Use the new /plan command in Antigravity to:

• Keep actions safe (biases toward research & read-only tools)

• Map out complex tasks before touching code

• Generate a clear, structured artifact plan every time

Give it a spin and tell us what you build!

TLDR: Antigravity 2.11.0 adds generative UI to render HTML artifacts inline in chat, alongside several UX improvements and bug fixes.

Improvements

• Added support for referencing and inlining external files directly using u/path/to/file syntax within AGENTS.md and custom rule files.
• Added support for rendering YAML frontmatter as a formatted metadata card in markdown files and artifacts.
• Added support for KaTeX math formatting, Chart.js, and Plotly charts in generative UI widgets and artifact previews.
• Added the ability to split terminals side-by-side in the terminal panel using a dedicated split button or keyboard shortcut (Ctrl+Shift+5 on Linux/Windows, Cmd+\ on Mac), complete with draggable divider resizing and tree indicators in the terminal list.
• Added the Darcula theme preset and syntax highlighting color palette.
• Added support for viewing specific page ranges (StartPage and EndPage) in multi-page documents (like PDFs) and specifying image resolution (MediaResolution) in the view_file tool.
• Added support for hover preview card on skills chip in chat.
• Added support for defining a rules: key in custom markdown agent frontmatter to bind specific rule files directly to an agent.
• Added support for discovering custom skills, agents, and rules defined in skills.json, agents.json, and rules.json configuration files located within project subdirectories.
• Several overall UX improvements for performance, accessibility, and project configurations.

Fixes

• Fixed an issue where slash commands unavailable to a selected custom agent were incorrectly shown in the command menu.
• Fixed an issue where image attachments in the message composer were lost when switching between conversations.
• Fixed an issue where Ogg audio and video files were assigned generic MIME types that could cause model errors.
• Fixed an issue where custom color themes with transparency or alpha channels caused UI text to render in red.
• Fixed an icon mismatch where the title bar overflow menu displayed a generic terminal icon instead of matching the tab icon in the side panel.
• Fixed a memory leak where uncommitted conversation renders could leave agent state updates running in the background.
• Fixed a memory leak where rapidly switching between conversations could leave background streaming connections open indefinitely.
• Fixed an issue where clicking a comment link would not scroll to the comment if it was located below the visible fold in virtualized file and diff viewers.
• Fixed an issue where project folder names with special characters or spaces were improperly decoded twice.
• Fixed an issue where newly added files in the review pane displayed an inaccurate added line count due to trailing newlines.
• Fixed an issue where users authenticated via Workforce Identity Federation (WIF / Enterprise sign-in) were logged out prematurely after one hour due to token refresh endpoint routing.
• Fixed a z-index stacking conflict where inline file change panels overlapped the @-mention dropdown menu.
• Fixed an issue where the side panel overflow menu incorrectly marked the last-clicked action with a checkmark.
• Fixed an issue where form controls and scrollbars inside generative UI widgets did not adapt to the dark or light theme.
• Fixed an issue where Enterprise users authenticating via Application Default Credentials could erroneously see an error message.
• Fixed cold-start token endpoint routing to prevent HTTP 429 resource exhaustion errors on launch, and cached account authentication status to eliminate redundant backend checks on user interactions.
• Fixed an issue where long terminal commands in permission prompts caused the popup layout to overflow.
• Fixed an issue where rapid consecutive agent status updates caused repeated notification chime sounds.
• Fixed an issue where expanded tool steps and collapsible sections could collapse or remount when earlier steps updated during long conversations.
• Fixed an issue where disabled menu items displayed an active pointer cursor instead of a disabled cursor.
• Fixed an issue where previously configured command permission allow and deny rules were ignored after migrating to the updated permissions system.
• Disabled the Git Push action in repositories without a configured remote, and added clear tooltips explaining why VCS actions are disabled.
• Fixed an issue where pasting text into the integrated terminal using Ctrl+V failed unless clipboard read permissions were explicitly granted.
• Fixed an issue where artifacts created in nested subdirectories could fail to appear in the artifacts panel.
• Fixed an issue where signed-in users clicking "Manage" in Settings briefly saw an unauthenticated account overlay.
• Fixed an issue where selecting text in an assistant response failed to show the "Quote" button if the mouse drag released below the message bubble.
• Fixed authentication race conditions that could cause error loops when signing in again or leave stale account state after signing out.
• Fixed an issue where a single corrupted or empty project file could cause the entire projects sidebar to appear empty.
• Fixed an issue in the file viewer where selecting a single code token or word failed to display the comment button.
• Fixed an issue where temporary network disruptions or DNS lookup timeouts during startup caused permanent account failure screens instead of automatically retrying.

To explore the full list of improvements and bug fixes, read the complete release notes on our Changelog, or visit the Download page to update to the latest version of Antigravity.

TLDR: Adds /voice dictation with mic-serve SSH forwarding, embedded ripgrep for consistent agent code search, custom status line cost tracking, granular script runner permission suggestions, and UTF-8 file editing and MCP error fixes.

Improvements

• Added /voice dictation to transcribe speech directly into the prompt with f5 and /voice toggle shortcuts, introduced the mic-serve subcommand to forward a local microphone over SSH, and improved allotment messaging for concurrent stream #.
• Added a cost field to the status line data model, exposing unrounded estimated session costs to track running token spend.
• Improved agent code search by bundling an embedded ripgrep binary instead of invoking external system binaries, increasing search performance and cross-platform consistency.
• Improved the always-proceed permission mode to auto-approve MCP tool calls and page reads without prompting.
• Improved allow-always permission suggestions for script runners (npm run, yarn, pnpm, cargo run) by scoping approvals to specific script names.
• Improved conversation titles by generating them automatically on creation, providing meaningful session names in the resume picker.
• Improved error diagnostics when an MCP server configured for Google credentials cannot locate Application Default Credentials (ADC).

