Unclear It is not stated what this check controls or what the fallback value change does.
What
Organisation policy limits are rules an organisation sets on what its members can do in Claude Code. A new check now decides whether those limits need to be applied to the current session. The check is skipped when any of these is true:
CLAUDE_CODE_REMOTEis set, which means the session is running in Anthropic's cloudDISABLE_LOGIN_COMMANDis set- a trusted device token (
CLAUDE_TRUSTED_DEVICE_TOKEN) is present - the user is not a claude.ai subscriber
In every other case, Claude Code waits for the policy limits to load, up to a time limit.
The built-in fallback for tengu_org_policy_denied also changed from 60 to 1. A fallback is the value Claude Code uses when it has no value from Anthropic's servers. What this setting controls was not traced, and nobody has yet checked what Anthropic's servers return for it.
Why
In practice, claude.ai subscribers using Claude Code on their own machine may see Claude Code wait briefly for policy limits to load. Cloud sessions, trusted devices and setups with login disabled skip that wait. The new fallback value only matters when no server value is available, and its effect is not known.
tengu_org_policy_denied 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 v2.1.296: not a boolean we can read
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 v2.1.296. It isn't a statement about your account. What a flag value here can and cannot tell you
It is not stated what this check controls or what the fallback value change does.