Support a comment-based referencing feature for agent output.

Problem

During multi-turn interactions with an agent, users often need to repeatedly rephrase their requests to refer to or modify content from the agent’s previous response, which is time-consuming and cumbersome.

Solution

We hope to implement a commenting feature similar to Antigravity’s: users can select specific portions of the agent’s output and add comments directly. In the next interaction, these comments would be automatically included in the conversation context.

This would significantly improve usability and workflow efficiency.

Use Case

especially during extended multi-turn dialogues with the agent.

Priority

  • :red_circle: High - Blocking issue
  • :yellow_circle: Medium - Important improvement
  • :green_circle: Low - Nice to have

Additional Info

(Optional: screenshots, examples, links)

Thank you for the feedback — it’s a good suggestion. We’ll create an internal ticket to analyze and evaluate it.