playbooks

home / mcp / mcp ide bridge mcp server

MCP IDE Bridge MCP Server

Provides bidirectional MCP HTTP streamable messaging to enable client-to-client collaboration across IDEs and AI agents.

Installation
Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "mvp2o-ai-mcp-ide-bridge": {
      "url": "http://localhost:8111/mcp/"
    }
  }
}

You can connect multiple IDEs and tools in real time using the MCP IDE Bridge. This server enables client-to-client messaging so developers can collaborate across editors, share context, and coordinate AI agents as part of a unified MCP workflow.

How to use

Set up the MCP IDE Bridge as your central messaging layer. Create a messaging client in each IDE, configure each project to expose its own identity and recipients, and start exchanging messages in real time. You can fire-and-forget messages to multiple recipients, then poll for replies to keep everyone in sync. This enables cross-IDE collaboration, AI agent coordination, and streamlined tool integration.

How to install

Prerequisites you need before you start:

Option 1: Run with Docker (recommended) to quickly get a working instance.

docker run -d --name mcp-ide-bridge -p 8111:8111 mcp-messaging-server

Option 2: Run the Python development server directly.

# First-time setup (see Local Development section for full instructions)
pip install -r requirements.txt && pip install -e .

# Run server
python -m mcp_messaging.server --port 8111

Configuration and usage notes

Create a file named mcp_recipients.json in your project root. Each project owns a single file with a unique ID and a list of recipients it can communicate with. This is used by IDE agents to establish messaging routes.

{
  "my_id": "myproject_cursor",
  "recipients": {
    "teammate_vscode": {
      "name": "Teammate's Project",
      "description": "My teammate's project in VS Code"
    },
    "aiagent_windsurf": {
      "name": "AI Agent Project", 
      "description": "AI agent development in Windsurf"
    }
  },
  "server_info": {
    "url": "http://localhost:8111/mcp/",
    "transport": "http_streamable"
  }
}

Connect Your IDE

Follow these steps in each IDE to connect to the bridge.

Cursor IDE: create a local server reference and connect to the bridge URL.

VS Code: install the MCP extension, place mcp_recipients.json in the project root, and configure the MCP settings.

Non-IDE Clients

Non-IDE clients can use the same MCP protocol. Provide a recipients_config parameter to MCP tools to participate in the communication flow.

Security considerations

For desktop development and trusted networks, this bridge focuses on ease of use and rapid iteration. It does not enforce authentication, encryption, or auditing by default. If you require production-grade security, consider enterprise solutions that add SSO, encryption, and audit trails.

Testing

A lightweight test harness helps you validate identity, check-ins, and messaging flows. Use the provided CLI to exercise each core tool and verify end-to-end messaging.

Development

The project supports local development with Python and Docker. You can run a Docker container for quick starts or run the Python server in development mode to iterate on features.

Available tools

checkin_client

Register your presence and capabilities so others can discover and coordinate with you.

send_message_without_waiting

Fire-and-forget messaging to one or more recipients with an immediate return.

get_messages

Retrieve replies or messages sent to you after broadcasting.

get_my_identity

Obtain your configured identity and recipient information for setup guidance.

get_active_sessions

View current active connections and messaging sessions.