Claude Code Projects now adds explanatory text distinguishing messages sent directly in a thread from ones relayed via a coordinator, behind a killswitch
What
A new block of prompt-construction logic builds explanatory "lead" text that tells Claude how a project member's message reached it. Previously the prompt distinguished only messages relayed via a coordinator session or returned from thread-fetch tools; it now also marks a third case: a message the person sent directly in the current thread, with a lead saying it was "sent here, not relayed". This is wired into the message-formatting path and controlled by a gate function (tengu_projects_thread_reply_lead_killswitch) that decides whether the new behavior is active.
Why This helps Claude correctly understand how a message in a Claude Code Projects thread reached it, so it doesn't treat a directly-sent reply as if it had been relayed through a coordinator. The behavior sits behind a killswitch that defaults off, so it can be disabled if it causes problems.
tengu_projects_thread_reply_lead_killswitch 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 v2.1.274: 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 v2.1.274. It isn't a statement about your account. What a flag value here can and cannot tell you
Whether this labeling is active for any given account is unknown, since nothing has been read yet about the…
New in this build: tengu_projects_thread_reply_lead_killswitch