Dynamic workflow agents wait out a usage limit and automatically re-run once it resets, instead of failing
What
Workflow agent orchestration gained a new pausing mechanism (a state machine referred to as Cn internally) tied to a setting called autoContinueAtUsageLimit, controlled by a flag named tengu_linked_clover (on by default).
- Before spawning a background agent, Claude Code now waits on a "clear to proceed" check, which can hold up the spawn if a usage limit is currently in effect.
- If a usage limit is hit, the run computes a resume time from the limit's reset time, caps how long or how many times it will wait, and shows progress messages such as "Usage limit reached ... Workflow paused; waiting agents re-run shortly after the reset" and "Usage limit reset. Re-running N waiting agent(s)."
Why
Instead of a workflow run simply failing when it hits a usage limit, waiting agents are paused and automatically resumed once the limit resets, so multi-agent workflow runs are more resilient to hitting rate limits mid-run.
tengu_linked_clover 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 v2.1.271: on
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 v2.1.271. It isn't a statement about your account. What a flag value here can and cannot tell you
New in this build: tengu_linked_clover