Gemini CLI 0.32.0

What's Changed

• feat(plan): add integration tests for plan mode
• fix(acp): update auth handshake to spec
• feat(core): implement robust A2A streaming reassembly and fix task continuity
• feat(cli): load extensions in parallel
• Plumb the maxAttempts setting through Config args
• fix(cli): skip 404 errors in setup-github file downloads
• fix(cli): expose model.name setting in settings dialog for persistence
• docs: remove legacy cmd examples in favor of powershell
• feat(core): Enable model steering in workspace.
• fix: remove trailing comma in issue triage workflow settings json
• feat(core): implement task tracker foundation and service
• test: support tests that include color information
• feat(core): introduce Kind.Agent for sub-agent classification
• Changelog for v0.30.0
• Update changelog workflow to reject nightly builds
• Changelog for v0.31.0-preview.0
• feat(cli): hide workspace policy update dialog and auto-accept by default
• feat(core): rename grep_search include parameter to include_pattern
• feat(plan): support opening and modifying plan in external editor
• feat(cli): implement interactive shell autocompletion
• fix(core): allow /memory add to work in plan mode
• feat(core): add HTTP 499 to retryable errors and map to RetryableQuotaError
• feat(core): Enable generalist agent
• Updated tests in TableRenderer.test.tsx to use SVG snapshots
• Refactor Github Action per b/485167538
• fix(github): resolve actionlint and yamllint regressions from #19443
• fix: action var usage
• feat(core): improve A2A content extraction
• fix(cli): support quota error fallbacks for all authentication types
• fix(core): flush transcript for pure tool-call responses to ensure BeforeTool hooks see complete state
• feat(plan): adapt planning workflow based on complexity of task
• fix: prevent orphaned processes from consuming 100% CPU when terminal closes
• feat(core): increase fetch timeout and fix [object Object] error stringification
• [Gemma x Gemini CLI] Add an Experimental Gemma Router that uses a LiteRT-LM shim into the Composite Model Classifier Strategy
• docs(plan): update documentation regarding supporting editing of plan files during plan approval
• test(cli): fix flaky ToolResultDisplay overflow test
• ui(cli): reduce length of Ctrl+O hint
• fix(ui): correct styled table width calculations
• Avoid overaggressive unescaping
• feat(telemetry) Instrument traces with more attributes and make them available to OTEL users
• Add support for policy engine in extensions
• Docs: Update to Terms of Service & FAQ
• Fix bottom border rendering for search and add a regression test.
• fix(core): apply retry logic to CodeAssistServer for all users
• Fix extension MCP server env var loading
• feat(ui): add 'ctrl+o' hint to truncated content message
• Fix flicker showing message to press ctrl-O again to collapse.
• fix(cli): hide shortcuts hint while model is thinking or the user has typed a prompt + add debounce to avoid flicker
• feat(plan): update planning workflow to encourage multi-select with descriptions of options
• refactor(core,cli): useAlternateBuffer read from config
• fix(cli): ensure dialogs stay scrolled to bottom in alternate buffer mode
• fix(core): revert auto-save of policies to user space
• Demote unreliable test.
• fix(core): handle optional response fields from code assist API
• fix(cli): keep thought summary when loading phrases are off
• feat(cli): add temporary flag to disable workspace policies
• Disable expensive and scheduled workflows on personal forks
• Moved markdown parsing logic to a separate util file
• fix(plan): prevent agent from using ask_user for shell command confirmation
• fix(core): disable retries for code assist streaming requests
• feat(billing): implement G1 AI credits overage flow with billing telemetry
• feat: better error messages
• fix(ui): persist expansion in AskUser dialog when navigating options
• fix(cli): prevent sub-agent tool calls from leaking into UI
• fix(cli): Shell autocomplete polish
• Changelog for v0.31.0-preview.1
• Add slash command for promoting behavioral evals to CI blocking
• Changelog for v0.30.1
• Add low/full CLI error verbosity mode for cleaner UI
• Disable Gemini PR reviews on draft PRs.
• Docs: FAQ update
• fix(core): reduce intrusive MCP errors and deduplicate diagnostics
• docs: fix spelling typos in installation guide
• Promote stable tests to CI blocking.
• feat(core): enable contiguous parallel admission for Kind.Agent tools
• Enforce import/no-duplicates as error
• fix: merge duplicate imports in sdk and test-utils packages (1/4)
• fix: merge duplicate imports in a2a-server package (2/4)