[FEATURE] Keep Quick Entry Responses in Overlay on Current Screen

Preflight Checklist

• [x] I have searched existing requests and this feature hasn't been requested yet
• [x] This is a single feature request (not multiple features)

Problem Statement

Problem Statement

The current quick entry feature for Claude Desktop on Mac is a great shortcut, but it breaks the user's workflow context by switching to the full Claude app window when displaying responses. This contrasts with ChatGPT's Mac overlay, which keeps the user on their current screen with a compact response window.

Current behavior:

1. User double-taps Option key from any app
2. Quick entry overlay appears at bottom of screen
3. User types question or captures screenshot
4. User sends query
5. Claude Desktop app window opens, pulling focus away from the original task
6. User must switch back to their original app/screen to continue work

This extra context switch defeats some of the efficiency gains that quick entry provides, especially for quick lookups or brief consultations while focused on other work.

Proposed Solution

Desired Behavior

Keep the response in a compact overlay window on the user's current screen, similar to ChatGPT's implementation:

1. User double-taps Option key from any app
2. Quick entry overlay appears at bottom of screen
3. User types question or captures screenshot
4. User sends query
5. Claude processes and returns response in a floating overlay on the current screen (does not switch app focus)
6. User can read response, perform follow-up actions, or dismiss overlay to return to original task
7. Optional: User can click to expand response into full Claude app if needed for longer conversations

Benefits

• Reduced context switching: Users stay focused on their current task
• Faster workflow: No need to manually switch back to the original app after getting an answer
• Competitive parity: Matches ChatGPT's overlay behavior that many Mac users expect
• Better for quick lookups: Small questions get answers without disrupting focus
• Seamless multitasking: Ideal for code review, content review, or quick questions while working in other apps

Suggested Implementation Details

• Response overlay appears in same location as query input (bottom of screen or near original query)
• Overlay is resizable and dismissible
• User can click an expand button to open full Claude window for longer conversations
• Optional: Add a setting in Quick Access preferences to customize overlay behavior
• Optional: Set a timeout to auto-dismiss responses after a certain period of inactivity

Alternative Solutions

_No response_

Priority

High - Significant impact on productivity

Feature Category

Other

Use Case Example

Use Cases

1. Code review: Share error message with Claude, get instant explanation, stay in code editor
2. Writing assistance: Highlight text, ask for help, read suggestion without leaving document
3. Content research: Ask Claude a quick question while reading/writing, get answer on current screen
4. Debugging: Share console output, get diagnosis, stay in development environment
5. Quick lookups: Ask syntax questions, get explanations, remain in current workflow

Additional Context

Additional Context

Quick entry is already a valuable feature, but this enhancement would make it feel less interruptive to the user's primary workflow. The goal isn't to replace the full Claude app experience, but to provide a true "overlay-style" quick access that doesn't demand focus switches.

This aligns with how other AI overlays work on Mac (ChatGPT, Perplexity) where the user can get quick answers without context switching.