Pearl MCP

How to Install and Use Pearl MCP

Pearl MCP is a powerful server that connects AI and expert services through a simple interface. If you want to use Pearl MCP locally or with MCP clients, this guide will help you install it and run it with easy-to-follow steps and example code.

Prerequisites Before Installation

Before you start installing Pearl MCP, you need a few things ready. Pearl MCP requires Python 3.12 or higher to be installed on your computer. You also need a Pearl API key, which you can get by contacting Pearl through their website. Finally, make sure you have pip (a Python package installer) or uv package manager to handle dependencies.

Step 1: Download Pearl MCP

First, you need to get the Pearl MCP server files on your computer:

git clone https://github.com/Pearl-com/pearl_mcp_server.git
cd pearl_mcp_server

This command downloads the server code and moves into the folder to prepare for the next steps.

Step 2: Set Up the Python Environment

Next, create a special Python environment for Pearl MCP to keep everything organized:

python -m venv .venv
source .venv/bin/activate  # On Windows use: .venv\Scripts\activate

This creates and activates a virtual environment, meaning all the files you install won’t affect other programs.

Step 3: Install the Needed Packages

Now, you install all the software Pearl MCP needs to work properly:

pip install -e .

This command reads the project settings and adds all the required tools to your environment.

Step 4: Configure the Server

Before running the server, you have to add your Pearl API key. Create a file called .env inside the src directory of the project with this content:

PEARL_API_KEY=your-api-key-here

Replace your-api-key-here with the actual API key you received from Pearl.

Step 5: Run the Pearl MCP Server Locally

You can start the server on your computer using two options. The default way uses stdio transport:

pearl-mcp-server --api-key your-api-key

If you want to use a network connection with Server-Sent Events (SSE) and a custom port, use:

pearl-mcp-server --api-key your-api-key --transport sse --port 8000

Again, replace your-api-key with your actual key.

Example: Using Pearl MCP with Python Client

To connect your Python code to Pearl MCP, you can write a simple script like this:

import asyncio
from mcp.client.session import ClientSession
from mcp.client.stdio import StdioServerParameters, stdio_client

async def main():
    async with stdio_client(
        StdioServerParameters(command="pearl-mcp-server", args=["--api-key", "your-api-key"])
    ) as (read, write):
        async with ClientSession(read, write) as session:
            await session.initialize()

            tools = await session.list_tools()
            print(tools)

            result = await session.call_tool(
                "ask_pearl_ai",
                {
                    "question": "What is MCP?",
                    "session_id": "optional-session-id"
                }
            )
            print(result)

asyncio.run(main())

This script starts a session, lists available tools, and asks the “What is MCP?” question, then prints the answer.

---

By following these steps, you will have Pearl MCP installed, configured, and ready to use with your applications. It is simple and gives you access to AI and expert services from Pearl through a standard interface.