mcpField
M
mcp-proxy

mcp-proxy

Local
github.com/imandra-aiDeveloper Tools2Updated 73w ago
Repository

About mcp-proxy

mcp-proxy bridges remote MCP servers to clients that use stdio rather than Streamable HTTP. It is currently tailored to Imandra Universe and includes first-class support for its authentication.

Key features

  1. 01Expose Streamable HTTP through stdio
  2. 02Connect to remote MCP servers
  3. 03Support Imandra Universe authentication
  4. 04Configure servers with environment variables

Use cases

  1. 01Use a remote server from a stdio-only client
  2. 02Connect to Imandra Universe MCP servers
  3. 03Specify a custom development or deployment endpoint

MCP support

The proxy wraps a remote MCP server's Streamable HTTP transport in a stdio interface for broader client compatibility. It supports configuring Imandra Universe servers and credentials.

#mcp#mcp-server

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "proxy-imandra-ai": {
      "command": "npx",
      "args": [
        "-y",
        "github:imandra-ai/mcp-proxy"
      ]
    }
  }
}
Run commandnpx -y github:imandra-ai/mcp-proxy

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npmgithub:imandra-ai/mcp-proxy

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

mcp-proxy

A basic proxy for remote Model Context Protocol (MCP) servers, designed to make the Streamable HTTP transport usable by a broader set of clients by offering a stdio facade.

This is currently specialised to the Imandra Universe MCP servers (with first-class support for Imandra Universe authentication), but would be easy to generalise.

Usage

Recommended usage is via npx, which is included with npm. Similar functionality is available in other JavaScript toolchains (e.g. bunx, yarn dlx).

npx -y github:imandra-ai/mcp-proxy
Configuration

Configuration is possible via environment variables, typically specified in an env field in your MCP server configuration file.

  • IMANDRA_API_KEY - your Imandra API key, available from Imandra Universe.
  • IMANDRA_MCP_SERVER - name/ID of the Imandra Universe MCP server to use (e.g. reasoner_gateway). Find available MCP servers in the Imandra Universe gallery.
  • REMOTE_URL - the URL of your remote MCP server (e.g. for a dev or local deployment). If not specified, defaults to https://api.imandra.ai/v1beta1/tools/mcp/${IMANDRA_MCP_SERVER}.
Example Cursor/Windsurf configuration
{
  "mcpServers": {
    "imandra-universe/reasoners": {
      "command": "npx",
      "env": {
        "IMANDRA_API_KEY": "my_api_key",
        "IMANDRA_MCP_SERVER": "reasoner_gateway"
      },
      "args": [
        "-y",
        "[email protected]:imandra-ai/mcp-proxy"
      ]
    }
  }
}

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.