| title | Deskora |
|---|---|
| emoji | 🎧 |
| colorFrom | indigo |
| colorTo | purple |
| sdk | docker |
| pinned | false |
Deskora is a sophisticated, AI-driven customer support platform designed to transition traditional ticketing systems into high-performance, real-time communication hubs. Built using the MERN stack and integrated with Google Gemini AI, Deskora automates complex support workflows, provides deep analytical insights, and offers a premium user experience through modern design principles.
- Automated Triage: Tickets are automatically classified into Billing, Technical, or General categories using advanced NLP.
- Sentiment Analysis: Real-time detection of user sentiment to prioritize urgent or high-frustration cases.
- Intelligent Routing: Tickets are dynamically assigned to agents based on their specific category expertise and current workload.
- AI Knowledge Synthesis: Users can query the Knowledge Base via a natural language "Ask AI" interface for immediate, synthesized answers.
- Professional Article Library: Includes over 25 pre-authored support articles covering multiple domains.
- Hierarchical Organization: Seamless filtering by category, including Getting Started, Technical Support, Billing, and General FAQ.
- Global Accessibility: Multi-language support through AI-powered translation frameworks.
- Role-Based Data Views: Personalized dashboards for Users, Agents, and Admins.
- Real-time Monitoring: Visual tracking of ticket volume, resolution rates, and category distribution via Recharts.
- Workload Optimization: Admin-level visibility into agent capacity and average response times.
- Comprehensive Profiles: Support for detailed professional metadata including Job Titles, Company Names, and Phone Numbers.
- Real-time Communication: Instant messaging capabilities between agents and customers powered by Socket.io.
- Media Integration: Seamless management of screenshots and technical logs via Cloudinary integration.
- Unified Design System: A premium glassmorphism interface with fully synchronized light and dark modes.
- Frontend: React.js, Lucide Icons, Recharts, Socket.io-client.
- Backend: Node.js, Express.js.
- Database: MongoDB (Atlas).
- AI Engine: Google Gemini 1.5 Pro.
- Media: Cloudinary.
- Real-time: Socket.io.
Create a .env file in the /backend directory with the following variables:
| Variable | Description |
|---|---|
| PORT | The server port (e.g., 5000) |
| MONGO_URI | MongoDB Atlas connection string |
| JWT_SECRET | Secure string for authentication tokens |
| GEMINI_API_KEY | Google AI Studio API Key |
| CLOUDINARY_CLOUD_NAME | Cloudinary Cloud Identifier |
| CLOUDINARY_API_KEY | Cloudinary API Key |
| CLOUDINARY_API_SECRET | Cloudinary API Secret |
git clone https://github.com/Nikhitha-spec/deskora.git
cd deskoracd backend
npm install
# Populate .env with necessary credentials
# Seed the database with initial KB articles and test users
node seed.js
# Start the production-ready server
node server.jscd frontend
npm install
# Launch the development environment
npm run devDeskora is pre-configured for multi-service deployment using Vercel's experimentalServices architecture.
- Link the repository to your Vercel account.
- Set the Framework Preset to "Services" in Project Settings.
- Configure all environment variables identified in the Configuration section within Vercel's Environment Variables dashboard.
- Execute a full redeploy to initialize the joint frontend and backend services.
The seeding process (node seed.js) generates several test accounts for role-based exploration. The default password for all pre-seeded accounts is password123.
- Administrator: admin@deskora.com
- Billing Specialist: billing@deskora.com
- Technical Engineer: tech@deskora.com
- General Support: general@deskora.com
- Standard Customer: user@example.com
- Administrators: Full system governance, global analytics, and user management.
- Agents: Specialized ticket handling, AI co-pilot assistance, and workload tracking.
- Users: Knowledge base access, self-service AI queries, and ticket lifecycle management.