Production-grade AI conversational interface built to simulate modern LLM-powered SaaS tools.
CopilotUI is a system-level frontend engineering showcase designed to demonstrate conversational UX, streaming responses, markdown rendering, and AI copilot interface design.
This project is part of the GigDevOS capability lab — created to replace NDA-protected AI product work with anonymized capability demonstrations.
CopilotUI replicates the interface experience of modern AI products such as copilots, assistants, and knowledge tools.
It demonstrates how users interact with conversational AI systems across workflows like coding, documentation, analytics, and product operations.
The project focuses on:
- Conversational UI architecture
- Streaming response systems
- Markdown rendering
- Code block visualization
- Prompt interaction design
Many AI product interfaces cannot be publicly shared due to NDAs.
CopilotUI exists to demonstrate:
- AI SaaS frontend capability
- Real-time conversational UX
- LLM interface engineering
- Developer tooling UI patterns
It serves as a flagship demo for AI-focused freelance and gig positioning.
- User & AI message bubbles
- Avatar indicators
- Timestamp labels
- Message grouping
- Edit & regenerate messages
- Like/dislike feedback
- Copy & share actions
- Simulated real-time generation
- Word-by-word streaming
- Typing indicator states
AI responses support:
- Headings
- Lists
- Tables
- Links
- Rich formatting
- Syntax highlighting (Prism)
- Language detection
- Copy-to-clipboard
- Scrollable blocks
- Auto-resizing textarea
- Keyboard submission (Enter to send, Shift+Enter for newline)
- Prompt suggestions
- File attachments (images, PDF, text, etc.)
- Drag & drop support
- Chat history
- New chat creation
- Rename threads
- Delete threads
- Favorites / pinned chats
- Search conversations
- Collapsible on mobile
- Profile — User info, display name, email
- Settings — Theme, notifications, cookies, data controls, export, account deletion
- Subscription — Plan selection (Free, Pro, Team)
- Help & Support — FAQ accordion
| Category | Stack |
|---|---|
| Framework | Next.js 16 (App Router) |
| Language | TypeScript |
| UI | React 19 |
| Styling | Tailwind CSS 4 |
| Markdown | react-markdown, remark-gfm |
| Code | react-syntax-highlighter (Prism) |
| Icons | Lucide React |
| Deployment | Vercel |
Note: LLM integration is mocked — no real API required. Responses use realistic mock data.
CopilotUI follows a modern AI product interface language.
- Light/dark theme toggle
- Conversational canvas
- Glass input composer
- Gradient AI responses
- Sidebar workspace layout
Built to feel production-ready — not a chatbot widget.
/app
/(dashboard) # Authenticated routes
/chat # Chat list & conversation view
/help # Help & FAQ
/profile # User profile
/settings # App settings
/subscription # Plan management
/logged-out # Post-sign-out page
/components
/chat # ChatBubble, Sidebar, PromptInput, etc.
/layout # PageHeader, PageContent, PageFooterLinks
/ui # Button, Modal, Card, Spinner, etc.
/contexts # Auth, User, Chat, Sidebar, Subscription
/data # Constants, mock data, FAQ, prompts
/hooks # useChat, useTheme, useIsMobile, useStreamingText
/lib # streamText, settings, subscription, account, etc.
All conversations use realistic mock datasets.
Examples include:
- Product design prompts
- Code generation queries
- Analytics insights
- Technical explanations
No lorem ipsum or placeholder dialogue. Data is persisted in localStorage for demo continuity.
Clone the repository:
git clone https://github.com/dchobarkar/copilotui.git
cd copilotuiInstall dependencies (npm or pnpm):
npm install
# or
pnpm installRun the development server:
npm run dev
# or
pnpm devOpen http://localhost:3000.
npm run build
npm startOptimized for Vercel deployment.
- Push repo to GitHub
- Import into Vercel
- Deploy instantly
- Optimized markdown parsing
- Lazy syntax highlighting
- Smooth streaming rendering
- Memoized chat bubbles
CopilotUI contains no proprietary AI product interfaces or client data.
All conversational systems and UI flows are self-initiated demonstrations.
This project is part of a capability showcase, but forks and adaptations are welcome.
MIT License — free for personal and commercial adaptation.
If you found this project helpful, consider starring the repository.
Built to simulate real AI product experiences ⚡