How to use the CRHQ conversation view — sending messages, reading responses, viewing tool calls, and managing output.
The conversation interface is where you interact with agents. It's designed to show you exactly what the agent is doing, while keeping the experience clean and readable.
Type in the message input at the bottom and press Enter to send. For multi-line messages, use Shift+Enter for a new line.
The input supports:
Agent responses stream in real-time — you see text appear as the agent thinks and writes. Responses can include:
When an agent uses a tool (reads a file, runs a command, calls an API), the tool call is shown inline in the conversation. You can control how these are displayed:
Each tool call is shown as a separate, expandable block with:
Tool calls are grouped into collapsible sections. You see a summary of what happened without the verbose output. Expand to see details.
Switch between modes in Account → Display Settings.
When an agent produces a visual output (code file, HTML page, document), it appears in the artifact panel on the right side. The panel supports:
Conversations can be long. CRHQ uses pagination — when you scroll to the top, older messages load automatically. This keeps the interface responsive even for sessions with hundreds of messages.