What
Claude Code has a behaviour that places some content beside tool results, the output Claude gets back after using a tool.
- The setting that controls it is now read from the environment variable
CLAUDE_CODE_DAPPER_CUPCAKEand the switchtengu_dapper_cupcake. It still defaults to on. Nothing has been read about either switch yet. - The old variable
CLAUDE_CODE_SQUISHY_NEWTand the switchtengu_squishy_newtno longer appear anywhere, so setting the old variable has no effect. - The same helper supplies instruction text mentioning
read_page_dataand "to get its current version". - User messages now get a
besideToolResults: truemark when they meet two checks, one about where the message came from.
Why
If you set CLAUDE_CODE_SQUISHY_NEWT to turn this behaviour off, for yourself or in an automated setup, that setting is now ignored. Use CLAUDE_CODE_DAPPER_CUPCAKE instead.
tengu_dapper_cupcake 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 v2.1.301: 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.
tengu_squishy_newt 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 v2.1.301: 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 v2.1.301. It isn't a statement about your account. What a flag value here can and cannot tell you
New in this build: CLAUDE_CODE_DAPPER_CUPCAKE