[FEATURE] Add an opt-in/opt-out setting for the session feedback survey

Status Open
Maintainer reply None cached
Activity 0 comments · opened Sep 10, 2026

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

<img width="383" height="52" alt="Image" src="https://github.com/user-attachments/assets/48f3cad6-81d4-4213-9e10-b8e0e1db7343" />

The in-session feedback survey ("How is Claude doing this session?") fires automatically during active use, with no setting to disable it, it isn't even noticed by Claude Code when triggered. It can appear mid-task — interrupting focused work — regardless of whether the user wants to be prompted at that moment.

As an end user, paying for this for business purposes, this is a) irritating, b) distracting (heaven help I was mid-surgery with Claude driving and this prompt popped up, or serious creative flows with user and Claude 'vibing' away) and c) not being paid to stop what I am doing to click on a thing and think about your question (I am not Claude, I am a paying end user)

Related
Related to #88809, which asks for a manual trigger so the survey can be summoned on happy or unhappy paths alike. That's a different ask — this request is about disabling the automatic firing entirely, not changing when or how it's manually invoked. Giving end users control over (potentially) unsolicited and interrupting visual prompts.

Proposed Solution

Add a settings toggle: Opt In / Opt Out of the survey.

Opt In (default / current behavior) — survey continues to fire exactly as it does today, no other change needed.
Opt Out — the survey never fires for that user. No timing logic, no context-detection required — a single, one-time preference. Given the standards across various other Opt In in marketing and the like, do not impose your want for feedback, without offering a way to not engage. Think of the human elements when injecting UX desires, however subtle on a thing we see with our eyes, and the reactions a brain might have mid-focus.

Alternative Solutions

I asked Claude Code directly whether it could help address or dismiss the prompt when it appeared. It confirmed it has no visibility into this UI element at all — it's part of the host application's own interface, entirely outside anything the assistant can see, trigger, or interact with. So there's currently no workaround on either side: not a user setting, and not something the AI assistant can route around either.

A related issue, #88809, proposes a manual trigger for the survey (letting a user summon it on demand rather than waiting for the automatic prompt). That's a useful but different fix — it adds a way to invoke the survey, but doesn't offer a way to prevent it from firing automatically, so it wouldn't actually solve this problem.

Other software commonly handles this kind of periodic in-app prompt with a "don't show this again" or notification-preference toggle — this request is the same well-established pattern, applied here.

Priority

High - Significant impact on productivity

Feature Category

Interactive mode (TUI)

Use Case Example

Example scenario:

I'm a paying business user working with Claude Code on a real project — reviewing a codebase, running a local dev server, and comparing it against a related app in active development.
Mid-task, with a background process running and my attention on the actual work, an unsolicited "How is Claude doing this session?" prompt appears in the interface, with no way to have prevented it.
With this feature (an Opt In/Opt Out setting), I could have opted out in advance, so the prompt never fires at all for my account.
This would save me the actual cost of the interruption: stopping mid-focus to notice, process, and dismiss a UI element I didn't ask for, then re-orienting back to the task — time and attention I'm not being compensated to spend managing the tool's own feedback requests.

Additional Context

_No response_

View original on GitHub ↗