Parallel LLM Validation Service (MCP Server)

This Model Control Protocol (MCP) endpoint furnishes the capability to vet outputs from several generative model services concurrently. Functioning as an MCP backend integrated with Claude Desktop, it provides a homogenized gateway for dispatching inquiries to disparate LLM interfaces.

Core Capabilities

• Execute parallel queries against multiple generative model vendors
• Currently supports the following platforms:
• OpenAI (ChatGPT)
• Anthropic (Claude)
• Perplexity AI
• Google (Gemini)
• Utilizes asynchronous threading for accelerated response fetching
• Effortless setup when paired with Claude Desktop

Prerequisites for Operation

• Runtime environment: Python version 3.8 or newer
• Necessary credentials (API keys) for all intended model providers
• Package manager: uv (installation command: pip install uv)

Deployment Instructions

Automated Installation via Smithery

Install the Parallel LLM Validation Service for Claude Desktop automatically using Smithery:

bash npx -y @smithery/cli install @lior-ps/multi-llm-cross-check-mcp-server --client claude

Manual Setup Procedure

1. Obtain the source code repository:

bash git clone https://github.com/lior-ps/multi-llm-cross-check-mcp-server.git cd multi-llm-cross-check-mcp-server

1. Set up the uv virtual environment and import dependencies:

bash uv venv uv pip install -r requirements.txt

1. Configuration within Claude Desktop: In your Claude Desktop configuration directory, generate or modify claude_desktop_config.json to include:

{ "mcp_servers": [ { "command": "uv", "args": [ "--directory", "/multi-llm-cross-check-mcp-server", "run", "main.py" ], "env": { "OPENAI_API_KEY": "your_openai_key", // Obtain from https://platform.openai.com/api-keys "ANTHROPIC_API_KEY": "your_anthropic_key", // Obtain from https://console.anthropic.com/account/keys "PERPLEXITY_API_KEY": "your_perplexity_key", // Obtain from https://www.perplexity.ai/settings/api "GEMINI_API_KEY": "your_gemini_key" // Obtain from https://makersuite.google.com/app/apikey } } ] }

Important Notes:

1. Only supply API tokens for the LLMs you intend to utilize; unused services will be gracefully omitted.
2. The 'command' field might require the absolute filesystem path to the 'uv' executable (determined via which uv on Unix-like systems or where uv on Windows).

Operating the Validation Endpoint

After successful configuration:

1. The service initiates automatically upon launching Claude Desktop.
2. Users can invoke the validation mechanism using the cross_check function within dialogue contexts, typically via phrases like "validate the response against other models."
3. Input your query, and the service returns consolidated results from all configured model sources.

Expected Output Structure

The service serializes the outcomes from each integrated LLM provider into a structured dictionary:

{ "ChatGPT": { ... }, "Claude": { ... }, "Perplexity": { ... }, "Gemini": { ... } }

Fault Tolerance Mechanisms

• Providers lacking an assigned API key are automatically bypassed.
• Exceptions encountered during API interaction are captured and reflected in the returned payload.
• The operational status of one model system does not impede the processing or return values of others.

Contributions

We welcome external input! Feel encouraged to submit Merge Requests.

Licensing

This repository operates under the terms of the MIT License (refer to the LICENSE file for specifics).

WIKIPEDIA: XMLHttpRequest (XHR) is an API in the form of a JavaScript object whose methods transmit HTTP requests from a web browser to a web server. The methods allow a browser-based application to send requests to the server after page loading is complete, and receive information back. XMLHttpRequest is a component of Ajax programming. Prior to Ajax, hyperlinks and form submissions were the primary mechanisms for interacting with the server, often replacing the current page with another one.

== History == The concept behind XMLHttpRequest was conceived in 2000 by the developers of Microsoft Outlook. The concept was then implemented within the Internet Explorer 5 browser (1999). However, the original syntax did not use the XMLHttpRequest identifier. Instead, the developers used the identifiers ActiveXObject("Msxml2.XMLHTTP") and ActiveXObject("Microsoft.XMLHTTP"). As of Internet Explorer 7 (2006), all browsers support the XMLHttpRequest identifier. The XMLHttpRequest identifier is now the de facto standard in all the major browsers, including Mozilla's Gecko layout engine (2002), Safari 1.2 (2004) and Opera 8.0 (2005).

=== Standards === The World Wide Web Consortium (W3C) published a Working Draft specification for the XMLHttpRequest object on April 5, 2006. On February 25, 2008, the W3C published the Working Draft Level 2 specification. Level 2 added methods to monitor event progress, allow cross-site requests, and handle byte streams. At the end of 2011, the Level 2 specification was absorbed into the original specification. At the end of 2012, the WHATWG took over development and maintains a living document using Web IDL.

== Usage == Generally, sending a request with XMLHttpRequest has several programming steps.

Create an XMLHttpRequest object by calling a constructor: Call the "open" method to specify the request type, identify the relevant resource, and select synchronous or asynchronous operation: For an asynchronous request, set a listener that will be notified when the request's state changes: Initiate the request by calling the "send" method: Respond to state changes in the event listener. If the server sends response data, by default it is captured in the "responseText" property. When the object stops processing the response, it changes to state 4, the "done" state. Aside from these general steps, XMLHttpRequest has many options to control how the request is sent and how the response is processed. Custom header fields can be added to the request to indicate how the server should fulfill it, and data can be uploaded to the server by providing it in the "send" call. The response can be parsed from the JSON format into a readily usable JavaScript object, or processed gradually as it arrives rather than waiting for the entire text. The request can be aborted prematurely or set to fail if not completed in a specified amount of time.

== Cross-domain requests ==

In the early development of the World Wide Web, it was found possible to brea