🤖 GitHub Chat Agent

The GitHub Chat Agent is a Streamlit-based application that allows users to interact with a GitHub repository in a conversational manner. It leverages Semantic Kernel and Azure OpenAI to provide intelligent responses and assist with repository queries. Additionally, it supports creating GitHub issues directly from the app.

---

🚀 Features

• Chat with GitHub Repositories: Query and retrieve information from a GitHub repository in a conversational format.
• Create GitHub Issues: Easily create issues in a repository with a user-friendly form.
• Streamlit Interface: Interactive and responsive UI for seamless user experience.
• Azure OpenAI Integration: Powered by Azure OpenAI for intelligent and context-aware responses.

---

📂 Directory Structure

git-chat-agent/
├── .devcontainer/          # Dev container configuration
│   ├── Dockerfile          # Dockerfile for the development container
│   ├── devcontainer.json   # Devcontainer configuration
├── samples/                # Contains concepts samples
├── tutorial/               # Notebook file for step by step process of creating the kernel with plugins
├── app.py                  # Main Streamlit application
├── git_plugin.py           # Plugin for interacting with GitHub repositories
├── requirements.txt        # Python dependencies
├── .env.example            # Example environment variables
├── .env                    # Environment variables (not included in the repo)

---

✨ Features in Detail

💬 Chat with GitHub Repositories

• Query repository details in a conversational format.
• View chat history and interact with the assistant.

🐞 Create GitHub Issues

• Fill out a form to create issues in the repository.
• Add labels and descriptions for better issue tracking.

🧩 git_plugin.py - GitHub Interaction Plugin

The git_plugin.py file provides functionality to interact with GitHub repositories programmatically. It includes features such as:

• Fetching repository details (e.g., branches, commits, pull requests).
• Searching for files or content within the repository.
• Creating and managing GitHub issues.
• Authenticating with GitHub using a Personal Access Token (PAT).

This plugin is a core component of the application, enabling seamless integration with GitHub.

---

📓 Jupyter Notebook: tutorial/semantic-kernel-chat-agent.ipynb

Purpose

The notebook provides an interactive environment to explore and experiment with the Semantic Kernel framework. It demonstrates how to use Semantic Kernel to build intelligent, context-aware chat agents and plugins for interacting with GitHub repositories.

Features

• Introduction to Semantic Kernel:
  • Overview of key concepts like Kernel, Plugins, Agents, Chat Completion, and Vector Stores.
• Chat Completion Service:
  • Demonstrates how to create and interact with a chat completion service using Azure OpenAI.
• Chat Completion Agent:
  • Explains the difference between a chat completion service and an agent, and how to create an agent for context-aware interactions.
• GitHub Plugin:
  • Shows how to create a plugin to interact with the GitHub API for tasks like retrieving repository details, managing issues, and analyzing commits.
• Practical Examples:
  • Includes examples of querying user profiles, describing repositories, listing issues, and analyzing commits.

---

🛠️ Setup Instructions

🧑‍💻 1. Development with Dev Containers

This repository includes a .devcontainer configuration for Visual Studio Code. To use it:

1. Open the repository in VS Code.
2. Install the Dev Containers extension.
3. Reopen the project in the container.

---

🧑‍💻 2. Local setup

2.1 Clone the Repository

git clone https://github.com/hannapureddy_microsoft/git-agent.git
cd git-chat-agent

2.2 Create a .env File

Create a .env file in the root directory by copying the .env.example file:

cp .env.example .env

Update the values in the .env file with your own credentials:

GLOBAL_LLM_SERVICE="AzureOpenAI"
AZURE_OPENAI_ENDPOINT=""
AZURE_OPENAI_API_KEY="YOUR_API_KEY"
AZURE_OPENAI_CHAT_DEPLOYMENT_NAME="gpt-4.1"
AZURE_OPENAI_API_VERSION="2024-12-01-preview"
GITHUB_PAT="YOUR_GITHUB_PERSONAL_ACCESS_TOKEN"

Follow these steps to create an Azure OpenAI endpoint and keys for Azure Open AI Model:

1. Create an Azure OpenAI Resource

1. Log in to the Azure Portal.
2. Search for Azure OpenAI in the search bar and select Azure OpenAI.
3. Click Create to start creating a new resource.
4. Fill in the required details:
   • Subscription: Select your Azure subscription.
   • Resource Group: Create a new resource group or select an existing one.
   • Region: Choose a supported region (e.g., East US, West Europe).
   • Name: Provide a unique name for your OpenAI resource.
5. Click Review + Create and then Create.

2. Deploy a model (ex:GPT-4.1)

1. Navigate to your newly created Azure OpenAI resource.
2. Go to the Model Deployments section in the left-hand menu.
3. Click Create to deploy a new model.
4. Select the Model(ex:GPT-4.1) model from the list.
5. Provide a Deployment Name (e.g., gpt-4.1) and configure the model settings as needed.
6. Click Deploy to start the deployment process.

3. Retrieve Endpoint and API Key

1. Once the deployment is complete, go to the Keys and Endpoint section in your Azure OpenAI resource.
2. Copy the Endpoint URL and API Key. These will be used in your .env file.

Reference Image

Below is an example of the Keys and Endpoint section in the Azure portal:

---

2.3 Install Dependencies

Install the required Python packages:

pip install -r requirements.txt

2.4 Run the Application

Start the Streamlit app:

streamlit run app.py

---

🐳 3.Using Docker

3.1 Build the Docker Image

docker build -f .devcontainer/Dockerfile -t git-chat-agent .

3.2 Run the Docker Container

docker run -p 8501:8501 --env-file .env git-chat-agent

Access the app at http://localhost:8501.

---

🙌 Acknowledgments

• Streamlit
• Semantic Kernel
• Azure OpenAI