Agile-bot is an intelligent SaaS platform designed for Agile project management through advanced automation and AI-driven insights. The platform automatically converts software requirement documents into actionable user stories, validates tasks, and optimizes sprint planning using state-of-the-art AI technologies.
This project consists of three main components:
- Backend: Django REST Framework API with Celery for asynchronous task processing
- Frontend: Next.js application with modern UI components
- LLM: AI processing module for document analysis and task generation
- Automated User Story Generation: Convert software requirement documents (PDF, DOCX) into precise user stories with acceptance criteria
- AI-Powered Task Validation: Ensure task accuracy and alignment with project goals using LLM-based validation
- Sprint Planning: Streamline project workflows with intelligent task allocation and capacity planning
- Dependency Detection: Automatically identify task dependencies and relationships
- Team Management: Assign tasks based on team composition and expertise
- WebSocket Support: Real-time updates for task generation progress
- Team Invitations: Email-based team member invitations with role management
- Live Progress Tracking: Monitor document processing and task generation in real-time
- Multi-format Support: Process PDF and DOCX requirement documents
- Vector Embeddings: Use Qdrant vector database for semantic search and context retrieval
- Intelligent Parsing: Extract requirements, user stories, and technical specifications automatically
- Framework: Django 5.0 with Django REST Framework
- Authentication: JWT with Google OAuth2 integration
- Task Queue: Celery with Redis broker
- WebSockets: Django Channels for real-time communication
- Storage: Azure Blob Storage (production) / Local filesystem (development)
- Database: PostgreSQL (production) / SQLite (development)
- Framework: Next.js 14 with React 18
- UI Components: Shadcn/ui with Radix UI primitives
- Styling: Tailwind CSS with custom animations
- Authentication: NextAuth.js with Google OAuth
- State Management: React hooks and context
- PDF Viewing: React PDF Viewer for document preview
- LLM Inference: Groq AI (Llama 3.2)
- Vector Database: Qdrant for semantic search
- Embeddings: Cohere for text embeddings
- Document Processing: Docling for document parsing
- Orchestration: Google Gemini for task validation
Agile-bot/
├── Backend/ # Django REST API and Celery workers
├── NFrontend/ # Next.js frontend application
├── LLM/ # AI processing and vector database
├── ScreenShots/ # Application screenshots
└── docker-compose.yml # Docker orchestration
- Docker and Docker Compose
- Python 3.10+
- Node.js 18+
- Redis (for local development)
- Clone the repository:
git clone https://github.com/yourusername/Agile-bot.git
cd Agile-bot- Set up environment variables:
# Copy and configure Backend environment
cp Backend/.env.production.template Backend/.env
# Copy and configure Frontend environment
cp NFrontend/.env.production.template NFrontend/.env.local- Start all services:
docker-compose up -d- Access the application:
- Frontend: http://localhost:3000
- Backend API: http://localhost:8000
- API Documentation: http://localhost:8000/api/docs/
See individual README files for detailed setup instructions:
Key environment variables in Backend/.env:
GEMINI_API_KEY: Required for task validationGOOGLE_OAUTH_CLIENT_ID/SECRET: For Google authenticationREDIS_URI: Redis connection for Celery and WebSocketsUSE_AZURE_STORAGE: Enable Azure Blob Storage for productionUSE_POSTGRESQL: Enable PostgreSQL for production
Key environment variables in NFrontend/.env.local:
NEXTAUTH_URL: Application URLNEXTAUTH_SECRET: Secret for session encryptionNEXT_PUBLIC_API_URL: Backend API endpointNEXT_PUBLIC_WS_URL: WebSocket endpointGOOGLE_CLIENT_ID/SECRET: Google OAuth credentials
The project includes production-ready configurations:
docker-compose.prod.yml: Production Docker setupdeploy-celery-azure.sh: Azure deployment script- Azure Blob Storage integration
- PostgreSQL database support
- Gunicorn WSGI server
- Nginx reverse proxy ready
The project includes automated deployment workflows:
.github/workflows/azure-deploy.yml: Automatic deployment on push to main/newDesgin.github/workflows/manual-deploy.yml: Manual deployment trigger
Required GitHub Secrets:
AZURE_CREDENTIALS: Azure service principal credentialsACR_USERNAME: Azure Container Registry usernameACR_PASSWORD: Azure Container Registry passwordNEXTAUTH_SECRET: NextAuth.js secret for session encryption (generate withopenssl rand -base64 32)
- Configure Azure resources (App Service, Blob Storage, PostgreSQL)
- Update production environment variables
- Run deployment script:
./deploy-celery-azure.shInteractive API documentation is available at:
- Swagger UI: http://localhost:8000/api/docs/
- ReDoc: http://localhost:8000/api/redoc/
- Upload requirement document (PDF/DOCX)
- AI processes document and extracts requirements
- System generates user stories with:
- Title and description
- Acceptance criteria
- Story points estimation
- Priority level
- Dependencies
- Review and edit generated stories
- Organize into sprints
- Automatic sprint capacity calculation
- Task allocation based on team composition
- Dependency-aware scheduling
- Story point tracking
- Sprint velocity metrics
- Create and manage teams
- Invite members via email
- Role-based access control
- Team capacity planning
- Member expertise tracking
| Title | Screenshot |
|---|---|
| Landing Page |  |
| Projects List |  |
|