Follow Discord
Sweep 25 Sep 2026 · 19:33Z Build v2.1.283 504 read Stable v2.1.274 Latest v2.1.283 Next v2.1.283 Feeds RSS JSON llms.txt llms-full.txt Unofficial
HomeReleasesv2.1.281entry #325

Claude Code v2.1.281 ·

Warning when you switch account or organization mid-conversation, behind a flag

Claude Code records organization changes mid-conversation and can warn that the next reply may be slower and use more tokens, behind a flag

Group of 3 Nothing to try yet In Development
JSON All of v2.1.281
Nothing to try yetTier: how much it should matter to you
2Useful: my rating, 1 to 5
3Signal: worth watching, 1 to 5
AuthArea: what it touches
In DevelopmentKind: in v2.1.281,
In DevelopmentSection of the release

What

  • New conversation record types credential_org and advisor_stripped are added, and the session log accepts both.
  • credential_org records which organization you are signed in to. If it changes mid-conversation, a warning can appear: "Signed in to a different account or organization than earlier in this conversation · the next response may be slower and use more tokens · /login switches accounts".
  • The warning shows only when all of these hold: it is the main interactive session, earlier replies contain thinking blocks (the model's recorded reasoning), the model has the new org_locked_thinking capability, and the tengu_lexical_glade flag is on. Otherwise the change is only recorded. The flag's built-in default is off.
  • advisor_stripped is added when advisor content is removed from the conversation.

Why

If you sign in to a different account partway through a conversation, for example with /login, the next reply may take longer and use more tokens, and this warning is how Claude Code would tell you. Flag readings taken before this release found tengu_lexical_glade off, and none has been taken under this release.

Read from
Feature flag
tengu_lexical_glade Off by default, switched on for this account

The shipped code defaults this off, and the flag server returned on for the one account this site reads on this version. That is the reading that makes the entry above worth a second look, and it still says nothing about your account.

This account: on · anonymous baseline: on · compiled default in v2.1.281: 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 v2.1.281. It isn't a statement about your account. What a flag value here can and cannot tell you

What has happened since
Flag reading moved The flag server now returns on for tengu_lexical_glade, read as this 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.
Flag reading moved The flag server now returns off for tengu_lexical_glade, read as this 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.
Flag reading moved The flag server now returns on for tengu_lexical_glade, read as this 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.

See this across every release →

How sure we are
Two sources agreeTwo things we can check say the same as this entry.
Anthropic's release notes agreeAdded telemetry.resource_attributes to the Claude apps gateway config, to put fixed labels on the telemetry of Claude Desktop and /login…
The name it cites is new in this buildNew in this build: tengu_lexical_glade

See this entry in the whole of v2.1.281 →

Feedback