Overview
What is Hologres MCP Server?
Hologres MCP Server is a universal interface between AI Agents and Hologres databases. It enables seamless communication, letting AI Agents retrieve database metadata and execute SQL operations.
How to use Hologres MCP Server?
Install via pip (pip install hologres-mcp-server) or clone the repository. Configure environment variables for Hologres connection (HOLOGRES_HOST, PORT, USER, PASSWORD, DATABASE) and optionally SERVER_HOST/PORT for SSE and HTTP Stream modes. Run the server using hologres-mcp-server with --transport set to stdio, sse (default), or http-stream. Integration examples are provided for Python and Node.js.
Key features of Hologres MCP Server
- Multiple transport modes: STDIO, SSE, and HTTP Stream
- Retrieve Hologres database metadata (schemas, tables)
- Execute SQL queries and analyze results
- Collect and view table statistics
- Get query plans and execution plans
Use cases of Hologres MCP Server
- AI agents querying Hologres databases in real time
- Retrieving schema and table DDL for database exploration
- Analyzing table statistics to optimize query performance
- Fetching query logs by user, application, or recent activity
- Integrating with MCP clients like Cherry Studio
FAQ from Hologres MCP Server
What transport modes does Hologres MCP Server support?
STDIO (standard I/O), SSE (Server-Sent Events), and HTTP Stream (Streamable HTTP). The default is SSE.
What environment variables are required?
HOLOGRES_HOST, HOLOGRES_PORT, HOLOGRES_USER, HOLOGRES_PASSWORD, and HOLOGRES_DATABASE are all required. SERVER_HOST and SERVER_PORT are only needed for SSE and HTTP Stream modes.
How can I install Hologres MCP Server?
You can use pip (pip install hologres-mcp-server) or clone the repository and run it with uv. Configuration is added to the MCP client’s mcpServers section.
Can Hologres MCP Server work with Cherry Studio?
Yes. Cherry Studio can connect in SSE mode (e.g., http://localhost:8001), HTTP Stream mode (http://localhost:8001/mcp), or STDIO mode with the appropriate command.
What are the runtime requirements?
Python 3.10 or higher and uv for package management.
