Use Chat

​

Introduction

A Conversation in ELITEA represents a dynamic dialogue involving multiple participants. These can include language models (LLMs), Agents, Pipelines, Toolkits, MCPs, and human Users like yourself. You interact using natural language, and the chat maintains context, allowing you to refer to previous messages within the same conversation. Conversations are isolated; context is not shared between different conversations. All your conversations are securely stored on the ELITEA server, making them accessible from any device where you log in. You can find all your conversations listed under the Chat menu in the sidebar. Key Features

​

Getting Started

​

Creating a New Conversation

1. In the main sidebar on the left, locate the Chat section.
2. Click the + Conversation button at the top of the sidebar.
3. The chat input field appears and is highlighted to focus your attention.
4. You’ll see a welcome screen with the message “Hello, [Your Name]! What can I do for you today?”
5. The message input box at the bottom shows the placeholder: “Type your message. Use # to search and add AI assistants to conversation.”
6. Choose your approach:
   • Add a participant first: Type # to search and select an Agent, Pipeline, Toolkit, or MCP from the dropdown list that appears. Selected participants will appear as chips above the input box.
   • Select a model: Click the model selector dropdown to choose an LLM (e.g., GPT-4, Claude).
7. Type your message: Enter your initial message, question, or command (e.g., “Help me write a Python script”, “What are the best practices for API design?”, “Explain quantum computing”).
8. Send: Click the Send icon (paper airplane icon) or press Enter.
9. The new conversation is created and appears in the CONVERSATIONS sidebar on the left.
10. The conversation name is automatically generated based on your message content. During name generation (1-2 seconds), you’ll see a loader icon with “Naming” text next to the conversation item. You can manually rename it at any time after generation completes.

​

Creating a New Folder

Organize your conversations by grouping them into folders.

1. In the Conversations sidebar, locate the folder icon button next to “Conversations” at the top.
2. Click the folder icon (Create folder button).
3. A new folder entry will appear. Enter a descriptive Name for your folder.
4. Press Enter or click the checkmark to save.

The new folder will appear in your CONVERSATIONS sidebar.

​

Managing Conversations

​

Moving Conversations to Folders

To organize conversations into folders: Method 1: Drag and Drop

1. In the CONVERSATIONS sidebar, click and hold on the conversation you wish to move.
2. Drag the conversation over the destination folder.
3. Release to drop the conversation into the folder.

Method 2: Context Menu

1. In the CONVERSATIONS sidebar, right-click on the conversation you wish to move.
2. Select Move to from the context menu.
3. Choose the desired destination folder from the list, OR
4. Select Create folder to create a new folder and move the conversation into it simultaneously.
5. To move a conversation back to the main list, select Back to the list from the Move to menu.

​

Conversation Actions (Sidebar)

You can manage conversations directly from the CONVERSATIONS sidebar by right-clicking on a conversation or clicking the options menu (often …) associated with it. The following actions are available:

• Rename: Rename the conversation. Enter the new name and confirm.
• Move To: Move the conversation into a folder, as described above.
• Playback: Enter Playback mode for this conversation (See Playback Mode).
• Pin/Unpin: Select Pin on top to keep the conversation at the top of the list for easy access. Select Unpin on a pinned conversation to remove it from the top.
• Delete: Permanently remove the conversation. You will be asked to confirm this action.

​

Managing Folders

​

Folder Actions (Sidebar)

Folders can be managed directly from the CONVERSATIONS sidebar. Right-click on a folder or click its associated options menu (…) to access the following actions:

• Rename: Rename the folder. Enter the new name and click the checkmark (✔) or Save button.
• Pin/Unpin: Select Pin on top in the folder contextual menu to keep the folder at the top of the CONVERSATIONS sidebar for quicker access. Select Unpin to remove it from the top.
• Delete: Remove the folder.
  Important: Deleting a folder does not delete the conversations inside it. Conversations within a deleted folder are automatically moved back to the main conversation list (root level). You will be asked to confirm deletion.

