Gemini CLI 0.24.0

What's Changed

• chore(core): refactor model resolution and cleanup fallback logic
• Add Folder Trust Support To Hooks
• Record timestamp with code assist metrics.
• feat(policy): implement dynamic mode-aware policy evaluation
• fix(core): use debugLogger.debug for startup profiler logs
• feat(ui): Add security warning and improve layout for Hooks list
• fix #15369, prevent crash on unhandled EIO error in readStdin cleanup
• chore: improve error messages for --resume
• chore: remove clipboard file
• Implemented unified secrets sanitization and env. redaction options
• feat: automatic /model persistence across Gemini CLI sessions
• refactor(core): remove deprecated permission aliases from BeforeToolHookOutput
• fix: add missing type field to MCPServerConfig
• Make schema validation errors non-fatal
• chore: limit MCP resources display to 10 by default
• Add experimental in-CLI extension install and uninstall subcommands
• feat: Add A2A Client Manager and tests
• feat: terse transformations of image paths in text buffer
• Security: Project-level hook warnings
• Added modifyOtherKeys protocol support for tmux
• chore(core): fix comment typo
• feat: Show snowfall animation for holiday theme
• do not persist the fallback model
• Resolve unhandled promise rejection in ide-client.ts
• fix(core): handle checkIsRepo failure in GitService.initialize
• fix(cli): add enableShellOutputEfficiency to settings schema
• Manual nightly version bump to 0.24.0-nightly.20251226.546baf993
• refactor(core): extract static concerns from CoreToolScheduler
• fix(core): enable granular shell command allowlisting in policy engine
• chore/release: bump version to 0.24.0-nightly.20251227.37be16243
• refactor: deprecate legacy confirmation settings and enforce Policy Engine
• Migrate console to coreEvents.emitFeedback or debugLogger
• Exponential back-off retries for retryable error without a specified …
• feat(agents): add support for remote agents and multi-agent TOML files
• Update wittyPhrases.ts
• refactor(auth): Refactor non-interactive mode auth validation & refresh
• Revert "Update wittyPhrases.ts (#15697)"
• fix(hooks): deduplicate agent hooks and add cross-platform integration tests
• Implement support for tool input modification
• Add instructions to the extensions update info notification
• Add extension settings info to /extensions list
• Agent Skills: Implement Core Skill Infrastructure & Tiered Discovery
• chore: remove cot style comments
• feat(agents): Add remote agents to agent registry
• feat(hooks): implement STOP_EXECUTION and enhance hook decision handling
• Fix build issues caused by year-specific linter rule
• fix(core): handle unhandled promise rejection in mcp-client-manager
• log fallback mode
• Agent Skills: Implement Autonomous Activation Tool & Context Injection
• fix(core): improve shell command with redirection detection
• Add security docs
• feat: add folder suggestions to /dir add command
• Agent Skills: Implement Agent Integration and System Prompt Awareness
• chore: cleanup old smart edit settings
• Agent Skills: Status Bar Integration for Skill Counts
• fix(core): mock powershell output in shell-utils test
• Agent Skills: Unify Representation & Centralize Loading
• Unify shell security policy and remove legacy logic
• feat(core): restore MessageBus optionality for soft migration (Phase 1)
• feat(core): Standardize Tool and Agent Invocation constructors (Phase 2)
• feat(core,cli): enforce mandatory MessageBus injection (Phase 3 Hard Migration)
• Agent Skills: Extension Support & Security Disclosure
• feat(hooks): implement granular stop and block behavior for agent hooks
• Agent Skills: Add gemini skills CLI management command
• refactor: consolidate EditTool and SmartEditTool
• fix(cli): mock fs.readdir in consent tests for Windows compatibility
• refactor(core): Extract and integrate ToolExecutor
• Fix terminal hang when user exits browser without logging in
• fix: avoid SDK warning by not accessing.text getter in logging
• Make default settings apply
• chore: rename smart-edit to edit
• Opt-in to persist model from /model
• fix: prevent /copy crash on Windows by skipping /dev/tty
• Support context injection via SessionStart hook.
• Fix order of preflight
• Fix failing unit tests
• fix(cli): resolve paste issue on Windows terminals.
• Agent Skills: Implement /skills reload
• Add setting to support OSC 52 paste
• remove manual string when displaying manual model in the footer
• fix(core): use correct interactive check for system prompt
• Inform user of missing settings on extensions update
• feat(policy): allow 'modes' in user and admin policies
• fix: default folder trust to untrusted for enhanced security
• Add description for each settings item in /settings
• Use GetOperation to poll for OnboardUser completion
• Agent Skills: Add skill directory to WorkspaceContext upon activation
• Fix settings command fallback
• fix: writeTodo construction
• properly disable keyboard modes on exit
• Add workflow to label child issues for rollup
• feat(ui): add visual indicators for hook execution
• fix: image token estimation
• feat(hooks): Add a hooks.enabled setting.
• feat(admin): Introduce remote admin settings & implement secureModeEnabled/mcpEnabled
• Remove trailing whitespace in yaml.
• feat(agents): add support for remote agents
• fix: limit scheduled issue triage queries to prevent argument list too long error
• ci(github-actions): triage all new issues automatically
• Fix test.
• fix: hide broken skills object from settings dialog
• Agent Skills: Initial Documentation & Tutorial
• fix(patch): cherry-pick b54e688 to release/v0.24.0-preview.0-pr-16284 to patch version v0.24.0-preview.0 and create version 0.24.0-preview.1
• fix(patch): cherry-pick 356f76e to release/v0.24.0-preview.1-pr-16252 to patch version v0.24.0-preview.1 and create version 0.24.0-preview.2
• fix(patch): cherry-pick eda47f5 to release/v0.24.0-preview.2-pr-16557 [CONFLICTS]