pycaretaker 0.1.0

A smart Python package manager companion — manage, monitor, visualize, and analyze your environment with optional AI insights.

🛡️ PyCareTaker v0.1

A smart Python package manager companion — manage, monitor, visualize, and analyze your Python environment with optional AI insights.

Not just another pip wrapper, but a developer assistant.

---

✨ Features

Core Features

Command	Description
install <pkg>	Install a package and auto-update requirements.txt
remove <pkg>	Uninstall a package
deps	Visualize dependency graph (networkx + matplotlib)
outdated	Compare installed versions against PyPI latest
diff	Compare current env vs saved requirements.txt
memwatch	Live memory + CPU profiler with dual graphs
security	Vulnerability scan (pip-audit + PyPI + AI)
plugins	Run custom user plugins
interactive	Interactive shell with all commands

AI Features (optional)

Command	Description
ai "Add a web framework"	Natural language → pip actions
Auto-recommendations	AI suggests related packages after install
Security insights	AI analyzes CVE databases
Profiling analysis	AI interprets memory/CPU graphs

---

🚀 Quick Start

# Install dependencies
pip install networkx requests packaging colorama psutil matplotlib

# Run help
python package_manager.py --help

# Examples
python package_manager.py deps                          # Dependency graph
python package_manager.py deps --text                   # Text tree
python package_manager.py outdated                      # Version check
python package_manager.py diff                          # Env diff
python package_manager.py memwatch                      # CPU + Memory profiler
python package_manager.py memwatch --export usage.csv   # Export on exit
python package_manager.py security                      # Vulnerability scan
python package_manager.py plugins                       # Run user plugins
python package_manager.py interactive                   # Interactive mode

With AI

# Using OpenAI API
python package_manager.py --api-key sk-... ai "Add a web framework"

# Using local Ollama (auto-detected)
python package_manager.py ai "Remove all testing libraries"

# Environment variable
export OPENAI_API_KEY=sk-...
python package_manager.py ai "What packages do I need for data science?"

---

🔌 Plugin System

Drop .py scripts in the plugins/ folder. Each must define:

def run(context: dict) -> None:
    packages = context["installed_packages"]  # {name: version}
    # Your custom logic

See pycaretaker/plugins/examples/license_check.py for a reference.

---

📁 Project Structure

PyCareTaker/
├── package_manager.py          # Entry point
├── pycaretaker/
│   ├── cli.py                  # CLI dispatcher
│   ├── core/                   # Non-AI features
│   │   ├── packages.py         # Install/remove/freeze
│   │   ├── monitor.py          # Background change detection
│   │   ├── deps.py             # Dependency visualization
│   │   ├── outdated.py         # PyPI version tracking
│   │   ├── diff.py             # Environment diffing
│   │   ├── profiler.py         # Memory + CPU profiler
│   │   └── export.py           # CSV/JSON export
│   ├── ai/                     # AI features (optional)
│   │   ├── backend.py          # Flexible LLM backend
│   │   ├── nlp_commands.py     # Natural language → actions
│   │   ├── recommendations.py  # Package suggestions
│   │   ├── security.py         # CVE/vulnerability analysis
│   │   └── analysis.py         # Profiling interpretation
│   └── plugins/                # Plugin loader + examples
├── plugins/                    # User plugin drop folder
├── tests/                      # Unit tests
└── requirements.txt

---

🧪 Running Tests

python -m pytest tests/test_core.py -v

---

📄 License

MIT