mcp

This MCP server provides access to Huawei Cloud services through the Model Context Protocol, enabling conversational AI assistants to securely interact with cloud resources such as ECS, OBS, GaussDB, and more via MCP-compliant requests. The server is designed to be started as a Python module and can be integrated into MCP workflows or marketplaces that expect a standard MCP server endpoint. Once running, you can use the MCP client tooling to send intents and tasks that map to Huawei Cloud operations, including querying resource states, creating or deleting resources, and performing management actions in a structured, context-aware manner. The configuration example shown in the README demonstrates how the server can be registered with an MCP marketplace or orchestrator, with endpoints and behavior defined in a config.yaml-like structure. To interact with the server, provide your cloud credentials via environment variables and issue MCP calls through your agent or marketplace UI; the server will translate these into Huawei Cloud API calls and return structured results for your conversational workflow.

Prerequisites:

• Python 3.10 or higher
• pip (comes with Python)
• Optional: uv (recommended) for running Python projects in a fast, lightweight runtime

Step-by-step installation:

1. Clone or download the repository and navigate to the project root.
2. Create and activate a Python environment (recommended):
   • python -m venv venv
   • source venv/bin/activate (Linux/macOS) or venv\Scripts\activate (Windows)
3. Install the package in editable mode (so changes reflect immediately):
   • pip install -e .
4. Install any additional dependencies if needed (from the repository, e.g., in huaweicloud_services_server or submodules):
   • pip install -e huaweicloud_services_server
5. Run the MCP server via the Python module (as provided in mcp_config):
   • python -m huaweicloud_services_server.mcp_server
6. Alternative with uv (if you have uv installed and prefer running a specific service):
   • uv run mcp-server

Environment setup (required):

• HUAWEI_ACCESS_KEY: your Huawei Cloud Access Key
• HUAWEI_SECRET_KEY: your Huawei Cloud Secret Key

These should be kept secure and are used by the server to authenticate with Huawei Cloud APIs.

Tips and known considerations:

• The README suggests using uv for running the service, but a Python module-based run is supported via the mcp_config example. If you use uv, ensure your entry module exposes the MCP server as expected by the uv runner.
• Set HUAWEI_ACCESS_KEY and HUAWEI_SECRET_KEY in your environment before starting the server to enable Huawei Cloud API access.
• When registering with MCP marketplaces (e.g., Cline), you may need to provide a mcpServers entry similar to the example in the README, mapping the server name to its URL, transport type, and auto-approval rules.
• If you modify the service or its submodules, run the installation steps again to ensure dependencies are updated.
• For troubleshooting, check logs for authentication errors, missing dependencies, or misconfigured endpoints, and verify that the environment variables are loaded correctly in your runtime environment.