Skills
Claude Code v2.1.73: PermissionDenied Hook Fires on Auto-Mode Denials with Retry Control, Named Subagents in @ Mention Typeahead
The latest Claude Code changelog adds a PermissionDenied hook that fires after auto-mode classifier denials — returning {retry: true} tells the model it can retry the tool call, enabling custom policy override logic. Named subagents now appear in @ mention typeahead suggestions for easier multi-agent orchestration. Also: CLAUDE_CODE_NO_FLICKER=1 for alt-screen rendering, fixes for prompt cache misses in long sessions caused by tool schema bytes changing mid-session, and nested CLAUDE.md files being re-injected dozens of times.
↳ Follow the thread