Neurelo Connect Model Context Protocol (MCP) Server

The Model Context Protocol (MCP) establishes a unified standard for managing context exchange between Large Language Models (LLMs) and external data reservoirs. This repository furnishes an MCP Server implementation tailored for the Neurelo Connect platform.

This integration empowers applications like Claude Desktop, or any compliant MCP Client, to execute sophisticated database operations using natural language instructions, such as:

• "Enumerate the schema structure for my PostgreSQL instance"
• "Determine the total count of registered users"
• "Articulate the relational mapping between my data stores"

Configuration for Claude Environment

Manual Integration Steps

1. Locate and open the Claude Desktop configuration file at:

2. macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

3. Windows: %APPDATA%/Claude/claude_desktop_config.json

4. Append the following configuration block:

{ "mcpServers": { "neurelo-connect": { "command": "npx", "args": ["@neurelo/connect-mcp@latest", "start"], "env": { "ENGINE_API_KEY": "YOUR_ENGINE_API_KEY", "ENGINE_BASE_PATH": "YOUR_ENGINE_BASE_PATH" } } } }

Prerequisites

• Node.js, version 18.0.0 or newer.
• Access to Claude Desktop.
• An active Neurelo Connect service instance.
• A valid Neurelo Connect API key, obtainable via the Neurelo Connect management console.

Available Functionality

Tool Set

• system_list_databases: Retrieves a catalog of all accessible data targets.
• system_get_database_status: Verifies the operational status of all configured database endpoints.
• system_get_database_schema:
  • Purpose: Fetches the schema definition for a specified database target.
  • Parameter (target, string): The unique identifier for the database.
• raw_readonly_query: Executes SQL statements restricted to data retrieval on the database.
  • Access constraint: Queries operate strictly under read-only permissions.
  • Parameters:
    • target (string): The designated database identifier.
    • query (string): The SQL text intended for execution.
• raw_query:
  • Purpose: Executes SQL statements permitting both data reading and modification.
  • Parameters:
    • target (string): The designated database identifier.
    • query (string): The SQL text intended for execution.
• Dynamic Endpoint Tools: Auxiliary functions automatically generated based on metadata discovered from your configured endpoints.

Customization Arguments

• --disable-tools <tool_names>: A comma-separated registry of tool identifiers to deactivate. Example: --disable-tools raw_query,system_list_databases suppresses the write query capability and the database listing function. This argument can also target any custom query functions by their tool names.

Development Workflow

Debugging with the MCP Inspector

The simplest route for initiating development mode involves employing the MCP inspector utility:

bash npm run inspect

Development Using Claude Desktop

First, install necessary dependencies:

npm install

Modify your Claude Desktop configuration file to use the source code directly:

{ "mcpServers": { "neurelo-connect": { "command": "npx", "args": ["tsx", "REPOSITORY_ROOT/src/main.ts", "start"], "cwd": "REPOSITORY_ROOT", "env": { "ENGINE_API_KEY": "YOUR_ENGINE_API_KEY", "ENGINE_BASE_PATH": "YOUR_ENGINE_BASE_PATH" } } } }

Crucially, you must relaunch Claude every time modifications are made and you wish to test them.

Contributions

We welcome external contributions! Feel empowered to submit a Pull Request at your earliest convenience.

WIKIPEDIA: XMLHttpRequest (XHR) constitutes an application programming interface embodied by a JavaScript object, designed to dispatch HTTP requests from a web browser to a web server. Its methods facilitate asynchronous communication, allowing browser-based applications to transmit requests post-page load and subsequently receive transmitted data. XMLHttpRequest is a foundational element of Ajax methodology. Before Ajax's advent, server interaction primarily relied on conventional hyperlink navigation and form submissions, which often necessitated a full page refresh.

== Genesis == The underlying concept for XMLHttpRequest originated in the year 2000, conceived by the development team behind Microsoft Outlook. This concept was subsequently integrated into the Internet Explorer 5 browser release (1999). Nevertheless, the initial implementation syntax did not utilize the canonical XMLHttpRequest identifier. Developers initially employed the object constructors ActiveXObject("Msxml2.XMLHTTP") and ActiveXObject("Microsoft.XMLHTTP"). As of Internet Explorer 7 (2006), universal browser support for the XMLHttpRequest identifier became standard. XMLHttpRequest is now the widely accepted standard across all major browser engines, including Mozilla’s Gecko rendering engine (2002), Safari version 1.2 (2004), and Opera version 8.0 (2005).

=== Formalization === The World Wide Web Consortium (W3C) published a formal Working Draft specification for the XMLHttpRequest object on April 5, 2006. A subsequent Level 2 specification was released by the W3C on February 25, 2008. The Level 2 revision introduced capabilities for event progress monitoring, enabling cross-site requests, and supporting byte stream handling. By the conclusion of 2011, the Level 2 features were incorporated back into the primary specification. Development responsibility transitioned to the WHATWG at the close of 2012, which now maintains a continuously evolving document utilizing Web IDL definitions.

== Operational Procedure == Generally, initiating a server transaction using XMLHttpRequest involves several distinct programmatic phases.

1. Instantiate an XMLHttpRequest object via its constructor call.
2. Invoking the "open" method to define the request methodology (GET, POST, etc.), specify the resource endpoint, and choose between synchronous or asynchronous execution.
3. For asynchronous operations, establishing an event listener callback function that will be triggered upon state transitions of the request.
4. Commencing the data transmission process by calling the "send" method, optionally including payload data.
5. Handling state changes within the designated event listener. Upon reception of server data, it is typically stored in the responseText attribute. When the object finishes processing the response, its state transitions to 4, signifying the "done" status. Beyond these fundamental steps, XMLHttpRequest provides numerous configuration levers to govern request transmission and response processing. Custom header fields can be appended to tailor server handling directives, and data payloads can be transferred during the "send" invocation. The received response can be subjected to JSON parsing for immediate object utilization or streamed progressively rather than awaiting complete buffer accumulation. Furthermore, requests possess the ability to be prematurely terminated or configured with a time-out deadline.

== Cross-Origin Transactions ==

During the nascent stages of the World Wide Web's evolution, the inherent security model prohibited direct data fetching betw