StockAnalyzer

StockAnalyzer is an open-source, cross-platform desktop application for stock analysis, technical charting, screening, and portfolio management. It is built with .NET 8 and Avalonia UI, leveraging Python to provide advanced analytical services.

---

Supported Platforms

OS	Architecture	Python Setup
Windows 10 / 11	x64	Automatic (embedded Python installer included)
macOS (Intel & Apple Silicon)	x64 / ARM64	Manual (requires system Python 3.8+ and configuration)
Linux (Ubuntu / Debian-based)	x64	Manual (requires system Python 3.8+ and configuration)

---

Requirements

• .NET SDK: 8.0 or later
• Python: 3.8 to 3.11 (required for data retrieval and indicators)
  • Core libraries: yfinance, pandas, polars, pyarrow, onnxruntime.

---

Project Structure

StockAnalyzer/
├── StockAnalyzer.Avalonia/  # Desktop UI (Views, ViewModels, Assets)
├── StockAnalyzer.Core/      # Business logic, math, and Named Pipe IPC
├── StockAnalyzer.Python/    # Python scripts for data ingestion and indicators
└── Tests/                  # Unit, Integration, and UI Automation tests

---

Getting Started

1. Clone the Repository

git clone https://github.com/akuro17/StockAnalyzer.git
cd StockAnalyzer

2. Configuration

Windows users can run the app with default settings out of the box. For custom paths or non-Windows environments, you need an active configuration file:

# Copy the example configuration
cp StockAnalyzer.Avalonia/appsettings.example.json StockAnalyzer.Avalonia/appsettings.json

Important

appsettings.json is ignored by Git to keep your local paths secure.

3. Python Environment

Option A: Windows (Automatic)

No manual setup is needed. The app automatically initializes Python and installs dependencies on the first run. (The first launch may take a few minutes to run pip install in the background).

Option B: macOS / Linux / Custom Paths (Manual)

You need to create a virtual environment, install dependencies, and update your configuration.

# Create and activate a virtual environment
python3 -m venv .venv
source .venv/bin/activate

# Install requirements
pip install -r StockAnalyzer.Python/requirements.txt

Next, update StockAnalyzer.Avalonia/appsettings.json with the absolute path to the python executable in your virtual environment:

"Python": {
  "PythonPath": "/absolute/path/to/your/StockAnalyzer/.venv/bin/python3"
}

4. Run the Application

dotnet build StockAnalyzer.sln
dotnet run --project StockAnalyzer.Avalonia/StockAnalyzer.Avalonia.csproj

---

Workspace Layout

On the first launch, the app creates a Data/ directory to store your workspace files:

Data/
├── Daily/      # Daily historical data (Parquet)
├── Weekly/     # Weekly aggregated data
├── Monthly/    # Monthly aggregated data
├── Metadata/   # Symbol info and metadata
├── Config/     # User settings and UI state
└── Portfolios/ # Portfolio data

---

Testing

StockAnalyzer uses xUnit for C# and pytest for Python.

C# Tests

# Run unit and integration tests (excludes UI tests)
dotnet test Tests/StockAnalyzer.Tests.csproj --filter "FullyQualifiedName!~UITests"

Note

UI Tests are excluded by default as they require a GUI and are generally not suitable for headless CI environments.

Python Tests

cd StockAnalyzer.Python
pytest

---

FAQ & Troubleshooting

• Windows SmartScreen "Windows protected your PC" warning when running the executable. This warning appears because the pre-built binaries generated on GitHub Actions are not digitally signed with a code signing certificate, which can be costly to obtain. To run the app, click "More info" on the warning dialog, then click "Run anyway". The warning will not appear again on subsequent launches of the same file.
• Why does the first launch take longer on Windows? The app is downloading a portable Python environment and installing packages. Subsequent startups are immediate. Check the logs/ directory if you want to monitor the pip output.
• Can I use my own Python installation? Yes. Update the PythonPath in appsettings.json to point to your preferred Python 3.8-3.11 binary.
• What is trend_predictor.onnx? This is the machine learning model used for predictions. A dummy.onnx is provided to allow the app to compile and start, but actual predictions will fail until you replace it with a properly trained model.
• Python processes hang after a crash. If the app closes unexpectedly, kill lingering Python processes manually (Stop-Process -Name python -Force on PowerShell, or pkill -f update_pipeline.py on Bash).

---

Acknowledgments & Licenses

This project relies on several key libraries:

• Avalonia UI (MIT)
• DuckDB.NET (MIT)
• ONNX Runtime (MIT)
• yfinance (Apache 2.0)

See THIRD_PARTY_NOTICES.md for a complete list and license texts.

---

Disclaimer (Yahoo Finance API)

Warning

This software is provided "as-is". It connects to third-party endpoints using yfinance, an unofficial Yahoo Finance client.

• Use at your own risk. The maintainers are not responsible for service interruptions, IP blocks, or Terms of Service violations.
• This tool is intended for personal, educational, or research purposes. Complying with Yahoo Finance's API limits and Terms of Service is the sole responsibility of the user.

---

AI Contribution Notice

Important

StockAnalyzer is primarily developed and maintained using AI coding assistants. As such, it does not have a traditional team of human maintainers. Users are encouraged to utilize AI coding assistants themselves to troubleshoot issues, fix bugs, or implement new features locally.