As part of our ongoing commitment to improve the user experience on ASI:One, we've introduced a set of UI and functional enhancements across the web platform. These updates are rooted in direct feedback from our user community and aim to make agent interactions more intuitive, seamless, and efficient.
Agent Execution Enhancements
- Execution Status Indicators: Users can now view the correct execution statuses when interacting with agents, improving transparency around agent behavior and response cycles.
- Web Search by Default: All models on ASI:One now have web search enabled by default. This ensures users benefit from real-time information retrieval without having to manually enable the feature.
Agent Page Updates
- Consistent Chat UI: The chat experience on individual agent pages has been upgraded to match the full ASI:One chat interface.
- "My Agent" → "Agent Sandbox": We've renamed the side panel label to better reflect its purpose and reduce ambiguity.
- Image Glimpse Fixes: Resolved issues where images would briefly flash or misalign when navigating to the agent screen.
- Scroll Restoration: Addressed bugs related to improper scroll positions after switching views.
- Clutter free experience: Reduced unnecessary UI elements.
Improvements to @ Mentions
- Improved Input Behavior: Fixed a bug where pressing return after typing an @mention prematurely sent the message.
- Context Persistence: Navigating between chat sessions via the sidebar now maintains correct context, improving session continuity.
- Mention Shortcut Optimization: Clicking "Chat with Agent" now inserts the agent's @mention directly into the current chat input field instead of opening a new session.
UI/UX Adjustments
- Thumb-Down Icon Visibility: Updated the icon color for clearer feedback signaling.
- Padding and Layout Refinements: Adjustments in side panel padding enhance clarity and alignment.
- General Bug Fixes: Numerous small visual and functional fixes to improve overall system stability and polish.
These incremental updates reflect our focus on building an agent-native interface that prioritizes usability without compromising depth. Your feedback continues to shape our roadmap.
Keep it coming, and keep using asi1.ai.
