Opening the Web UI
Access the Web UI in multiple ways:- Keyboard Shortcut
- Direct URL
Press
Ctrl+Shift+L (Windows/Linux) or Cmd+Shift+L (Mac) to open the Web UI in a new tab.Interface Overview
The Web UI consists of several key areas:Main Chat Area
The central workspace where conversations take place:- Message Display: Full conversation history with formatted responses
- Code Blocks: Syntax highlighting and copy buttons
- Markdown Support: Rich text rendering for better readability
- Image Previews: Inline image display for vision-enabled models
- Message Actions: Copy, edit, regenerate, or delete individual messages
Navigation Sidebar
Located on the left side (toggle withCtrl+B):
- Chat History: Browse and search through past conversations
- New Chat: Create new conversation threads
- Folders/Organization: Group related chats together
- Search: Find specific conversations quickly
Input Area
At the bottom of the chat interface:- Text Input: Multi-line text area with auto-resize
- Attachment Options: Upload images, documents, or files
- Mode Toggles: Quick access to chat modes (webpage, search, knowledge base, vision)
- Model Selector: Switch AI models on the fly
- Submit Options: Send with Enter or Ctrl+Enter (configurable)
Settings Panel
Comprehensive configuration options:General Settings
- Wide screen mode for larger displays
- Resume last chat on opening
- Restore last used model for previous chats
- Generate titles using AI
Model Settings
Configure AI models from various providers:
- Ollama (local models)
- OpenAI (GPT-4, GPT-3.5)
- Anthropic (Claude)
- Google (Gemini)
- Custom API endpoints
RAG Settings
Configure Retrieval Augmented Generation:
- Embedding model selection
- Chunk size and overlap
- Number of retrieved documents
- Splitting strategy
Key Features
Chat Management
Multiple Conversations
Multiple Conversations
Manage unlimited conversations simultaneously:
- Each chat maintains its own context and history
- Switch between chats without losing progress
- Organize chats with custom titles
- Export or delete conversations as needed
Chat History
Chat History
Full conversation history with:
- Automatic saving to local storage
- Search and filter capabilities
- Export to markdown or JSON
- Import previous conversations
Message Editing
Message Editing
Edit and refine your conversations:
- Edit previous messages and regenerate responses
- Delete individual messages
- Copy message content
- View message metadata (tokens, model, timestamp)
Advanced Features
Wide Screen Mode
For larger displays, enable wide mode for better readability:- Go to Settings → General Settings
- Enable “Enable wide screen mode”
- The interface expands to utilize more screen space
Wide mode is recommended for screens 1920px or wider.
Resume Last Chat
Automatically continue where you left off:- Go to Settings → General Settings
- Enable “Resume the last chat when opening the Web UI”
- Your most recent conversation loads automatically on startup
Model Restoration
Restore the specific model used in each chat:- Go to Settings → General Settings
- Enable “Restore last used model for previous chats”
- When switching between old chats, the model automatically changes to match
AI Title Generation
Automatically generate descriptive titles for conversations:- Go to Settings → General Settings
- Enable “Generate Title using AI”
- New chats automatically receive contextual titles based on content
Keyboard Shortcuts
Boost your productivity with keyboard shortcuts:| Action | Shortcut | Description |
|---|---|---|
| Open Web UI | Ctrl+Shift+L | Open in new tab |
| New Chat | Ctrl+Shift+O | Start new conversation |
| Toggle Sidebar | Ctrl+B | Show/hide chat history |
| Focus Input | Shift+Esc | Jump to message input |
| Toggle Chat Mode | Ctrl+E | Switch chat modes |
Keyboard shortcuts can be customized in your browser’s extension management page.
Settings and Configuration
Managing Models
The Web UI provides comprehensive model management:- Ollama
- OpenAI
- Other Providers
Configure local Ollama models:
- Go to Settings → Ollama Settings
- Enter your Ollama server URL (default: http://localhost:11434)
- Click “Fetch Models” to load available models
- Configure default models for chat, embedding, and vision
Knowledge Base Management
Upload and manage your documents:- Go to Settings → Manage Knowledge
- Click “Add New Knowledge”
- Upload files (PDF, DOCX, TXT, CSV, MD)
- Wait for processing and embedding generation
- Select knowledge bases to use in conversations
Prompt Library
Create reusable prompts:- Go to Settings → Manage Prompts
- Choose prompt type:
- System Prompts: Set conversation context
- Quick Prompts: Reusable templates
- Copilot Prompts: Context menu actions
- Create and save your prompts
Data Management
Export and Backup
Export Conversations
- Right-click on any chat in history
- Select “Export” to save as JSON or Markdown
- Choose location and save
Export All Data
- Go to Settings → About
- Use backup/export options
- Save all conversations, settings, and knowledge bases
Privacy and Storage
All data is stored locally in your browser. Page Assist does not send data to external servers except to your configured AI providers.
- Chat History: IndexedDB (browser database)
- Knowledge Base: Browser local storage + embeddings
- Settings: Chrome storage API
- Vector Data: Indexed in browser memory
Tips and Best Practices
Troubleshooting
Web UI won't open
Web UI won't open
- Check if the extension is enabled
- Try the keyboard shortcut
Ctrl+Shift+L - Click the extension icon and select “Open Web UI”
- Check browser console for errors
Chat history not loading
Chat history not loading
- Check browser storage settings
- Ensure you haven’t cleared browser data
- Try refreshing the page
- Check IndexedDB in browser DevTools
Settings not saving
Settings not saving
- Grant storage permissions to the extension
- Check if you’re in private/incognito mode
- Try restarting the browser
- Verify storage quota isn’t exceeded
Next Steps
Internet Search
Add real-time web search to your conversations
Knowledge Base
Upload and chat with your documents
Prompts
Create custom prompts and templates
Vision
Analyze images with vision-capable models