Backup

You can run and configure the spinqit MCP tools server to efficiently invoke SpinQ’s quantum hardware resources from AI models. This MCP server exposes a local, standardized interface that your MCP clients can connect to, manage execution, and submit quantum tasks to the cloud backend securely.

How to use

Set up the MCP server and connect your MCP client to the local stdio endpoint. Use the provided command configuration to launch the server in your environment, then configure your MCP client with the same environment variables to supply your private key and SpinQ Cloud username. Once running, you can submit 2-qubit or larger quantum circuits as QASM submissions through the MCP interface and receive results from the cloud platform.

How to install

Prerequisites you need before installation: Python 3.10 or higher. If Python is not installed, you can install it from the official Python website or use Conda to create an appropriate environment.

Install spinqit_mcp_tools from the Python package index.

Verify installation by running the QASM submitter module to ensure the tool initializes correctly.

Configure the MCP server by using the stdio configuration shown in the example below. Save this configuration in your client settings to enable interaction with the local MCP server.

{
  "mcpServers": {
    "qasm_sub": {
      "type": "stdio",
      "command": "cmd",
      "args": [
        "/C",
        "C:\\Users\\ylin\\.conda\\envs\\mcp-server-py310\\python.exe",
        "-m",
        "spinqit_mcp_tools.qasm_submitter"
      ],
      "env": {
        "PRIVATEKEYPATH":"<Your Privatekey Path>",
        "SPINQCLOUDUSERNAME":"<Your SpinQ Cloud Username>"
      }
    }
  }
}

Additional sections

Configuration and usage notes: you can adapt the stdio configuration block to your operating system. If you run on Windows, you may use the CMD-based command shown in the example. If you run on macOS or Linux, you can replace the command path with your Python executable path inside the environment and keep the same -m qasm_submitter module.

Security considerations: protect your PRIVATEKEYPATH and SPINQCLOUDUSERNAME. Only share configuration with trusted MCP clients and keep private keys in a secure location. Rotate keys periodically and monitor access to the MCP endpoint.

Troubleshooting tips: ensure Python 3.10+ is in your PATH or your Conda environment is active when launching the server. If the submitter fails to initialize, re-check the environment path used in the command and verify that the environment contains the spinqit_mcp_tools package.

Usage examples

To submit a QASM circuit through the MCP server, start the stdio server with the configured command, then use your MCP client to provide the PRIVATEKEYPATH and SPINQCLOUDUSERNAME in the environment settings for that connection. The client will route QASM submissions to the SpinQ cloud backend and return results.

Available tools

qasm_submitter

Submits QASM circuits to SpinQ cloud via the MCP server and returns execution results.