Metadata-Version: 2.1
Name: githubanalysis
Version: 0.1.5
Summary: A powerful tool for analyzing Git repositories using LLM-powered insights
Home-page: UNKNOWN
Author: Rimon
Author-email: info@rimon.com.au
License: UNKNOWN
Description: # GitHub Analysis Tool
        
        A powerful tool for analyzing Git repositories using LLM-powered insights. This tool provides detailed analysis of repository patterns, milestones, technical challenges, team dynamics, and more.
        
        ## What's New
        
        - The report now always includes at least **three technical challenges** identified and described in detail.
        - **All commits** (hash, author, message) are exported to a separate Excel file (`.xlsx`) alongside the main report.
        - **Commit references** in the technical challenges section are now clickable URLs, allowing you to directly view the referenced commit on GitHub/GitLab.
        - The **default report format is now Markdown** (if `--format` is not specified).
        
        ## Installation
        
        ```bash
        pip install githubanalysis
        ```
        
        ## Usage
        
        ### Basic Analysis
        
        ```bash
        githubanalysis https://github.com/username/repo.git
        ```
        
        ### Analysis with Date Range
        
        ```bash
        githubanalysis https://github.com/username/repo.git \
            --start-date 2023-01-01 \
            --end-date 2024-03-20
        ```
        
        ### OpenAI API Key Configuration
        
        You can provide your OpenAI API key in three ways:
        
        1. Command line argument:
        ```bash
        githubanalysis https://github.com/username/repo.git --openai-key your-api-key
        ```
        
        2. Environment variable:
        ```bash
        export OPENAI_API_KEY=your-api-key
        githubanalysis https://github.com/username/repo.git
        ```
        
        3. .env file:
        Create a `.env` file in your working directory:
        ```
        OPENAI_API_KEY=your-api-key
        ```
        
        ### Custom Prompts
        
        You can customize the analysis prompts by providing a JSON file:
        
        ```bash
        githubanalysis https://github.com/username/repo.git \
            --custom-prompts path/to/prompts.json
        ```
        
        ### Advanced Options
        
        ```bash
        githubanalysis https://github.com/username/repo.git \
            --start-date 2023-01-01 \
            --end-date 2024-03-20 \
            --format markdown \
            --output-dir custom_reports \
            --openai-key your-api-key \
            --model gpt-4 \
            --custom-prompts path/to/prompts.json
        ```
        
        ## Output
        
        The tool generates:
        
        - A comprehensive report in Markdown (default) or JSON format, including:
          - Key technical challenges (minimum three, with detailed analysis)
          - Milestones
          - Team and contributor analysis
          - Code and commit statistics
          - Clickable commit links for easy navigation
        - An **Excel file** with all commit hashes, authors, and messages for the analyzed period.
        
        Reports and Excel files are saved in the specified output directory (default: `reports/`).
        
        ## Requirements
        
        - Python 3.7+
        - Git
        - OpenAI API key (for LLM analysis)
        
        ## Dependencies
        
        - gitpython
        - openai
        - python-dotenv
        - argparse
        - tqdm
        - markdown
        - pandas
        - openpyxl
        - matplotlib
        - seaborn
        - scikit-learn
        - numpy
        - nltk
        - requests
        - tiktoken
Keywords: git,analysis,repository,llm,openai,gpt,commit,log,analysis
Platform: UNKNOWN
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Topic :: Software Development :: Version Control :: Git
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.7
Description-Content-Type: text/markdown
