Introduction
The chat input is your starting point for any work in Sana, allowing you to get started with your work. Type, paste, drop images and files, access and edit tasks, choose sources, and select a model—all in one place.
The new chat input makes every workflow faster, easier, and more flexible and replaces previous features such as Pro mode and chat filters.
Using the Chat Input
Here is an overview of the different functionalities you can access from the chat input window on the Start page.
1. Input Field
The main point of interaction for the user.
Write or paste texts or images to the agent
Drag and drop files to interact with them directly.
2. Tasks
Within the chat input, you can quickly access tasks to automate routine workflows and enhance your productivity without leaving the conversation.
Quick Access: Tasks are pre-built workflows with instructions you can run as often as you need. Instantly see and select your most recently used tasks for convenience. See here for more details.
Reusable Workflows: Selecting a task will automatically fill the input field for you, so you can launch it with one click.
Search and Discover: Use the search bar to quickly find the exact task you need, saving time and reducing manual effort.
Browse the Full Library: Click “All tasks” to explore the entire collection of available workflows and discover new ways to work efficiently.
New features:
Editable Before Sending: You can review and edit the task instructions in the input field before sending, giving you flexibility to customize each run.
Use Tasks Inside Chats: Tasks inside chat aren’t just for starting new work, you can also access and run them directly within any ongoing conversation in chat.
3. Create
Use the chat input to tell Sana exactly what you want to create. If you already know what you want to create, select between a doc, presentation, image, or sheet. If not, or leave the field blank if you want to explore freely.
Doc: Generate structured documents for reports, proposals, or notes.
Presentation: Craft engaging slides for meetings, pitches, or training.
Image: Create custom visuals, diagrams, or illustrations on demand.
Sheet: Build data tables, trackers, or calculations in spreadsheet format.
4. Sources
Control where Sana looks for information and tailor the agent’s responses.
Source Filters: Select specific folders, apps, files, or meetings to focus the agent’s search.
Broad or Specific Search: By default, Sana searches broadly, but you can narrow the focus to avoid irrelevant results.
Browse and Upload: Use the search icon to find and add assets, or click “Upload files” to add new files from your computer.
5. Model selection
Choose the LLM behind your chat.
On the far right, you can select which model you want to use in the chat.
Your workspace administrator controls which models are available to you.
Frequently Asked Questions
Q: Do I need to use sources to filter the responses?
A: No, but setting the source filters yourself, you can help the agent become more accurate to your needs.
Q: Do I need to select a mode to create an output?
A: No, in any chat you can also ask Sana to create e.g. a doc or presentation. Sana will try to understand your intent automatically. By selecting the mode yourself, you can faster go into the output mode you want.
If you have questions or need help, contact [email protected].