Apex

AI-Powered Project Requirements Management Platform

Apex is a full-stack application that leverages AI agents and Retrieval-Augmented Generation (RAG) to help teams manage project requirements, documents, and tasks intelligently. Upload your project documents, and let Apex automatically extract requirements, identify conflicts, and answer questions about your project.

Key Features

• AI-Powered Document Analysis - Upload PDFs, Word documents, and text files. Apex automatically extracts, chunks, and embeds content for semantic search.
• Intelligent Q&A Chat - Ask questions about your project documents and get AI-generated answers grounded in your uploaded content with source citations.
• Requirements Extraction - Automatically identifies functional and non-functional requirements from your documents.
• Tech Stack Detection - AI identifies and categorizes technology stacks mentioned in your project documentation.
• Conflict Detection - Detects conflicting requirements and flags them by severity (low, medium, high).
• Missing Information Tracking - Identifies gaps in your project documentation that need to be addressed.
• Task Management - Create and track project tasks with origin tracking (user-created vs. AI-suggested).
• Progress Tracking - Monitor project progress based on completed tasks.
• Real-time Streaming - Chat responses stream in real-time using Server-Sent Events (SSE).

Tech Stack

Frontend

• React 18 with TypeScript
• Vite for fast development and building
• React Router 6 for client-side routing
• Zustand for state management
• Tailwind CSS for styling
• Radix UI for accessible UI components
• Better Auth (React client) for authentication

Backend

• Node.js with Express.js
• TypeScript for type safety
• Prisma ORM with MongoDB
• LangGraph for AI agent orchestration
• Google Generative AI for LLM capabilities
• MongoDB Atlas Vector Search for semantic search
• Better Auth for authentication
• Multer for file uploads

AI & Document Processing

• LangGraph - State machine-based AI agent workflows
• Google GenAI - Text embeddings and LLM responses
• pdf.js - PDF text extraction
• Mammoth - Word document (.docx) text extraction
• LangChain Text Splitters - Intelligent document chunking

Project Structure

apex/
├── frontend/          # React + Vite frontend application
│   ├── src/
│   │   ├── components/    # Reusable UI components
│   │   ├── pages/         # Page components
│   │   ├── services/      # API service functions
│   │   ├── stores/        # Zustand state stores
│   │   └── lib/           # Utilities and auth client
│   └── ...
├── backend/           # Express.js API server
│   ├── src/
│   │   ├── agents/        # LangGraph AI agents
│   │   │   ├── graphs/    # Agent workflow definitions
│   │   │   ├── nodes/     # Individual agent nodes
│   │   │   └── tools/     # Agent tools
│   │   ├── modules/       # Feature modules (projects, tasks, etc.)
│   │   ├── config/        # Configuration files
│   │   ├── middlewares/   # Express middlewares
│   │   └── utils/         # Utility functions
│   ├── prisma/
│   │   └── schema.prisma  # Database schema
│   └── ...
└── README.md          # This file

Getting Started

Prerequisites

• Node.js 18+
• MongoDB (local or MongoDB Atlas)
• Google AI API Key (for Generative AI features)

Installation

1. Clone the repository

   git clone https://github.com/yourusername/apex.git
   cd apex

2. Set up the backend

   cd backend
   npm install

3. Configure environment variables

   Create a .env file in the backend directory:

   DATABASE_URL="mongodb+srv://username:[email protected]/apex"
   PORT=5000
   BASE_URL="http://localhost:5000"
   NODE_ENV="development"
   CORS_ORIGIN="http://localhost:5173"
   BETTER_AUTH_SECRET="your-secret-key-min-32-characters"
   GOOGLE_API_KEY="your-google-ai-api-key"

4. Generate Prisma client and push schema

   npm run prisma:generate
   npm run prisma:push

5. Set up the frontend

   cd ../frontend
   npm install

6. Configure frontend environment

   Create a .env file in the frontend directory:

   VITE_API_URL="http://localhost:5000"

Running the Application

1. Start the backend server

   cd backend
   npm run dev

   The API server will start at http://localhost:5000

2. Start the frontend development server

   cd frontend
   npm run dev

   The frontend will be available at http://localhost:5173

API Overview

Endpoint	Description
/api/auth/*	Authentication endpoints (Better Auth)
/api/projects	Project CRUD operations
/api/project/:id/documents	Document upload and management
/api/project/:id/data	Project requirements, tech stacks, conflicts
/api/assistant	AI chat assistant (SSE streaming)
/api/tasks	Task management
/api/project/:id/progress	Project progress tracking

AI Agent Workflows

Apex uses LangGraph to orchestrate AI agents for various tasks:

Information Upload Pipeline

1. Extract Text - Parse uploaded documents (PDF, DOCX, TXT)
2. Summarize - Generate document summary and extract requirements
3. Chunk Text - Split into semantic chunks
4. Embed Chunks - Generate vector embeddings
5. Store - Save to MongoDB with vector indexes

Query Pipeline

1. Decide Source - Determine if query needs database or vector search
2. Vector Search - Semantic similarity search on document chunks
3. Database Query - Structured data retrieval
4. Summarize Response - Generate final answer with sources

License

This project is licensed under the ISC License.

---

For detailed documentation on each component, see:

• Frontend Documentation
• Backend Documentation