Sairis

Appearance

AI Assistant User Guide ​

Introduction ​

The AI Assistant is an advanced conversational interface that provides intelligent responses to your questions using large language models, knowledge bases, and content integration. This guide explains how to use the AI Assistant effectively, including its core features, navigation, and advanced capabilities.

Getting Started ​

Accessing the Assistant ​

The AI Assistant is available from the main navigation menu. When you first access the Assistant, you'll see either:

  • The LaunchPad screen with quick access to common resources and Blueprints
  • A welcome message inviting you to start a conversation

First Interaction ​

To begin using the Assistant:

homepage

  1. Type a question in the prompt field at the bottom of the screen
  2. Press Enter or click the Send button (paper airplane icon)
  3. The AI will process your question and provide a response

Interface Overview ​

The AI Assistant interface consists of these major components:

  1. Header Bar: Contains the home button, Blueprint selector, and utility controls
  2. Conversation Area: The main section where your conversation with the AI appears
  3. Prompt Field: The input area at the bottom where you type your questions
  4. Side Panel: Optional panel for chat history, content selection, or AI settings
  5. Action Menu: Available through the "+" button beside the prompt field

Conversation Basics ​

Asking Questions ​

To ask the AI a question:

  1. Type your question in the prompt field at the bottom of the screen
  2. Press Enter or click the Send button (paper airplane icon)
  3. Your question will appear in the conversation area
  4. The AI will generate a response which appears below your question

Tips for effective questions:

  • Be specific about what you need
  • Provide relevant context
  • Ask one question at a time for best results
  • Use follow-up questions to refine or expand on earlier responses

Viewing Responses ​

AI responses appear in the conversation area with:

  • The AI's avatar icon (determined by the active Blueprint)
  • The response text with formatting (headings, lists, etc.)
  • Citations and references where applicable
  • Action buttons for response management

During response generation:

  • A "Generating your response" indicator appears
  • Text is streamed in real-time as it's generated
  • You can stop generation with the stop button (square icon)
  • Citations may be gathered after the main response is complete

Using the Microphone ​

Instead of typing, you can use voice input:

  1. Click the microphone icon next to the prompt field
  2. Begin speaking clearly
  3. The system will transcribe your speech in real-time
  4. When you stop speaking, the question will be submitted automatically
  5. To cancel voice input, click the microphone icon again

Attaching Files ​

You can attach files to provide context for your questions:

  1. Click the paperclip icon in the prompt field
  2. Select one or more files from your device (up to 10MB total)
  3. Supported file types include PDFs, Word documents, Excel files, text files, and images
  4. The files will appear as chips below the prompt field
  5. Type your question about the files and send
  6. The AI will process the files and use their content to inform its response

Blueprint Feature ​

Selecting Blueprints ​

Blueprints are specialized AI configurations for different purposes:

  1. Click the Blueprint selector in the header (shows current Blueprint name)
  2. Choose from available Blueprints in the dropdown menu
  3. Each Blueprint has specific knowledge, capabilities, and conversation starters
  4. The AI avatar will change to match the selected Blueprint
  5. You can also click a Blueprint avatar in previous responses to switch to that Blueprint

Blueprint Home Screen ​

When you select a Blueprint, you may see a home screen with:

  1. Blueprint name and description
  2. Blueprint avatar/initials
  3. Available conversation starters (if configured)
  4. Option to start typing or use the microphone

This screen helps you understand the Blueprint's purpose and available capabilities.

Conversation Starters ​

Many Blueprints include conversation starters - predefined prompts for common tasks:

  1. These appear on the Blueprint home screen
  2. Click a conversation starter to immediately send that prompt
  3. The AI will process the conversation starter as if you had typed it
  4. You can also access conversation starters via the shortcuts menu (top right of conversation area)

Working with Responses ​

Citations ​

When the AI uses specific information sources, citations appear:

  1. Citations are numbered in the response text (e.g., [1], [2])
  2. Click "Citations" at the bottom of the response to view sources
  3. Each citation includes:
    • Title of the source
    • Excerpt of the relevant content
    • Document type icon
    • Page number (if applicable)
    • Link to view the full source

You can:

  • Expand individual citations to see excerpts
  • Click "Expand All" to view all citations at once
  • Click "View" to open referenced documents or web pages

