Nine guard bugs fixed in one wave.
- Question cards render again on current Claude Code builds
- Guards and tools that reported success while doing nothing now fail properly
- /careful catches chained, substituted, and capital-flag deletes instead of waving them through
- Design CLI fails loudly on bad flags instead of billing for a guess
- Shared team brains get brain-aware planning instead of silent suppression
- AskUserQuestion orphaned on Claude Code 2.1.89+
- /context-restore loading a sibling worktree's checkpoint
- /sync-gbrain drift re-register on gbrain 0.42+
- Developer-profile double counting
- One-way-door credential net plurals and runtime wiring
- Thin-client brains misclassified as broken
Every fix ships with a tripwire that proves the guard actually guards. This release closes out the silent-failure class across gstack: guards and tools that reported success while doing nothing. Question cards render again on current Claude Code builds. /careful catches chained, substituted, and capital-flag deletes it used to wave through. The design CLI fails loudly on bad flags instead of billing you for a guess. Shared team brains (thin clients) get brain-aware planning instead of silent suppression. Four of the fixes came from community PRs, absorbed with authorship intact and hardened on top. Fixed: AskUserQuestion orphaned on Claude Code 2.1.89+. /careful chained-rm bypass. /context-restore loading a sibling worktree's checkpoint. /sync-gbrain drift re-register on gbrain 0.42+. Developer-profile double counting. One-way-door credential net: plurals + runtime wiring. design CLI silent NaN flags. Thin-client brains misclassified as broken.