MCP Demo Streamable HTTP Bridge

mcp-demo-streamable-http-bridge

typescript

GitHub Stars

typescript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "masx200-mcp-demo-streamable-http-bridge": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-memory"
      ],
      "env": {
        "BRIDGE_API_PWD": "/path/to/workdir",
        "BRIDGE_API_PORT": "3000",
        "BRIDGE_API_TOKEN": "your-secret-token",
        "MEMORY_FILE_PATH": "/path/to/custom/memory.json",
        "BRIDGE_STREAMABLE_HTTP_PATH": "<BRIDGE_STREAMABLE_HTTP_PATH>"
      }
    }
  }
}

You will run a bridge server that converts stdio inputs into a streamable-http MCP interface, letting you manage MCP tools, prompts, and resources through standard HTTP/WebSocket/SSE endpoints. This bridge supports multiple MCP servers simultaneously and offers configuration options, hot reload, and secure access patterns to streamline development and testing workflows.

How to use

You can run multiple MCP servers through a single bridge and access them over HTTP or SSE endpoints. Use the memory server for in‑memory MCP tools and the time server for time-related MCP functionality, all bridged into a consistent HTTP API. Start the bridge with the configured stdio MCP servers, then connect your MCP client to the HTTP endpoints exposed by the bridge.

How to install

Prerequisites: install Node.js and npm, then set up the bridge project with the required dependencies.

Step 1: Install Node.js and npm on your system.

Step 2: Prepare your project directory and install dependencies.

Step 3: Start the bridge with the MCP servers you want to run.

Additional sections

Configuration overview: you can enable hot reload of the configuration, specify a listening host and port, and set API tokens for HTTP access. The bridge also supports a path prefix for the MCP HTTP endpoints and CORS origin restrictions. See the following server configurations for concrete examples.

Environment variables you may use to customize the bridge behavior include BRIDGE_STREAMABLE_HTTP_PATH, BRIDGE_API_TOKEN, BRIDGE_API_PORT, and BRIDGE_API_PWD. Command-line arguments also enable hot reload, config file paths, API keys, ports, hosts, and CORS origins, and you can enable the SSE server mode with dedicated endpoints.

MCP server configurations (stdio) you can run via the bridge include multiple servers defined in a single mcpServers grouping. The following examples show two servers wired through the bridge: one memory server and one time server.

{
  "mcpServers": {
    "memory": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-memory"],
      "env": {
        "MEMORY_FILE_PATH": "/path/to/custom/memory.json"
      }
    },
    "time": {
      "command": "uvx",
      "args": ["mcp-server-time", "--local-timezone=America/New_York"]
    }
  }
}

Built by

VeilStrat

AI signals for GTM teams