CSCE4350_gradebot

README

CSCE 4350 Gradebot

Automated code grading system for CSCE 4350 assignments.

Features

• Server Mode: HTTP server for receiving and grading code submissions
• Client Mode: CLI tool for submitting assignments for grading
• OpenAI Integration: Uses GPT-4o Mini for code analysis and feedback
• Web Interface: HTML dashboard for viewing submissions and grades
• Cross-Platform: Native binaries for Linux, macOS, and Windows
• Koyeb Deployment: Optimized for cloud deployment

Architecture

Overview

The gradebot consists of:

• Server: HTTP API server handling grading requests
• Client: CLI tool for submitting assignments
• Rubrics: Evaluation logic and test runners

Local Development

Setup

1. Clone the repository:

   git clone https://github.com/jh125486/CSCE4350_gradebot.git
   cd CSCE4350_gradebot
   

2. Set up environment variables:

   # Copy the example environment file
   cp .env.example .env
   
   # Edit .env with your actual secrets
   nano .env  # or your preferred editor
   

3. Initialize development environment:

   go mod tidy
   make init    # Installs git hooks
   make build
   

   The make init command installs a pre-push hook that runs tests and linting before allowing a push.

Testing Locally

The application automatically loads environment variables from a .env file when running locally.

Start the server:

# Using Makefile
make local-test

# Or manually
./bin/gradebot server --port 8080

Test the client:

# Submit a project for grading
./bin/gradebot project1 --dir /path/to/your/project --run "python main.py"

Environment Variables

The following environment variables are required for full functionality:

• OPENAI_API_KEY: Your OpenAI API key
• BUILD_ID: Unique build identifier for authentication
• R2_ENDPOINT: Cloudflare R2 endpoint URL
• AWS_ACCESS_KEY_ID: R2 access key
• AWS_SECRET_ACCESS_KEY: R2 secret key

Optional variables:

• R2_BUCKET: Custom bucket name (defaults to "gradebot-storage")
• AWS_REGION: AWS region (defaults to "auto")
• USE_PATH_STYLE: Use path-style S3 URLs (for LocalStack testing)

Development Workflow

Run tests:

make test          # Run all tests with race detection
make test-verbose  # Run tests with verbose output

Run linting:

make lint          # Run golangci-lint and security checks

Git Hooks: The pre-push hook automatically runs before each git push to ensure:

• All tests pass (including race detection)
• Code passes all linting checks

To bypass the hook (not recommended):

git push --no-verify

Usage

Submit assignments for grading:

# Project 1
./gradebot project1 --dir /path/to/project --run "python main.py"