string"}
},
{
"name": "query",
"in": "query",
"required": true,
"schema": {"type": "string"}
},
{
"name": "limit",
"in": "query",
"schema": {"type": "integer"}
}
]
}
}
}
}
Note: For Altinity Cloud deployments, use the provided endpoint URL with your organization-specific token.
JWE Authentication
When JWE authentication is enabled, the server expects tokens encrypted using AES Key Wrap (A256KW) and AES-GCM (A256GCM). These tokens contain ClickHouse® connection parameters:
{
"host": "clickhouse.example.com",
"port": 8123,
"database": "analytics",
"username": "user123",
"password": "secret",
"protocol": "http",
"secure": "false"
}
Generate tokens using the provided utility.
go run ./cmd/jwe_auth/jwe_token_generator.go \
--jwe-secret-key "your-jwe-encryption-secret" \
--jwt-secret-key "your-jwt-signing-secret" \
--host "clickhouse.example.com" \
--port 8123 \
--database "analytics" \
--username "user123" \
--password "password123" \
--expiry 86400
More details in jwe_authentication.md
TLS Configuration
ClickHouse® TLS
./altinity-mcp \
--clickhouse-tls \
--clickhouse-tls-ca-cert /path/to/ca.crt \
--clickhouse-tls-client-cert /path/to/client.crt \
--clickhouse-tls-client-key /path/to/client.key
Server TLS
./altinity-mcp \
--transport https \
--server-tls \
--server-tls-cert-file /path/to/server.crt \
--server-tls-key-file /path/to/server.key
Testing
Test ClickHouse® Connection
./altinity-mcp test-connection \
--clickhouse-host localhost \
--clickhouse-port 8123 \
--clickhouse-database default
Run Tests
go test ./...
Integration Tests
Integration tests use Docker containers and require Docker to be running:
go test -v ./cmd/altinity-mcp/...
Development
Prerequisites
- Go 1.24 or later
- Docker (for integration tests)
- ClickHouse® server (for development)
Building
go build -o altinity-mcp ./cmd/altinity-mcp
Running Tests
# Unit tests
go test ./pkg/...
# Integration tests (requires Docker)
go test -v ./cmd/altinity-mcp/...
CLI Reference
Global Flags
--config: Path to configuration file (YAML or JSON)--log-level: Logging level (debug/info/warn/error)--clickhouse-limit: Default limit for query results (default: 1000)--openapi: Enable OpenAPI endpoints (disable/http/https) (default: disable)
ClickHouse® Flags
--clickhouse-host: ClickHouse® server host--clickhouse-port: ClickHouse® server port--clickhouse-database: Database name--clickhouse-username: Username--clickhouse-password: Password--clickhouse-protocol: Protocol (http/tcp)--read-only: Read-only mode--clickhouse-max-execution-time: Query timeout in seconds--clickhouse-http-headers: HTTP headers for ClickHouse requests (key=value pairs)
Server Flags
--transport: Transport type (stdio/http/sse)--address: Server address--port: Server port--allow-jwe-auth: Enable JWE authentication--jwe-secret-key: Secret key for JWE token decryption (must be 32 bytes for A256KW).--jwt-secret-key: Secret key for JWT signature verification
Commands
version: Show version informationtest-connection: Test ClickHouse® connection
Contributing
- Fork the repository
- Create a feature branch
- Make your changes
- Add tests
- Run the test suite
- Submit a pull request
License
This project is licensed under the Apache License 2.0. See the LICENSE file for details.
Support
For support and questions:
- GitHub Issues: https://github.com/altinity/altinity-mcp/issues
- Email: support@altinity.com
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.