Chat Interface
The chat panel is your primary way of interacting with your AI agent. It sits on the right side of the workspace and is available whenever your workspace is open.
Sending Messages
Type your message in the input field at the bottom of the chat panel and press Enter to send. To insert a newline without sending, use Shift + Enter.
Attaching Files
Click the paperclip icon (📎) next to the input field to attach files to your message. Attached files are shared directly with your agent for that message — images, documents, spreadsheets, and code files are all supported. For persistent document knowledge, consider adding files to your Knowledge Base instead.
Streaming Responses
Your agent's responses appear in real time as they are generated. You do not need to wait for the full response before reading — text streams in word by word as the model produces it.
Managing Running Operations
If your agent is performing a long-running task — such as executing code, searching the web, or processing a large file — you will see a progress indicator in the chat. Click the Cancel button that appears during active runs to stop the operation at any point.
Clearing Chat History
To clear your conversation history, click the trash icon (🗑️) at the top of the chat panel. A confirmation dialog will appear before anything is deleted. Note that clearing history does not affect your agent's configuration or knowledge base — only the visible conversation is removed.
Per-Message Usage Details
Each message in the chat includes a usage summary that expands when you click on it. This shows:
- Input tokens — the tokens sent to the model (your message plus context).
- Output tokens — the tokens generated by the model in its response.
- Cached tokens — tokens served from the model's context cache, which reduces cost and latency.
- Time — how long the model took to generate the full response.
- Tool calls — a log of any tools the agent invoked while processing your message.
For more on what tokens mean for your plan, see Token Usage.