Skip to main content
The Profy chat interface is your primary workspace for collaborating with Experts. The entire interface consists of three major areas: the message area on the left, the input area at the bottom, and the workspace panels on the right. Each area is designed around the philosophy of “conversation as collaboration,” enabling you to complete the full loop from communication to delivery within a single conversation.

​
Interface Layout

Message Area

The complete timeline of conversation history, including your messages, Expert replies, tool call processes, phase groupings, and more

Input Area

The message editor at the bottom, supporting text, voice, file uploads, as well as advanced features like model switching and plan mode

Workspace Panels

Expandable multi-tab panels on the right side, displaying previews, websites, files, browser, and other real-time work outputs from the Expert

​
Message Area

The message area displays the complete conversation process in a timeline format.

​
Message Types

  • Your messages: Text, images, file attachments, and voice-transcribed text
  • Expert replies: Markdown rich text, code blocks, tables, images, and more
  • Tool calls: Every tool the Expert invokes during task execution is displayed as a card showing the process and results
  • Phase groupings: Complex tasks are automatically grouped into multiple phases that you can collapse or expand for details

​
Interaction Options

Each Expert reply supports the following actions:
  • Copy: Copy the reply content with one click
  • Read aloud: Click the speaker icon to have the Expert’s reply played as audio
  • Feedback: Help the Expert continuously improve by giving a thumbs up or thumbs down
  • Regenerate: Have the Expert re-answer when you’re not satisfied with the current reply
During a conversation, if the Expert’s reply is still being generated, you can click the “Stop” button at any time to interrupt the current output. After interrupting, you can choose “Continue generating” to resume the unfinished content.

​
Input Area

The input area is located at the bottom of the interface and is where you send instructions to the Expert.

​
Text Input

Type your request directly in the input box. Multi-line text is supported — press Enter to send and Shift + Enter for a new line.

​
”+” Attachment Menu

Click the ”+” button on the left side of the input box to access the following features:
Select files from your device to upload into the conversation. Supports images, documents, code files, archives, and many other formats. Files are automatically stored in your cloud drive.
Select existing files from your Profy cloud drive to insert into the conversation without re-uploading.
Take a photo or select images from your gallery to send to the Expert for analysis.
Enable voice input mode to transcribe speech into text before sending.

​
Model Selector

The model selector in the input box allows you to switch the AI model before sending each message. Different models offer different trade-offs between speed, accuracy, and cost.
Model lists and rates are operational data that track upstream providers, so the selector’s current values are authoritative. The stable rule is that models are gated by plan tier, ranked free < pro(Plus) < max(Pro) < flagship < team < enterprise.

​
Plan Mode Toggle

You can activate “Plan Mode” from the ”+” menu. When enabled, the Expert won’t execute tasks directly — instead, it will first create a detailed plan document for you to review and iterate on before execution. See Plan Mode for details.

​
Conversation Budget

The conversation budget is not a fixed quota configured somewhere in the backend. It is computed from your balance every time you send a message:
In other words, “how many output tokens can my balance still afford.” So the budget moves with your balance, and switching to a more expensive model shrinks it immediately.
When the budget runs out, the streaming reply is deliberately cut off rather than left to fail at deduction time. An amber notice appears with two actions: start a new conversation, or top up.Two kinds of exhaustion are worth distinguishing, because they call for different fixes:The first is routinely mistaken for the second. It happens because a long conversation’s context keeps growing, so each turn costs more — you may still hold plenty of credits and simply not have enough to feed that large context once more. Which is why “start a new conversation” genuinely solves it.
Budget exhaustion is a bounded cut-off, not a silent failure: the runtime emits a budget_exhausted event and stops the stream once cumulative usage passes the budget, which prevents overdraft. There is also an internal protection you never see — tool results are capped in aggregate (80,000 tokens by default) and individually (20,000 tokens by default), and are truncated before entering the model context. That truncation does not interrupt the conversation.Verified 2026-08-11. Sources: services/core/src/routes/agent-proxy/invoke-lifecycle.ts (tokenBudget derivation), services/agent-runtime/src/base/llm/budget_cap.py, services/agent-runtime/src/harness/context/compaction.py.

​
Expert Selection

If you have multiple Experts, you can switch the currently active Expert at the top of the conversation. Each Expert has its own independent skills, memories, and conversation history.

​
Conversation Management

1

Create a New Conversation

Click the “New Conversation” button in the upper left corner to start a fresh exchange.
2

Switch Conversations

In the conversation list on the left, click any historical conversation to switch to it.
3

Fork a Conversation

Create a branch from any message in the conversation to explore different directions. See Conversation Replay & Forking for details.

​
Mobile Adaptation

The Profy chat interface automatically adapts to mobile devices: workspace panels switch to full-screen mode, and the input area stays at the bottom of the screen for easy access. The core interaction experience remains consistent with the desktop version.