quran_cloud_mcp_server logo

quran_cloud_mcp_server

by marwanWaly

The quran_cloud_mcp_server is an MCP (Machine Comprehension Platform) server designed to help LLMs (Large Language Models) access the Quran API (https://alquran.cloud/api) and reduce hallucination when working with Quran text. It enables models to retrieve only the necessary data, improving efficiency and accuracy.

View on GitHub

Last updated: N/A

quran_cloud_mcp_server

MCP server to help LLMs to get access to Quran API (https://alquran.cloud/api) to prevent the hallucination with Quran text.

hallucination is a big problem specially when you are working on sensitive data that each character is important.

one way of reducing the hallucination is by providing the context to your LLM but of course with large chunk of text like the holy Quran it's not efficient if you put all text in each request.

So, in this repo I have created an MCP server that's connect your LLM to a free API https://alquran.cloud/api that enables your model to retrieve only the data he needs.

Also, I will show to you how we can connect this MCP server to Claude desktop application.

Example of Claude the original response

Claude original response

Claude original response

Example of Claude the new response after connecting to Search-Quran MCP server

Claude New response

Claude New response

Installation

make sure you have python 3.13 & pip

Open your terminal and write these commands

Cloning

git clone https://github.com/marwanWaly/quran_cloud_mcp_server.git

Move to project directory

cd quran_cloud_mcp_server

Create virtual environment

python -m venv .venv

Activate venv

On Windows

.\venv\Scripts\activate

On Mac or linux

source .venv/bin/activate

Python packages installation

Use the package manager pip to install requirements.txt.

pip install -r requirements.txt

Create .env file

OPENAI_API_KEY=Your-secret-key

Run in terminal

python client.py

now you can directly chat with GPT4o in your terminal

server running in terminal

server running in terminal

Connect the server to Claude Desktop

Download Claude desktop and open it

Step 1

Select setting from the file menu

step 1

step 1

Step 2

Click on Developer then Edit Config

step 2

step 2

Step 3

Open claude_desktop_config.json

step 3

step 3

Step 4

Write this configuration in the file

{
  "mcpServers": {
    "Search-Quran": {
      "command": "python",
      "args": [
        "PROJECT_PATH_ON_YOUR_PC\\server.py"
      ],
      "host": "127.0.0.1",
      "port": 8080,
      "timeout": 30000
    }
  }
}

Don't forget to replace PROJECT_PATH_ON_YOUR_PC with the absolute path to your project server

step 4

step 4

Step 5

Restart Claude app (make sure it's completely closed from your taskbar by right click on Claude icon and select Quit)

Check if the new MCP has been added

step 5

step 5

Click on tools icon

step 6

step 6