Claude Code can now check whether a shell command (a command run in your terminal) changed any of its settings files. It takes a fingerprint of each file before and after the command, based on details such as the file's size and last-modified time, and compares them.
It does this for each level of settings:
- user
- project
- local
- managed (set by an administrator)
- launch
When a level changed, Claude Code records it and tells Claude which levels were affected.
This is off unless it is switched on remotely, and an override can turn it off. Readings taken for this site's account and for an anonymous account both returned off, though none has been taken under this release yet.
A command run during a session could quietly edit the settings that session depends on. This lets Claude know that happened.
tengu_violin_endpin 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: off · anonymous baseline: on · compiled default in v2.1.293: 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 v2.1.293. 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.
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_violin_endpin