OmniMind

OmniMind is an open-source Python library that implements the Model Context Protocol (MCP) to connect AI Agents with MCP Servers and run AI workflows, tools, and automations. It provides a plug-and-play experience with a ready-to-use set of tools, including Terminal access, web Fetch capabilities, Memory storage, and Filesystem interactions, all accessible through a simple Python interface. The library is designed to be easy to drop into Python projects and to be extended with custom MCP servers, making it suitable for developers building AI tools, automation pipelines, or educational experiments with MCP.

To use OmniMind, install the package and instantiate the OmniMind client in your Python code. You can start an agent, connect it to MCP servers, and load or add servers as needed. For example, after installation you typically create an OmniMind instance and call run to begin interacting with the configured MCP servers and tools. The project ships with examples showing how to add your own servers and customize the agent’s behavior, so you can tailor it to your workflow or automation needs.

Prerequisites:

• Python 3.8+ (recommended to use a virtual environment)
• Internet access to install dependencies

Step-by-step installation:

1. Create and activate a virtual environment (optional but recommended):

python -m venv venv
# Windows
venv\Scripts\activate.bat
# macOS/Linux
source venv/bin/activate

2. Install OmniMind from PyPI:

pip install omnimind

3. (Optional) Verify installation by importing and printing version:

python -c "from omnimind import __version__; print(__version__)"

4. Run a quick demo or integrate into your project by importing OmniMind and creating an instance as shown in the examples.

Environment variables and configuration options:

• OmniMind is designed to be extended with custom MCP servers. You can add servers programmatically using agent.add_server(name, command=..., args=...).
• If you use external tools or APIs, ensure proper authentication tokens and secrets are managed securely (consider using environment variables or a secrets manager).
• When running multiple MCP servers, you can order or prioritize them by how you add them to the OmniMind instance and by how you route prompts to specific servers.
• Common issues include missing dependencies for optional features or network restrictions when connecting to remote MCP servers. Ensure your Python environment has network access and that required dependencies are installed.
• Documentation and examples are available under docs/ and examples/ in the repository for deeper dives into server configuration and usage.