System Requirements & Dependencies

Cleo is built with flexibility in mind and is designed to operate across a variety of environments, including local development machines, private servers, and containerized platforms.

This section outlines the minimum and recommended requirements to run Cleo effectively, along with a breakdown of its core and optional dependencies.

---

Minimum System Requirements

Resource	Minimum
OS	macOS, Linux, or Windows (WSL)
Python Version	3.9 or higher
RAM	4 GB
Disk Space	500 MB
Network Access	Required for external APIs

Note: For Windows users, it is highly recommended to use Windows Subsystem for Linux (WSL) for better compatibility.

---

Recommended Development Environment

Resource	Recommended
Python Version	3.10+
RAM	8 GB or more
Docker	Optional, for isolated builds
Virtualenv	For dependency management
Git	Latest version for source control

---

Python Dependencies

The base dependencies required to run Cleo are listed in the requirements.txt file:

textCopyEditopenai>=1.0.0
requests>=2.31.0
python-dotenv>=1.0.0

Install them via pip:

bashCopyEditpip install -r requirements.txt

These dependencies enable foundational functionality such as task routing, API integration, and environment configuration.

---

Optional Dependencies (Advanced Features)

To enable advanced agent capabilities, install the following modules based on your intended use case:

Feature	Dependency	Install Command
Vector Memory	faiss-cpu or chromadb	pip install faiss-cpu
Web Browsing Agent	playwright	pip install playwright
HTML Parsing	beautifulsoup4	pip install beautifulsoup4
LLM Chaining	langchain	pip install langchain
Speech I/O	pyttsx3, SpeechRecognition	pip install pyttsx3 SpeechRecognition

Example:

bashCopyEditpip install faiss-cpu langchain beautifulsoup4

To install all optional dependencies at once:

bashCopyEditpip install -r requirements-extended.txt

requirements-extended.txt is a convenience file provided in the repo for full feature installations.

---

GPU Support (Optional)

If you're planning to run local language models or neural inference, GPU support can be enabled via:

• CUDA-compatible GPU

• PyTorch with GPU support

• Accelerated backends for Faiss or Hugging Face Transformers

These are not required for cloud-based models (like OpenAI), but are useful for self-hosted inference.

---

Environment Variables

Configuration values such as API keys and runtime preferences should be placed in a .env file at the project root.

Example .env:

iniCopyEditOPENAI_API_KEY=your_api_key
CLEO_AGENT_NAME=Atlas
CLEO_ENABLE_MEMORY=True

Environment variables are automatically loaded by Cleo on startup using python-dotenv.

---

Dependency Management Best Practices

• Always use a virtual environment for isolation:

bashCopyEditpython -m venv venv
source venv/bin/activate

• Freeze current dependencies if modifying:

bashCopyEditpip freeze > requirements.txt

• For production, pin versions strictly to avoid upstream changes.