​

Understanding Conversation/Folder Visibility

ELITEA has two project types — Private and Team. Visibility and access rules for conversations and folders depend on which type you’re working in. Example: Collaboration in a Team Project

View collaboration scenario

Scenario: The QA, PM, and BA team collaborate on a new file upload feature using a shared team project conversation.Steps:

• BA adds requirements:
  @BA: "The new feature should allow users to upload files up to 10MB. The system must validate file types and provide error messages for unsupported formats."
  
• PM sets the timeline:
  @PM: "Development completes by May 15th. QA starts testing May 16th. Release target: May 20th."
  
• QA plans testing:
  @QA: "I'll create test cases for uploads, file types, size limits, and error handling. @BA, does the system support drag-and-drop uploads?"
  
• BA clarifies:
  @BA: "Yes, drag-and-drop should be supported. Also add a progress bar during uploads."
  
• PM tracks progress:
  @PM: "QA, please share test results by May 18th for pre-release review."
  

The PM creates a “Feature: File Upload” folder and moves the conversation there so all team members can find it easily.

Tips for Effective Collaboration in Team Projects

• Use @ mentions to notify specific teammates in a conversation.
• Group related conversations into folders for better navigation.
• Define roles, responsibilities, and deadlines directly in the chat.
• Regularly post progress updates and flag blockers in the conversation.

​

Working with Participants

Participants are the core components you interact with within a conversation.

​

What are Participants?

Participants are the “tools” or “entities” you add to your chat:

​

Adding Participants to a Chat

​

Creating New Participants from Chat

You can create new participants directly from the chat interface:

1. Click the + icon in the input toolbar.
2. Hover over the relevant category (Agents, Pipelines, Toolkits, or MCPs).
3. Click the Create New option at the bottom of the submenu.
4. The corresponding Canvas editor opens inline.

• Agents: See How to Create and Edit Agents from Canvas.
• Pipelines: See How to Create and Edit Pipelines from Canvas.
• Toolkits: See How to Create and Edit Toolkits from Canvas.
• MCPs: See How to Create and Edit MCPs from Canvas.

​

Adding Users to a Conversation

Users (team members) can only be added to in team projects. They cannot be added to private conversations. To Add Users:

1. In the chat input toolbar, click the + icon.
2. Select Invite Users from the popup menu.
3. The Add users modal will appear with a search bar.
4. Use the search bar to find teammates by name.
5. Select one or more users from the list by clicking on them.
6. Click Add to confirm. The selected users will appear in the Users section of the PARTICIPANTS panel.

Important Notes:

• Added users will receive notifications about being added to the conversation.
• Users can also join a public conversation implicitly by interacting with it.
• Once a conversation is public, it cannot be converted back to private.

For detailed information on adding teammates, see Adding Teammates to Conversation.

​

Using Participants in a Conversation

Once added, participants are ready to process your messages:

1. Check Participants: Ensure the desired participant is listed in the Participants section.
2. Select Active Participant(s):
   • Click: Click the participant’s name/icon in the Participants list to make it active for your next message.
3. Mention Users (Team Projects only): You can mention other team members using @ followed by their name (e.g., @John Doe). This notifies them and brings their attention to specific parts of the conversation.
4. Send Message/Command: Type your message or a simple command (like “Go”, “Execute”, “Run it”) and press Send. The active participant(s) will process your input.

Example Usage

• To ask a general question using a specific model:
  1. Select a model from the model selector dropdown (e.g., GPT-4o).
  2. Type: "Explain the concept of recursion in programming." -> Send.
• To use a specific agent:
  1. Click the + icon in the input toolbar and hover over Agents.
  2. Select Data Analysis Agent from the submenu, or type #Data Analysis Agent in the input box.
  3. Type your request: "Analyze the latest sales data." -> Send.
• To use a toolkit:
  1. Click the + icon and hover over Toolkits. Select the desired toolkit (e.g., Artifact Toolkit).
  2. Type: "Store this data in artifacts." -> Send.
• To mention a team member (Team Projects only):
  1. In your message, type @ followed by the team member’s name (e.g., @John Doe).
  2. Type: "@John Doe, can you review this analysis?" -> Send.

