> ## Documentation Index
> Fetch the complete documentation index at: https://docs.useklaws.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Chat

> Talk to your agent from the web.

The Chat screen is the first thing you see after deploying your agent. It's where you have conversations, ask questions, and give instructions.

## Key features

* **Streaming responses** — answers appear word by word
* **Image support** — click the **+** button to attach an image. Your agent can see and analyze it
* **Voice mode** — click the mic icon to speak. Your message is transcribed and sent automatically. Click the speaker icon to hear responses read aloud
* **Quick suggestions** — when the conversation is empty, tap a suggestion to get started
* **Smart actions** — after your agent completes a task, action buttons appear automatically (e.g. "View in Canvas", "View in Tasks")
* **Tool progress** — while your agent works, you'll see what it's doing ("Browsing web...", "Writing file...")
* **Smart models** — your agent automatically picks the best AI model for each message
* **Chat history** — your conversations are saved and persist between sessions

## Tips

<Tip>
  Press **Enter** to send a message. Use **Shift+Enter** for a new line.
</Tip>

* You can stop a response mid-generation by clicking the stop button
* Click **Clear chat** to start a fresh conversation
* Your agent remembers context from previous messages in the same session

<Info>
  Your agent shares the same brain and memory across web chat and Telegram. A conversation on Telegram will inform responses on the web, and vice versa.
</Info>