What
When Claude Code reads something into the conversation, the result says whether it is worth reading again. Behind the tengu_candid_ledger flag, a switch Anthropic can turn on or off from its servers, a read result can now carry an extra inItsPlace field:
- On a read that is not worth reading again, the result carries
inItsPlacewhen the flag is true and a check passes. Before, it returned onlyblockandisWorthReadingAgain. - On a failed projects-context read, meaning a failed read of a project's context, the result carries
inItsPlacewhen the read is worth reading again, a check passes and the flag is exactly true.
The flag defaults to false. The flag server returned off for this site's account and for an anonymous check, but those readings were not taken under this release.
Why
While the flag is off nothing visible changes. If it is turned on, some reads can come back with something offered in place of the original result.
tengu_candid_ledger Off in both readingsThe flag server returned off for the account this site reads and for the anonymous baseline. A reading of off cannot rule out a rollout these two readings sit outside of.
This account: off · anonymous baseline: off · compiled default in v2.1.319: 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 v2.1.319. It isn't a statement about your account. What a flag value here can and cannot tell you
New in this build: tengu_candid_ledger