​

Selecting and Configuring Models

Selecting a Model:

1. Click the model selector dropdown at the bottom of the chat.
2. Select a desired LLM model from the available options (e.g., GPT-4o, GPT-5.1, Claude).
3. The selected model name is displayed on the button. Models that support image analysis or reasoning show small capability icons next to their name in the dropdown.

Model Settings: Click the Settings (⚙️) icon next to the model selector to fine-tune response generation. Settings vary by model type: Capabilities (shown when supported by the selected model) Max Completion Tokens (All Models) Steps Limit (Conversations only) Controls the maximum number of execution steps (tool calls) the AI can take before the loop is forcefully stopped. This prevents runaway agent executions that could consume excessive resources.

• Each time the AI invokes a tool or performs an internal reasoning step, the counter increments by one. When the limit is reached, the execution loop ends and the AI returns whatever results it has collected so far.
• The Steps Limit field is shown in the model settings panel only in Conversations (chat). It is not available on the Agents or Pipelines pages, where step limits are configured at the agent/pipeline level.

​

Configuring Participants

You can configure and edit participants (Agents, Pipelines, Toolkits, and MCPs) directly within the conversation using the Canvas editor. Accessing Participant Settings:

1. Option 1: Click the participant in the Participants list, then click the ⚙️ (settings) icon.
2. Option 2: Hover over the participant element in the Participants list and click the Edit icon that appears.
3. The appropriate Canvas editor will open based on the participant type.

What You Can Configure: Applying Changes:

1. Make your edits in the Canvas editor.
2. Click the Save button to apply your modifications.
3. The updated configuration will be used for subsequent messages in the conversation.

​

Displaying Configured Conversation Starters

When you add a participant (like an Agent, Pipeline, Toolkit, or MCP) that has a pre-configured “conversation starter” message or instruction set, this message will automatically appear in the chat. This helps guide you on how to interact with the participant effectively.

​

Interacting with Conversation Outputs

​

Editing Your Last Message

If you want to refine your prompt instead of sending a brand-new follow-up, ELITEA lets you edit your last user message and regenerate the answer from that updated text.

1. Locate your most recent user message.
2. Use the action icons shown under that message, then click the Edit (pencil) icon.
3. The message switches into inline edit mode.
4. If the original message included attachments, they remain visible in edit mode and are not lost automatically.
5. Update the text as needed.
6. If needed, remove an existing attached file from the edit view, or use the attach icon on the left side to add a new file.
7. Click Save and apply to submit the edited message and regenerate the answer, or click Cancel to discard your changes.

​

Like/Dislike and Commenting

1. Below each generated response, you’ll see Thumbs Up (👍) and Thumbs Down (👎) icons.
2. Click Thumbs Up to indicate satisfaction.
3. Click Thumbs Down to indicate dissatisfaction.
4. After clicking Thumbs Down, a Leave comment field appears. Click it, type your specific feedback or reason for disliking, and press Send (or Enter). This feedback is valuable for improving models and prompts.

​

Regenerating the Last Output

1. If you’re not satisfied with the very last response generated by a participant, you can ask it to try again.
2. Ensure a response has been generated.
3. Click the Regenerate icon 🔄 usually located near the last message or the input box.
4. The system will use the same input/prompt that generated the last response and attempt to create a new, potentially improved, output.

