What
Claude Code gains a survey that an organisation can send to its members:
- It fetches a pending survey from an organisation survey service, then records milestones, dismissals and submissions.
- The survey has select, multi-select and text questions, then a review step that says answers won't be used to improve Anthropic's services.
- It runs only for people signed in directly with Anthropic on an enterprise subscription, and is skipped when
CLAUDE_CODE_DISABLE_FEEDBACK_SURVEYis set. - Checks for a new survey are spaced out using a stored
orgSurveyCheck.nextCheckAt, every 6 hours by default. - The survey area above the prompt gains an
orgSurveycard with start, dismiss and ignore actions, wired in next to the memory, feedback and frustration surveys. The only trigger found is an entry{ survey: "orgSurvey", passiveCard: "open" }.
The feature is controlled by the server flag tengu_steady_mitten, whose built-in fallback value is false, so it shows only where the server enables it. The same release also adds a low_resource_notice name to a list of notices.
Why
If you are on an enterprise plan, your organisation may be able to ask you questions inside Claude Code. Setting CLAUDE_CODE_DISABLE_FEEDBACK_SURVEY keeps it away along with the other surveys.
tengu_steady_mitten Not enough to sayNothing here resolved what this flag was doing on this version, so nothing here should be read as on or off.
This account: no value returned · anonymous baseline: no value returned · compiled default in v2.1.304: off
These values were read against a different version of Claude Code, so treat them as the nearest reading available instead of one taken on this release.
Read once, for one account on one subscription tier, against v2.1.304. It isn't a statement about your account. What a flag value here can and cannot tell you
A reading is one sample. Claude Code evaluates its flags remotely, so no client sees the targeting rule behind a value and this says nothing about your account.
A reading is one sample. Claude Code evaluates its flags remotely, so no client sees the targeting rule behind a value and this says nothing about your account.
A reading is one sample. Claude Code evaluates its flags remotely, so no client sees the targeting rule behind a value and this says nothing about your account.
A reading is one sample. Claude Code evaluates its flags remotely, so no client sees the targeting rule behind a value and this says nothing about your account.
A reading is one sample. Claude Code evaluates its flags remotely, so no client sees the targeting rule behind a value and this says nothing about your account.
A reading is one sample. Claude Code evaluates its flags remotely, so no client sees the targeting rule behind a value and this says nothing about your account.
New in this build: tengu_steady_mitten