MCP Server - CLI & IDEs

Use the Xweather MCP Server from Cursor, VS Code, a supported JetBrains IDE, Claude Code, or GitHub Copilot CLI. Follow the setup for your development environment to access Xweather tools alongside your code.

VS Code

1. Update to the latest GitHub Copilot and Copilot Chat extensions.
2. Open the command palette and run MCP: Add Server.
3. Choose HTTP or HTTP(s), then enter https://mcp.api.xweather.com/mcp.
4. Give the server a label such as xweather for quick identification.
5. Add your API key when prompted for the credential.
6. Review and accept the trust prompt so VS Code can start the server and discover its tools.
7. If you want to limit the server to a specific repo or workflow, add an allowed_tools allowlist in the workspace MCP config instead of leaving the full tool set enabled.

Cursor Desktop

1. On the leftand panel, select Customize

2. Select MCPs, then New MCP Server

3. Add the following configuration to the newly opened .cursor/mcp.json. If the file already contains servers, add Xweather to the existing mcpServers object.

   {
       "mcpServers": {
           "Xweather": {
               "url": "https://mcp.api.xweather.com/mcp"
           }
       }
   }

4. Save the file. Then complete the OAuth sign-in in your browser using your Xweather client ID and client secret.

5. Alternatively, authenticate directly with an API key by adding an Authorization header:

{
    "mcpServers": {
        "Xweather": {
            "url": "https://mcp.api.xweather.com/mcp",
            "headers": {
                "Authorization": "Bearer ${env:XWEATHER_API_KEY}"
            }
        }
    }
}

JetBrains IDEs

1. Install or update the GitHub Copilot plugin for your JetBrains IDE.
2. Open the GitHub Copilot settings or MCP configuration page in the IDE preferences.
3. Add a new HTTP MCP server with the URL https://mcp.api.xweather.com/mcp.
4. Enter your API key when prompted.
5. Save the config and reload or restart the IDE if the server does not appear immediately.

CLI Workflows

Claude Code and GitHub Copilot CLI follow the same standard MCP workflow for remote HTTP servers:

1. Launch a session in your project directory (claude or copilot).
2. Add the Xweather MCP server using your CLI’s mcp add command:

claude mcp add --transport http xweather https://mcp.api.xweather.com/mcp \
  --header "Authorization: Bearer <client_id>_<client_secret>"

copilot mcp add --transport http xweather https://mcp.api.xweather.com/mcp \
  --header "Authorization: Bearer <client_id>_<client_secret>"

3. Run /mcp in your active session to verify that the server is connected and authenticated.

Prompt Ideas

• “Draft a README section explaining the lightning tool output format.”
• “Update .env.example to include the MCP API key with helpful comments.”
• “Suggest integration tests for the custom agent that polls xweather_get_forecast_weather.”