> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/n4ze3m/page-assist/llms.txt
> Use this file to discover all available pages before exploring further.

# Settings Overview

> Configure Page Assist with model settings, RAG, TTS, web search, and more

Page Assist provides comprehensive settings to customize your AI experience. Access settings through the extension options page.

## General Settings

### Web UI Configuration

* **Language**: Choose your preferred interface language from multiple supported languages
* **Theme**: Switch between light and dark mode
* **Speech Recognition Language**: Select the language for voice input
* **Default Prompts**: Set default prompts for SidePanel (Copilot) and Web UI
* **Resume Last Chat**: Continue your previous conversation when opening SidePanel or Web UI
* **Wide Screen Mode**: Enable wider layout for better readability
* **Sidebar Position**: Choose left or right sidebar placement

### Chat Behavior

* **Chat with Website**: Enable chat with current webpage by default in SidePanel
* **Generate Title**: Automatically generate chat titles using AI
* **Restore Last Model**: Automatically select the last used model for previous chats
* **Hide Model Settings**: Hide the current chat model settings panel
* **Temporary Chat**: Enable temporary (non-saved) chats in SidePanel or Web UI by default
* **Persist Chat Input**: Save unsent messages automatically

### Message Display

* **User Chat Bubble**: Display user messages in chat bubble style
* **Enable Markdown for User Messages**: Format user messages with Markdown
* **Show More for Large Messages**: Collapse large messages with a "show more" button
* **Text Wrapping in Tables**: Enable text wrapping in Markdown tables
* **Optimize for Small Screens**: Adjust chat UI for smaller displays

### Advanced Features

* **Tab Mentions**: Enable @tab mentions to reference browser tabs (Beta)
* **Paste Large Text as File**: Automatically convert large pasted text to file attachments
* **OCR Language**: Set default language for optical character recognition
* **Thinking Mode**: Show thinking mode state in forms by default
* **Reasoning Widget**: Control visibility of reasoning collapse panel
* **Enable Memory**: Experimental memory feature for context retention (Beta)
* **Message Queue**: Enable message queuing while streaming responses

### Copy & Clipboard

* **Auto Copy to Clipboard**: Automatically copy AI responses to clipboard
* **Copy as Formatted Text**: Copy text with formatting preserved
* **Remove Reasoning Tag**: Strip reasoning tags from copied text

## Model Settings

### Ollama Configuration

* **Ollama URL**: Set your Ollama server URL (default: `http://localhost:11434`)
* **Global Enable/Disable**: Toggle Ollama integration globally
* **Connection Status Check**: Enable/disable Ollama connection status monitoring
* **Custom Origin URL**: Advanced URL rewriting for custom setups
* **Auto CORS Fix**: Automatic CORS configuration for Ollama connections
* **Custom Headers**: Add authentication or custom headers for Ollama requests

### OpenAI Compatible APIs

Configure OpenAI-compatible endpoints like LM Studio, llamafile, or other providers:

* **API Endpoint**: Set custom API base URL
* **API Key**: Provide authentication key
* **Model Selection**: Choose from available models

### Chrome AI (Gemini Nano)

Use built-in Chrome AI capabilities when available.

## RAG (Retrieval) Settings

### Pipeline Configuration

* **Enable Embedding and Retrieval**: Toggle RAG functionality
* **Embedding Model**: Select embedding model (recommended: `nomic-embed-text`)
* **Chunk Size**: Set text chunk size for document splitting
* **Chunk Overlap**: Configure overlap between chunks
* **Text Splitter**: Choose splitting strategy
* **Separator**: Define text splitting separator (e.g., `\n\n`)
* **Retrieved Documents**: Set number of documents to retrieve per query
* **File Upload Limit**: Maximum files per knowledge base
* **Maximum Content Size**: Max content size for full context mode (default: 4028)

### Chat with Webpage

* **Full Context Mode**: For smaller pages, use complete page content
* **Embedding Mode**: For larger pages, use RAG with embeddings

