Load Browser MCP Tools as Semantic Kernel Plugins

Prerequisites

• .NET 8+ or Python 3.10+
• Node.js 18+
• Agent MCP Studio at agentmcp.studio

Step-by-Step Setup

1. Start bridge.js

   curl -O https://agentmcp.studio/bridge.js && npm install ws
   node bridge.js wss://agentmcp.studio/api/relay/YOUR-UUID

   Open the studio → Settings → MCP Relay Bridge → click Connect.

2. Python: Load MCP tools into a kernel

   import asyncio
   from semantic_kernel import Kernel
   from semantic_kernel.connectors.ai.open_ai import OpenAIChatCompletion
   from semantic_kernel.connectors.mcp import MCPStdioPlugin
   
   async def main():
       kernel = Kernel()
       kernel.add_service(
           OpenAIChatCompletion(ai_model_id="gpt-4o")
       )
       
       # Load all tools from Agent MCP Studio
       async with MCPStdioPlugin(
           name="agent_mcp_studio",
           command="node",
           args=["/path/to/bridge.js",
                 "wss://agentmcp.studio/api/relay/YOUR-UUID"]
       ) as mcp_plugin:
           kernel.add_plugin(mcp_plugin)
           
           result = await kernel.invoke_prompt(
               "Use the available tools to query the sales database"
           )
           print(result)
   
   asyncio.run(main())

3. C# (.NET): Load MCP tools into a kernel

   using Microsoft.SemanticKernel;
   using Microsoft.SemanticKernel.Connectors.MCP;
   
   var builder = Kernel.CreateBuilder();
   builder.AddOpenAIChatCompletion("gpt-4o", Environment.GetEnvironmentVariable("OPENAI_API_KEY")!);
   var kernel = builder.Build();
   
   // Load tools from Agent MCP Studio's relay
   await using var mcpClient = await McpClientFactory.CreateStdioClientAsync(
       "agent-mcp-studio",
       new() {
           Command = "node",
           Arguments = ["/path/to/bridge.js",
                        "wss://agentmcp.studio/api/relay/YOUR-UUID"]
       }
   );
   
   var plugin = await kernel.ImportPluginFromMcpClientAsync(
       "AgentMCPStudio",
       mcpClient
   );
   
   var result = await kernel.InvokePromptAsync(
       "Use the available tools to analyse our sales data"
   );
   Console.WriteLine(result);

Related Integrations