Viewing Referenced Content ​

When viewing referenced content:

  1. For documents (PDF, Word, etc.):

    • A full-screen viewer opens
    • The relevant excerpt is highlighted
    • You can navigate through the document
    • For PDFs, the specific page is shown

  2. For web pages:

    • A new browser tab opens with the page
    • A "Loading" dialog briefly appears

  3. For videos:

    • A video player opens with playback controls
    • Transcription may be available in an accordion below

Text Selection Actions ​

When you select text in an AI response:

  1. A floating menu appears with options:

    • Reply: Uses the selected text as context for your next question
    • Copy As: Copy text in regular or formatted form
    • Send To: Send to Microsoft Word or a new email

  2. To use these options:

    • Select the text you want to interact with
    • Choose the desired action from the floating menu
    • The action will be performed immediately

Response Actions ​

Each AI response has action buttons:

  1. Replay Question: Copies the original question to the prompt field
  2. Copy Answer: Copies the entire response to your clipboard
  3. Modify Answer: Shows a menu of ways to change the response:
    • Convert to step-by-step directions
    • Format data using tables
    • Elaborate on the answer
    • Summarize content
    • Translate to various languages
    • Change tone (Professional, Casual, Friendly, Marketing)
    • Change style (Business, Legal, Journalism, Creative)
  4. Audio: Toggle speech synthesis to read the response aloud

Advanced Features ​

Chat History ​

Access your conversation history through the side panel:

  1. Click the chat history icon in the header (document icon)
  2. View a list of your past conversations
  3. Click on any conversation to resume it
  4. Use the New Chat button to start a fresh conversation
  5. The three-dot menu for each conversation allows:
    • Renaming the conversation
    • Adding tags
    • Deleting the conversation

Content Panel ​

The Content panel allows you to chat with specific documents:

  1. Select "Content" in the side panel menu
  2. Browse available content or search for specific items
  3. Select documents you want the AI to reference
  4. Ask questions about the selected content
  5. The AI will focus on the chosen documents when answering

AI Settings ​

Customize the AI's behavior through Settings:

  1. Click the settings icon in the header or select "Settings" in the side panel
  2. Adjust parameters such as:
    • Model selection (Haiku or Sonnet)
    • Creativity level
    • Web search settings
    • Citation preferences
    • Knowledge base selection
    • Web pages to reference
  3. Changes apply to the current session only unless saved as a Blueprint

Keyboard Shortcuts ​

Keyboard shortcuts for power users:

  • Ctrl + N: Start new chat
  • Ctrl + K: New chat keeping current settings
  • Ctrl + E: Clear current conversation
  • Ctrl + R: Rename current chat
  • Ctrl + D: Delete current chat
  • Ctrl + S: Show chat settings
  • Ctrl + A: Attach file
  • Ctrl + C: Switch to content panel

Mobile Experience ​

When using the Assistant on mobile devices:

  1. The interface adapts to the smaller screen
  2. Side panels become full-screen overlays
  3. Access the menu by tapping the three-line icon
  4. Scroll through responses by swiping
  5. Long-press text to bring up the selection menu
  6. Tap the microphone for voice input

Troubleshooting ​

Issue: AI response stops generating

  • Click the refresh button to retry
  • Check your network connection
  • Try starting a new conversation

Issue: Attached files not being processed

  • Ensure file format is supported
  • Check that file size is under 10MB
  • Try with a smaller or different file format

Issue: Citations not appearing

  • Some responses may not require citations
  • Check if "Cite Sources" is enabled in settings
  • Try asking more specific questions about documents

Issue: Voice input not working

  • Check microphone permissions in your browser
  • Speak clearly and at a moderate pace
  • Try using shorter phrases

Issue: Response not relevant to my question

  • Try rephrasing your question for clarity
  • Provide more specific context
  • Ask follow-up questions to guide the AI
  • Check which Blueprint is active
  • In settings, verify knowledge bases are correctly selected

Issue: Unable to find past conversations

  • Check you're signed in with the correct account
  • Use the search function in chat history
  • Filter by tags if you've applied them
  • Some older conversations may have been automatically archived

For additional assistance, contact your system administrator or support team.