QUANT-IQ

A quantitative analysis tool for intelligent financial review.

About

QUANT-IQ analyzes financial statements using the Anthropic Claude API and presents results through a Streamlit web interface. Users upload financial documents in PDF, DOCX, or ZIP format, and the tool generates structured reports covering revenue growth, profitability, liquidity, and overall financial health. Reports are scored on a standardized rubric and exported as PDF.

Features

• Upload and analyze financial documents in PDF, DOCX, CSV, XLSX, or ZIP format
• Bulk processing of multiple documents, individually or grouped by directory
• Two analysis modes: standard analysis or prompt-customized analysis (structured outputs)
• Editable analysis prompt with save, restore, and download support
• Automated PDF report generation with styled HTML output
• Downloadable ZIP archive of all analysis results
• Customizable logo for branded reports
• Deployable locally or via Docker

Getting Started

Prerequisites

• Python 3.10+
• An Anthropic API key
• wkhtmltopdf or WeasyPrint system dependencies (used for HTML-to-PDF conversion)

Installation

git clone https://github.com/sieverett/QuantIQ.git
cd QuantIQ
pip install -r requirements.txt

Configuration

Create a .env file in the project root:

ANTHROPIC_API_KEY=your_anthropic_api_key

Or set the key through the Settings page in the app after launch.

Run

streamlit run app.py

Docker

docker build -t quantiq-app .
docker run -d -p 8501:8501 --name quantiq-app quantiq-app

Then open http://localhost:8501 in your browser.

Usage

1. Set your Anthropic API key in the Settings tab.
2. Upload financial documents through the Analyze tab.
3. Click Analyze to generate reports.
4. Download individual reports or a ZIP archive of all results.

Project Structure

QuantIQ/
├── app.py                  # Streamlit application entry point
├── components/             # UI components (sidebar, analyzer, settings, prompt editor)
├── quantiq/                # Core logic (analysis, reporting, file handling, logging)
├── prompts/                # Prompt templates and output format definitions
├── utils/                  # Session and auth utilities
├── styles/                 # Custom CSS
├── imgs/                   # Logo assets
├── templates/              # HTML report template
├── Dockerfile
└── requirements.txt

License

This project is licensed under the MIT License.