​

Using Canvas for Content Editing

Canvas is your all-in-one workspace for editing, refining, and collaborating on AI-generated content in ELITEA. Instead of copying results into other tools, you can work directly with code, tables, and diagrams—right where the conversation happens. What is Canvas? Canvas is a built-in editor that appears automatically when ELITEA generates code, tables, or Mermaid diagrams in a chat. It allows you to edit, refine, and export AI-generated content without leaving the conversation. Canvas Features: How to Use Canvas:

1. Ask ELITEA to generate code, a table, or a Mermaid diagram.
2. When ELITEA generates the content, look for the pencil icon (✏️) in the top-right corner of the content block.
3. Click the icon to open the Canvas editor in a modal window.
4. Make your edits using the available tools:
   • Copy: Copy content to clipboard
   • Undo/Redo: Revert or reapply actions
   • Save: Save your changes
   • Export: Download in various formats
5. Click Save to preserve your changes in the conversation or Export to download files.

For detailed information, real-world examples, and best practices, see Canvas in Conversation.

​

Attaching Files and Images

Attach files and images directly to chat conversations for AI-powered analysis. This feature enables multimodal interactions where AI can process visual content, documents, and data files alongside text-based queries. Every file you attach is automatically uploaded to the default attachments bucket — no manual Artifact Toolkit configuration is required. Files are accessible from the Artifacts section.

• Images are sent directly to the LLM for real-time vision analysis.
• Non-image files (documents, code, data) are indexed into a vector database and retrieved via semantic search.

Attachment Limits: How to Attach Files:

1. In the chat input toolbar, click the + icon.
2. Click Attach Files.
3. The native file browser opens — select one or more files.
4. Selected files appear as chips above the input field. To remove a file, click the × on its chip.
5. Type your message and press Send.

You can also attach files by:

• Drag and drop — drag files directly onto the chat input area.
• Paste — paste files or images from your clipboard with Ctrl+V / Cmd+V.

Availability by Chat Type: For detailed information about attachments, including agent configuration and file management, see Attach Files.

​

Using Modules

Modules provide built-in capabilities that enhance your conversations without requiring external integrations. They can be enabled directly from the chat interface or configured as part of an agent’s default setup. Available Modules Enabling Modules in Chats

1. Go to your chat.
2. In the chat toolbar, select +.
3. Select Modules.
4. Toggle the selected module on or off to enable or disable it.

Once enabled, the AI assistant can automatically use these tools during the chat when appropriate. Enabling Modules in Agents You can also enable modules as part of an agent’s default configuration. This makes the tools available in all new chats using that agent.

1. Navigate to Agents in the main menu and select the agent.
2. Expand MODULES.
3. Toggle the desired tools ON.
4. Click Save.
5. The enabled modules will be available in all new chats using this agent.

Using Modules Once enabled, the AI assistant can use the modules during chats:

• Python Sandbox: The assistant can execute Python code, install packages, perform calculations, and generate visualizations
• Planner: The assistant can break down complex tasks, create structured plans with priorities and due dates, and track task progress
• Data Analysis: The assistant can perform comprehensive data analysis on uploaded files using natural language commands

For detailed usage examples, troubleshooting, and best practices, see:

• Python Sandbox Guide
• Planner Guide
• Data Analysis Guide
• Smart Tools Selection Guide
• Swarm Mode Guide
• Agent & Pipeline Builder Guide

​

Sensitive Action Authorization

When an agent or pipeline attempts to call a sensitive tool (such as deleting a repository, running a shell command, or dropping a database table), the conversation automatically pauses and displays an authorization card before anything is executed. Authorization Dialog Elements: Actions:

• Authorize — Approves the tool call. Execution resumes and the tool runs as planned.
• Block — Rejects the tool call. The tool is skipped entirely; the agent receives a cancellation message and continues or stops based on its logic.

For full details including configuration, pipeline behavior, and troubleshooting, see Sensitive Action Authorization Guardrail.

​

Voice Capabilities

ELITEA Chat includes voice input and voice output capabilities that let you speak your messages and have AI responses read back to you.

​

Voice Input

Voice Input lets you dictate messages directly into the chat input field using a microphone. The transcribed text is inserted at the cursor position, so you can combine typed and spoken content in the same message. How to use Voice Input:

1. Click the microphone icon in the message input toolbar to start recording.
2. Speak your message. A live transcript appears in the input field as you talk — interim results update in real time.
3. Click the Stop (■) button to finish recording. The final transcript is committed and focus returns to the input field.

