GitInsight AI

GitInsight AI is an intelligent assistant for analyzing GitLab projects and team performance. This project was implemented during the LLM-Agents Hackathon at Sharif University of Technology.

Screenshot

Architecture

GitInsight AI - Simplified Architecture

graph TD
    User[User] --> CL[Chainlit Interface]
    CL --> MA[Main Agent]
    
    MA --> Tools
    MA --> Analyzers
    
    subgraph Tools
        T1[Project & User Activity]
        T2[Metrics & Time]
    end

    subgraph Analyzers
        A1[Commit Analyzer]
        A2[Issue Analyzer]
        A3[MR Analyzer]
    end

Loading

Simple Component Overview

1. Interface Layer

   • User interacts through Chainlit chat interface

2. Main Agent

   • Coordinates all analysis activities
   • Manages tool execution

3. Tools & Analyzers

   • Activity tracking and metrics collection
   • Specialized analysis for commits, issues, and merge requests

How to Run

1. Install the dependencies:

   pip install -r requirements.txt

2. Run the application:

   python -m chainlit run chainlit_app.py

Environment Variables

To run this project, you will need to add the following environment variables to your .env file

• GITLAB_URL: The URL of your GitLab instance (e.g., https://gitlab.com).
• GITLAB_TOKEN: Your GitLab personal access token with api scope. This is used to authenticate with the GitLab API.
• GPT_API_KEY: Your API key for the GPT service (e.g., OpenAI).
• GPT_BASE_URL: The base URL for the GPT API. For OpenAI, this is typically https://api.openai.com/v1.

Contributors

• Mehrshad
• Hana