Skip to main content

​
Layout overview

When you open a chat, the screen splits into two panels:
  • Left — Chat panel: Where you ask questions and receive AI-generated answers with citations.
  • Right — Document viewer / Insights: Where you read the original document or browse AI-generated insights.
When you first open a document, Sharly automatically generates a summary and displays suggested questions — clickable prompts you can use to start exploring immediately.

​
AI modes

At the top of the chat panel, a model selector lets you choose how the AI processes your questions:
Fastest responses. Best for straightforward lookups and simple questions.
A middle ground between speed and depth. Good for most everyday questions.
The most thorough mode, using advanced reasoning for complex analysis. Best for nuanced research questions that require deeper interpretation.

​
Chat settings

Click the gear icon at the bottom of the chat to open settings that control how the AI responds:
  • Language — Set the response language for this specific chat. Each chat can use a different language.
  • Custom Instructions — Provide specific instructions the AI should follow for all responses in this chat (e.g., “Always respond in bullet points” or “Focus on financial data”).
  • Answer Length — Choose between Short, Medium (default), or Long responses.
  • Content Source:
    • Focused — Responses are strictly based on your uploaded document content. Best when you need verified, source-only answers.
    • Expanded — The AI supplements document content with general knowledge. Useful when you need broader context beyond what’s in the file.
  • Prompt Bar — Toggle the quick-access prompt bar on or off.

​
Working with AI responses

Every AI response includes three action buttons:
  • Text-to-Speech (speaker icon) — Have the response read aloud.
  • Copy (clipboard icon) — Copy the response text to your clipboard.
  • Citations — View the exact source excerpts the AI used, including page numbers and direct links to navigate to that location in the document viewer.

​
Document viewer

The right panel’s Viewer tab displays your document inline. A toolbar along the right edge provides:
  • Capture a Section — Screenshot a portion of the document.
  • Document List — Switch between documents if multiple files are loaded.
  • Zoom In / Zoom Out / Actual Size — Adjust the view.
  • Search — Find text within the document.
If you have multiple documents in the chat, use the dropdown at the top of the viewer to switch between them.

​
Insights panel

Switch to the Insights tab (next to Viewer) to see AI-generated analysis organized into expandable sections:
  • Summary — A concise overview of the document.
  • Citations — Key referenced sources within the document.
  • Author(s) — Detected author information.
  • Highlights — The most important points and takeaways.
  • Findings — Key conclusions and results.
  • Topic & Keywords — Main subjects and recurring terms.
Each section can be expanded or collapsed to focus on what matters most.

​
Managing documents in a chat

Click the folder icon at the bottom of the chat to open the Collection panel. From here you can:
  • See all documents currently loaded in the chat session.
  • Search through your documents.
  • Add or remove documents from the conversation.
This is useful when you want the AI to answer questions across multiple documents at once.

​
Sharing a conversation

Click the Share Conversation button in the top-right corner of the chat to share the entire conversation with others.