Installation

Welcome to OmniMind! This Python library makes MCP (Model Context Protocol) and AI easy and fun for everyone. Whether you’re a coder, a complete beginner, a solopreneur, or a business owner, installing OmniMind is simple and quick—no tricky steps here! This guide walks you through everything you need to get started, and we’ve added ways for you to help make OmniMind a top project on GitHub—maybe even “repo of the day”!

Why Installing OmniMind is Easy

OmniMind is built to get you going fast. Here’s what makes it great:

Simple Steps: Just a few things to do, and you’re ready—no hard stuff to figure out.
Works on Any Computer: Windows, Mac, or Linux—it’s all good.
Free to Use: No cost, and it’s open-source, so you can try it risk-free.
Quick Setup: Takes just minutes, even if you’ve never coded before.
Helpful Tools Included: Comes with MCP servers like Fetch, Memory, and Filesystem—ready to go!

Let’s get started—it’s easier than you think!

What You’ll Need

Before we begin, you need two things. Don’t worry—we’ll make it simple!

Python: This is what runs OmniMind. Most computers have it, but we’ll check.
A Terminal: A place to type commands—like Command Prompt (Windows) or Terminal (Mac/Linux). It’s already on your computer!

Step 1: Check or Install Python

OmniMind needs Python to work, and it’s super easy to get.

For Non-Technical Users

Windows: Press Windows Key + R, type cmd, and hit Enter. In the black box, type python --version. If you see something like “Python 3.8.0,” you’re set! If not, download it from python.org. Click the big “Download Python” button, run the file, and check “Add Python to PATH” during setup.
Mac: Open Terminal (search “Terminal” in Spotlight), type python3 --version, and press Enter. If it shows a version, you’re good! If not, it’ll ask to install it—say yes and follow along.
Linux: Open Terminal, type python3 --version, and hit Enter. If it’s there, great! If not, type sudo apt install python3 (Ubuntu) or ask a friend to help.

Why It’s Great: Python is free, quick to install, and OmniMind works with any version 3.8 or higher—super flexible!

Step 2: Install OmniMind

Now that Python’s ready, let’s add OmniMind!

Open Your Terminal:
- Windows: Press Windows Key + R, type cmd, and hit Enter.
- Mac/Linux: Search “Terminal” and open it.
Type This Command:
```
pip install omnimind
```
Press Enter, and it’ll download OmniMind in seconds. If it says “pip is not recognized,” try:
```
python -m pip install omnimind
```
Check It Worked:
```
python -c "import omnimind; print('OmniMind is ready!')"
```
If you see “OmniMind is ready!”—you’re done!

Why It’s Great: One command gets you everything—no complicated downloads or setups. It’s fast and works every time!

Step 3: Set Up Your Google Gemini API Key

OmniMind uses Google Gemini for smart AI answers, and it needs a key to connect. Here’s how to do it—super simple!

Get Your Key:

Go to Google Cloud Console, sign in (or sign up—it’s free to start), and create a project (call it “OmniMind” if you like!). Search for “Gemini API,” enable it, and click “Create Credentials” to get an API key. Copy it—it’s a string like your-api-key-here.
Save It on Your Computer:
- Windows: In Command Prompt, type:
```
set GOOGLE_API_KEY=your-api-key-here
```
  and press Enter.
- Mac/Linux: In Terminal, type:
```
export GOOGLE_API_KEY=your-api-key-here
```
  and press Enter.
  
  This only lasts until you close the window—see “Make It Permanent” below if you want it to stick.
Make It Permanent (Optional):
- Windows: Search “Environment Variables” in the Start menu, click “Edit environment variables for your account,” add a new variable named GOOGLE_API_KEY, and paste your key.
- Mac/Linux: Open Terminal, type:
```
echo "export GOOGLE_API_KEY=your-api-key-here" >> ~/.bashrc
```
  (or ~/.zshrc for Mac), and restart Terminal.

Why It’s Great: This key connects OmniMind to Google Gemini for fast, smart answers—it’s free to try, and setup takes just a minute!

Step 4: Add a Config File

OmniMind needs a list of MCP servers (helpers) to work with. Let’s make one—it’s easy!

Create a File:

Open Notepad (Windows) or TextEdit (Mac), or any text editor. Paste this:

{
    "mcpServers": {
        "fetch": {
            "command": "uvx",
            "args": ["mcp-server-fetch"]
        },
        "memory": {
            "command": "C:\\Program Files\\nodejs\\npx.cmd",
            "args": ["-y", "@modelcontextprotocol/server-memory"],
            "env": {
                "MEMORY_FILE_PATH": "C:\\Users\\YourName\\Desktop\\memory.json"
            }
        },
        "filesystem": {
            "command": "C:\\Program Files\\nodejs\\npx.cmd",
            "args": ["-y", "@modelcontextprotocol/server-filesystem", "C:\\Users\\YourName\\Desktop", "C:\\Users\\YourName\\Desktop\\workspace"]
        }
    }
}

Change YourName to your actual username (e.g., C:\\Users\\John\\Desktop). Use / instead of \ on Mac/Linux (e.g., /home/john/Desktop). Save it as my_config.json on your Desktop.

Why These Servers?:
- fetch: Grabs web info—like summaries of pages.
- memory: Saves things—like your favorite color.
- filesystem: Works with files—like creating or listing them.
Need Node.js?:

The memory and filesystem servers use Node.js. If they don’t work, download it from nodejs.org—click the big green button, install it, and you’re set!

Why It’s Great: This file gets you started with useful helpers right away, and it’s simple to copy and paste—no coding needed yet!

Step 5: Try It Out

You’re ready! Let’s test OmniMind with a quick chat.

Make a File:

Open Notepad or TextEdit, paste this:
```
from omnimind import OmniMind

agent = OmniMind(config_path="C:\\Users\\YourName\\Desktop\\my_config.json")
agent.run()
```
Fix the path to match where you saved my_config.json (e.g., /home/john/Desktop/my_config.json on Mac/Linux). Save it as test.py on your Desktop.
Run It:

Open your terminal, type:
```
cd Desktop
python test.py
```
and press Enter. You’ll see “Ready! Type ‘quit’ to exit.” Type “What’s on example.com?” or “List files in my directory”—it’ll answer!
Have Fun:

Try more—check Quickstart for ideas like saving memories or fetching web summaries.

Why It’s Great: You’re using MCP and AI in minutes, it’s free, and it works for everyone—no experience required!

Help Make OmniMind Even Better

Love OmniMind? Here’s how you can help it grow and maybe get it trending on GitHub:

Star It: Click the ⭐ button at GitHub—it helps others find us!
Share It: Tell friends on X with #OmniMindAI or in groups—it’s a cool way to spread the word.
Join In: Have an idea? Post in GitHub Discussions or add your own servers—see Building a Custom MCP Server.
Say Hi: Email techiralthefuture@gmail.com with feedback—we’d love to hear from you!

Why It’s Great: Your support makes OmniMind a go-to tool for MCP and AI, and it’s fun to be part of something big!

Extra Tips

Stuck?: Check Frequently Asked Questions or ask in GitHub Discussions—we’re here to help.
More Servers?: Find tons of MCP servers at Awesome MCP Servers to add to your config.
Keep It Simple: Start with the basics—you can grow from there as you get comfy!

Why It’s Great: You’ve got all the help you need, it’s free to explore, and it’s made for everyone—no barriers here!

Ready to Go?

That’s it—OmniMind is installed and ready! It’s your easy way into MCP and AI with Python—fast, free, and fun. Try it out, share it, and help make it a top repo. Let’s make MCP simple for everyone together!