#2124: Feature Request: Conversation Outline / Jump-to-Question Sidebar Panel (对话提纲)

Problem

When conversations grow long (50+ messages), finding earlier discussion points requires extensive scrolling. Users have to manually search through the entire history to locate specific questions or topics discussed earlier. This is especially painful in long research/analysis sessions where many distinct topics are covered.

Proposed Solution

Add a Conversation Outline panel (对话提纲) accessible from the sidebar that:

1. Automatically extracts each user question/message as a clickable summary entry
2. Floats as a fixed-position panel (not inline, so it stays visible while scrolling the main conversation)
3. Clicking any entry jumps directly to that message in the conversation history
4. Supports lazy-loading of truncated older messages when jumping to them

UX Details

• Trigger: a small floating button (e.g. 📋 icon) in the top-right corner of the message area
• Panel: right-side overlay or sidebar panel showing a numbered list of user messages
• Each entry shows: the first ~60 chars of the user's question + a timestamp/duration indicator
• Click behavior: scroll to the target message, auto-load truncated history if needed, briefly highlight the target message
• Toggle: clicking the outline button again or clicking outside the panel closes it
• Active entry: the entry corresponding to the currently visible question is highlighted

Inspiration

This is similar to DeepSeek's web chat interface which provides a sidebar index of all questions asked in a session. It significantly improves navigation efficiency in long conversations.

Implementation Notes

We've prototyped this in our own fork and it works well. The core logic:

• Parse S.messages for all role: 'user' entries
• Store message index positions for jump targeting
• Use jumpToMessage() for navigation with auto-load
• CSS position: fixed for the floating panel

Would love to see this land in an upcoming release!

Priority

High — this is one of the most impactful UX improvements for users who rely on Hermes for extended research/analysis sessions.