Tasker

Tasker MCP lets you expose and control Tasker tasks from your MCP-enabled applications. It provides a lightweight bridge that lets you run Tasker actions remotely, query tool descriptions, and execute actions through a consistent API. This enables automation workflows that combine Tasker capabilities with other systems and services.

How to use

You connect an MCP-enabled app to a running Tasker MCP server and point the app at the server’s transport. You can run tasks, pass arguments, and receive results through the MCP channel. Use the CLI transport for a locally hosted server or the stdio transport for a direct integration with other tools. In practice, you’ll import a Tasker profile, start the MCP server with the correct tools metadata, then configure your client to reach that server.

Typical usage patterns include starting the server in SSE mode to support web-based dashboards or using the stdio transport to integrate directly with a desktop or scripting environment. Once the server is running, you connect your MCP-enabled app by pointing it to the server address and transport. You’ll provide the necessary API key or token to authorize access.

How to install

Prerequisites you need before you install and run the Tasker MCP server:

• A Tasker-enabled device with Tasker installed and a Tasker profile ready to export.

• A computer or device where you can run the MCP server binary and copy the tool descriptions file.

Step-by-step commands you will execute to run the CLI server locally and make it available to your MCP client:

1. Copy the CLI server binary and tool descriptions to your device. For example, if your device is accessible at a path, copy the files there and rename the binary to mcp-server after copying.

2. Start the server in SSE mode using the tool descriptions file and your Tasker API key.

3. Alternatively, start the server through the stdio transport by passing the appropriate tool descriptions file and Tasker API key, so your MCP client can communicate via stdio.

./mcp-server --tools /path/to/toolDescriptions.json --tasker-api-key=tk_... --mode sse

# Or for stdio transport
payload='{"jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": { "name": "tasker_flash_text", "arguments": { "text": "Hi" }  } }'
echo $payload | ./mcp-server --tools /path/to/toolDescriptions.json --tasker-api-key=tk_...

Configure your MCP client example

Configure your MCP-enabled application to connect to the running server. If you are using a desktop app like Claude Desktop, you can specify the stdio transport by pointing to the server binary and the tool descriptions file, along with the Tasker host and API key.

{
  "mcpServers": {
    "tasker": {
      "command": "/home/luis/tasker-mcp/dist/tasker-mcp-server-cli-x86_64",
      "args": [
        "--tools",
        "/home/luis/tasker-mcp/dist/toolDescriptions.json",
        "--tasker-host",
        "192.168.1.123",
        "--tasker-api-key",
        "tk_...",
        "--mode",
        "stdio"
      ]
    }
  }
}

Available tools

tasker_flash_text

Sends a text payload to Tasker for display or processing as part of an MCP-enabled workflow.

MCP#parse_args

Internal argument parsing action used to extract and map MCP arguments for task execution.