Suggestions regarding the AI ​​agent usage box in Zed

[weimin69]

Body

What are you proposing?

Add clearer visual distinction between the user’s messages and the AI agent’s responses in the Agent Panel chat interface.

Why does this matter?

What problem does this solve?

Currently, user prompts and agent responses look very similar in the conversation thread. There’s almost no visual separation (color, background, alignment, or styling), which makes long conversations hard to scan. It’s easy to lose track of where your input ends and the AI’s output begins, especially during multi-turn agent sessions.
What becomes easier or possible?

Faster scanning of long threads
Clearer understanding of conversation flow
Reduced cognitive load when reviewing previous context or debugging agent behavior
Better overall readability and usability of the Agent Panel

Are there any examples or context?

Most modern AI chat interfaces already solve this well:

Cursor, Claude.ai, ChatGPT, and Windsurf all use different background colors, text colors, or side indicators for user vs assistant messages.
Even simple approaches like left/right alignment, subtle background tints, or avatar/label differences make a big difference.

Possible approach

A few lightweight ideas that would already help a lot:

Different background colors (e.g. slightly darker/lighter panels for user vs agent)
Distinct text color or opacity for user messages
A small label or avatar (“You” / “Agent”) on each message
Left-align user messages and right-align (or differently style) agent responses
A thin accent bar or border on one side of agent messages

Any combination of the above would make the conversation much easier to follow without requiring major UI changes.