​

Text-to-Speech

Text-to-Speech reads AI responses aloud. A Read out button (megaphone icon) appears in the action bar below each AI message when the message contains speakable text. How to use Text-to-Speech:

1. Click the Read out (megaphone) button below an AI message to start playback.
2. The text is read aloud and a playback pill appears in the input area with:
   • A Stop (■) button to halt audio playback at any time.
   • A Voice settings (⚙) gear icon — opens the Voice Settings dialog (disabled while playback is active).
3. After stopping, a Play (▶) button replaces the Stop button to restart playback from the beginning.
4. Click the ⚙ gear icon (only available when playback is stopped) to open the Voice settings dialog.

Voice Settings: Click the ⚙ gear icon in the playback pill to open the Voice settings dialog. Settings are saved to your browser and persist across sessions. Click Apply to save, or Cancel to discard changes.

​

Speaking Mode

Speaking Mode is a continuous, hands-free voice conversation loop. Once activated, ELITEA automatically records your speech, sends your message after a pause, plays back the AI’s response, and then starts listening again — without any manual interaction between turns. How to activate Speaking Mode:

1. Leave the message input field empty.
2. Click the voicewave icon (shown in place of the Send button when the input is empty) to enter Speaking Mode.
3. A voicewave pill with an Exit (✕) button replaces the send button, indicating Speaking Mode is active.
4. Speak your message. After 3 seconds of silence, your message is automatically sent.
5. The AI response streams in and is read aloud via Text-to-Speech.
6. Once the response finishes playing, recording starts again automatically for the next turn.
7. Click the ✕ button in the voicewave pill to exit Speaking Mode at any time.

​

Playback Mode

Playback mode allows you to step through an existing conversation turn by turn, exactly as it happened, without actually sending requests to the Models, Toolkits, etc. Example: Using Playback Mode for a Demo Scenario:
A Product Manager is preparing a demo for stakeholders to showcase how the team collaborated on a new feature. They use Playback Mode to simulate the conversation and highlight key decisions and actions. Steps to Use Playback Mode: Access the Conversation:
The Product Manager navigates to the conversation in the CONVERSATIONS sidebar where the team discussed the feature. Activate Playback Mode:
The Product Manager right-clicks on the conversation and selects Playback from the context menu. Simulate the Conversation:

• The playback starts with the BA’s initial message outlining the feature requirements.
  Example:
  @BA: "The new feature should allow users to upload files up to 10MB in size. The system must validate file types and provide error messages for unsupported formats."
  
• The PM uses the Next control to move to the next message, where they set the timeline for development and testing.
  Example:
  @PM: "The development team will complete the implementation by May 15th. QA can start testing on May 16th, and we aim to release the feature by May 20th."
  
• The QA’s message about testing strategies is reviewed next.
  Example:
  @QA: "I will create test cases for file uploads, including valid and invalid file types, file size limits, and error handling."
  

Highlight Key Decisions:

• The Product Manager pauses playback to explain the rationale behind certain decisions, such as the timeline or testing strategy.
• They resume playback to show the BA’s clarification about drag-and-drop uploads.
  Example:
  @BA: "Yes, drag-and-drop uploads should be supported. Additionally, the system should display a progress bar during uploads."
  

Conclude the Demo:

• The Product Manager stops playback after showcasing the final message tracking progress.
  Example:
  @PM: "QA, please share the test results by May 18th so we can review them before the release."
  

Benefits of Playback Mode for Demos

• Polished Presentations: Playback Mode ensures a smooth and professional demo without interruptions or delays.
• Clarity: Stakeholders can see the exact flow of discussions and decisions.
• Engagement: The step-by-step simulation keeps the audience engaged and focused on key points.

By using Playback Mode, teams can effectively demonstrate their collaboration and decision-making processes to stakeholders, ensuring transparency and alignment.