GenAIScript

You use the GenAIScript MCP server to access Power Platform, Graph API, and Azure REST API capabilities through a unified MCP interface. This server lets you run MCP-enabled tooling locally or via a remote configuration, enabling streamlined AI-assisted access to these data sources and APIs.

How to use

To run the MCP server, you configure a local client to launch the MCP-enabled toolset and connect to the GenAIScript MCP server. You can start the server from your development environment using the pre-defined stdio configurations. This lets you interact with Power Platform APIs, Graph API, and Azure REST API via MCP prompts and commands.

How to install

Prerequisites and setup steps you need to follow are: install NodeJS, install the genaiscript CLI, and ensure you have a working shell to launch the MCP server. The server configuration shown below is intended for standard local usage with the GenAIScript MCP tooling.

{
  "servers": {
    "genaiscript-mcp-remote-win": {
      "type": "stdio",
      "command": "cmd",
      "args": [
        "/c",
        "npx",
        "genaiscript",
        "mcp",
        "--remote",
        "https://github.com/rajyraman/genaiscript-pac-az-mcp/",
        "--remote-branch",
        "main",
        "--groups",
        "mcp"
      ],
      "env": {
        "DEBUG": "*"
      }
    },
    "genaiscript-mcp-remote-linux": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "genaiscript",
        "mcp",
        "--remote",
        "https://github.com/rajyraman/genaiscript-pac-az-mcp/",
        "--remote-branch",
        "main",
        "" --groups",
        "mcp"
      ],
      "env": {
        "DEBUG": "*"
      }
    }
  }
}
``-please ignore the stray quote in the linux block above and use the intended pair of arguments shown in the source blocks.

Additional setup notes

Two practical ways to run the MCP server are shown for Windows and Mac/Linux. If you are on Windows, launch the MCP server with a command shell via cmd as shown in the Windows snippet. On macOS or Linux, launch via npx directly. You can also configure a VSCode or Claude/Desktop workflow by saving the appropriate stdio entry in your mcp configuration.

// Windows config example (cmd launcher)
{
  "type": "stdio",
  "command": "cmd",
  "args": [
    "/c",
    "npx",
    "genaiscript",
    "mcp",
    "--remote",
    "https://github.com/rajyraman/genaiscript-pac-az-mcp/",
    "--remote-branch",
    "main",
    "--groups",
    "mcp"
  ],
  "env": {"DEBUG": "*"}
}

// macOS/Linux config example (npx launcher)
{
  "type": "stdio",
  "command": "npx",
  "args": [
    "genaiscript",
    "mcp",
    "--remote",
    "https://github.com/rajyraman/genaiscript-pac-az-mcp/",
    "--remote-branch",
    "main",
    "--groups",
    "mcp"
  ],
  "env": {"DEBUG": "*"}
}

Troubleshooting and tips

If you experience authentication or startup issues, ensure you have the required credentials and environment values in place, and use the DEBUG environment variable to gather verbose logs. If you use a DevContainer workflow, you can populate the devcontainer environment file with PAC_CLIENT_ID, PAC_CLIENT_SECRET, AZ_CLIENT_ID, AZ_CLIENT_SECRET, and related settings as shown in the sample. This helps isolate issues related to external service authentication.

If Copilot or a similar tool interferes with authentication, running the MCP server inside a DevContainer or using Claude Desktop can help avoid conflicts on Windows systems.

Notes on usage

The MCP server enables a unified flow to query and act on Azure REST API, Graph API, and Power Platform APIs. You can request environment details, subscription information, or end-to-end data operations through MCP-enabled prompts and workflows.

Available tools

GenAIScript MCP server

Runs the MCP-enabled tooling to connect to Power Platform APIs, Graph API, and Azure REST API through MCP prompts and workflows.

Power Platform CLI access

Interacts with Power Platform Admin APIs/Dataverse APIs via MCP-enabled tooling.

Graph API access

Queries Graph API data and endpoints through the MCP framework.

Azure REST API access

Accesses Azure REST API endpoints via MCP tooling and the Azure CLI integration.