A new gate, tengu_parsed_willow (on by default), marks certain queued-command turns as human-authored, overridable via CLAUDE_CODE_PARSED_WILLOW
What
Claude Code now marks certain queued-command turns with humanTurn: true, meaning they are treated as if a human typed them. This behavior is controlled by a new gate called tengu_parsed_willow, which defaults to true (on).
- The gate can be overridden with a new environment variable,
CLAUDE_CODE_PARSED_WILLOW. - If that environment variable is not set, Claude Code falls back to the
tengu_parsed_willowgate's default value of true.
Why
This lets Claude Code (or an admin) control whether queued commands are treated as human-authored turns, with the environment variable giving a way to override the default behavior directly rather than waiting on the gate's rollout.
tengu_parsed_willow 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 v2.1.277: 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 v2.1.277. It isn't a statement about your account. What a flag value here can and cannot tell you
New in this build: CLAUDE_CODE_PARSED_WILLOW