## Text-to-Speech (TTS)

* **Enable TTS**: Toggle text-to-speech functionality
* **Auto Play**: Automatically play voice response after completion
* **TTS Provider**: Select text-to-speech provider
* **Voice Selection**: Choose from available voices
* **SSML Support**: Enable Speech Synthesis Markup Language
* **Response Splitting**: Split long responses for better TTS performance
* **Remove Reasoning Tag**: Strip reasoning content from TTS output

## Speech-to-Text (STT)

* **Auto Stop Timeout**: Set timeout in milliseconds for automatic speech detection stop
* **Auto Submit**: Automatically submit voice messages after recording

## Web Search

### Search Configuration

* **Enable Web Search**: Perform internet searches from chat
* **Search Engine**: Choose from multiple providers:
  * Google
  * Brave Search
  * Tavily
  * Exa
  * SearXNG (self-hosted)
* **Search Results**: Set total number of search results to fetch
* **Search ON by Default**: Enable internet search by default
* **Visit Specific Website**: Automatically visit websites mentioned in messages
* **Google Domain**: Configure Google search domain

### Search Filters

* **Domain Filter**: Only show results from specific domains
* **Blocked Domains**: Exclude results from certain domains

### API Keys

* **Brave API Key**: For Brave Search integration
* **Tavily API Key**: For Tavily search integration
* **Exa API Key**: For Exa search integration
* **Firecrawl API Key**: For web scraping with Firecrawl
* **SearXNG URL**: Self-hosted search engine URL

## Sharing Settings

* **Page Share URL**: Configure custom share server URL for privacy
* **Enable/Disable Sharing**: Toggle chat sharing feature
* **Manage Shares**: View and delete shared conversations

<Note>
  For privacy reasons, you can self-host the page share server and configure your own URL.
</Note>

## Knowledge Base

Manage your local knowledge base:

* **Upload Documents**: Add PDFs, text files, and other documents
* **File Limit**: Control maximum files per knowledge base
* **Indexing Notifications**: Get notified when indexing completes
* **Process Settings**: Configure how documents are processed and embedded

## System Settings

### Data Management

* **Export All Data**: Backup chat history, knowledge base, prompts, and settings
* **Import All Data**: Restore from backup file
* **System Reset**: Clear all data and return to default settings

<Warning>
  System reset is permanent and cannot be undone. Export your data before resetting.
</Warning>

### Extension Behavior

* **Extension Icon Action**: Set default action for extension icon clicks
* **Context Menu Action**: Configure right-click menu behavior
* **Show Web UI Button**: Display Web UI button in SidePanel
* **Font Size**: Adjust interface font size
* **Chat Background**: Set custom background image
* **Storage Sync**: Sync settings across devices using browser storage

## YouTube Integration

* **Auto Summarize**: Show summarize button on YouTube videos automatically (Beta, Chromium only)

## Prompts Management

Create and manage custom prompts:

* **System Prompts**: Define AI behavior and personality
* **Quick Prompts**: Reusable prompt templates with variables
* **RAG Prompts**: Specialized prompts for knowledge base queries
* **Copilot Prompts**: Prompts for SidePanel usage

## Model Management

For Ollama users:

* **Pull Models**: Download new models from Ollama library
* **Delete Models**: Remove unused models
* **Re-pull Models**: Update existing models
* **Model Nicknames**: Assign custom names to models
* **View Details**: See model information (family, size, quantization, etc.)

## Accessing Settings

### From Extension Icon

1. Click the Page Assist extension icon
2. In the Web UI, click the settings icon (gear)
3. Navigate through different settings tabs

### From Browser Extension Management

1. Go to your browser's extension management page
2. Find Page Assist
3. Click "Options" or "Extension Options"

## Privacy

Page Assist stores all data locally in your browser. The only external communication occurs when:

* Connecting to your AI provider (Ollama, OpenAI-compatible endpoints, etc.)
* Using the share feature (can be disabled)
* Performing web searches (when enabled)

All settings and chat history remain on your device unless you explicitly enable browser storage sync.