OpenStack MCP Server is an OpenStack resource query service built on the Model Context Protocol (MCP). It exposes API interfaces to retrieve compute, storage, network, image, and other OpenStack resources across a cloud environment. The service supports real-time resource queries, multi-dimensional information across domains, and flexible filtering by attributes such as name, ID, and other conditions. Users can control the level of detail returned with options: basic, detailed, and full information. The MCP interface is fully compatible with MCP protocol, enabling seamless integration with large language models and downstream tooling. The server is implemented as a high-performance asynchronous HTTP service based on Starlette and Uvicorn, interacting with the OpenStack API via the OpenStack SDK. It can deliver structured OpenStack resource information to language models via MCP, and supports SSE streaming output for real-time feedback. The project provides installation instructions (Python 3.10+ and an OpenStack environment) and a simple usage example for starting the service and querying resources at /openstack. It also outlines a development path for adding new resource query tools.