Fixes

• Fixed corrupted file edits in documents containing non-ASCII text (CJK characters, accented letters, emoji) caused by inexact offset matches producing invalid UTF-8.
• Fixed sessions stalling mid-response when a tool result or file diff contained invalid UTF-8.
• Fixed file writes being reported as failures after successfully saving to disk, preventing unnecessary agent retry loops on large files and notebooks.
• Fixed precedence resolution so explicitly configured skill and plugin paths take priority over automatically discovered customizations during name collisions.

To explore the full list of improvements and bug fixes, read the complete release notes on our Changelog, or visit the Download page to update to the latest version of Antigravity.

[effacé]

TLDR: This update focuses on core performance, workspace polish, and reliability—delivering faster startup times, improved sidebar navigation, and cleaner error recovery alongside bug fixes.

Improvements

• Improved performance across the app: faster startup and project switching, faster loading of conversations with long histories, less lag during active sessions, and smoother opening and collapsing of the sidebar and side panels.
• Added the ability to collapse top-level sidebar sections. The ungrouped list now shows up to 30 conversations, and conversation actions appear instantly on hover.
• Notification toasts now appear in the bottom-right corner and collapse into a compact stack that expands on hover, so they no longer cover the screen.
• Hovering over a file's comment count pill now previews all comments on that file alongside their line numbers, and clicking a comment opens the referenced file, diff, or artifact with that comment focused.
• Refreshed the Customizations settings screen with collapsible sections and item counts for skills, rules, plugins, and custom agents, consistent back navigation across sub-pages, and clearer card layouts and status indicators for MCP server configurations.
• Clicking Retry on an agent error card now resumes the task directly, without adding an artificial "Continue" message to your conversation.
• Custom agents can now reuse the skills, rules, and subagents you already have, using a single inheritCustomizations setting in the agent's frontmatter.
• Added support for uploading and attaching WebP images in chat.
• Added syntax highlighting for PowerShell (.ps1, .psm1, .psd1, and pwsh/ps1 code fences) and Zsh (.zsh, zsh) in chat and the file viewer.
• Refreshed the Google Docs, Sheets, Slides, and Drive icons to the 2026 Workspace logos, and URL artifact cards for those products now show the product icon instead of a generic link glyph.

Fixes

• Fixed an issue where Find (Ctrl+F / Cmd+F) in the Review pane missed matches that were outside the visible area or hidden inside collapsed files and folded line blocks. Search now scans all open files and automatically expands and scrolls to each match.
• Fixed an error that could stop you from rewinding a long conversation.
• Fixed an issue where a conversation could get stuck showing as still running and never return to idle, most often after pressing Stop.
• Large binary attachments returned by connected MCP tools — images, PDFs, and audio — are now saved to a file and referenced in the conversation instead of being pasted inline, where they could previously overflow the conversation and break the turn.
• Fixed an issue where one invalid or unsupported MCP server configuration would prevent all other configured MCP servers from loading.
• Fixed an issue on Windows where the Proceed button failed to appear on artifact review cards and plan editor headers.
• Fixed an issue where titles for long-running conversations could unexpectedly change to an older message.
• Fixed an issue where pinning, archiving, or restoring a conversation that was no longer available on disk appeared to succeed. The action is now reverted and an explanatory notification is shown.
• Fixed an issue where newly created conversations in the sidebar temporarily displayed raw link text instead of clean @-mention names.
• Added visible error notifications when file attachments fail because of file size constraints, attachment counts, or unsupported formats.
• Fixed an issue where confirmation dialogs (such as reverting actions or deleting conversations) focused the close button by default, so pressing Enter dismissed the dialog instead of confirming it.
• Fixed an issue where updated images in chat could keep showing stale, cached versions; they now refresh on reload.
• Fixed an issue where comments added to markdown files or artifacts in preview mode attached to the first line instead of the selected text.
• Fixed an issue where images uploaded or pasted into the chat composer alongside a comment on an artifact were left out of the message sent to the model.
• Fixed an issue in the Review pane where dragging file header pills into the chat composer did not correctly reference files beyond the first.

To explore the full list of improvements and bug fixes, read the complete release notes on our Changelog, or visit the Download page to update to the latest version of Antigravity.

TLDR: Artifact outline viewer navigation, print-mode read-only slash commands, machine-readable output for models and agents, alert callout and carousel rendering in markdown artifacts, terminal hyperlink improvements, and fixes for headless --mode, startup diagnostics, subagent messaging, and usage reporting.

Improvements

• Added a heading outline to the artifact viewer, opened with t, so a long markdown document's structure is visible at a glance and you can jump straight to a section instead of scrolling.
• Added non-interactive answers for more read-only slash commands in print mode, so -p "/permissions", /hooks, /help, /changelog and /config each emit one tab-separated record per line — or a structured payload under --output-format json and stream-json — without starting an agent turn, spending usage or leaving a conversation behind, with /help listing exactly the commands print mode answers and /changelog printing the release notes.
• Added machine-readable output to the models and agents subcommands through an --output-format flag accepting json and stream-json, and moved their error messages and progress spinner off stdout so captured output contains only the list.
• Added a disable-slash-command: true flag for a skill's SKILL.md frontmatter, which hides that skill from the / menu and from /name resolution while leaving it discoverable and invocable by the model, so a large skill library no longer floods the command menu.
• Added rendering for alert callouts ([!NOTE], [!TIP], [!IMPORTANT], [!WARNING], [!CAUTION]) and for carousel blocks in markdown artifacts, which previously showed as raw markup.
• Improved terminal hyperlink support, enabling clickable links over SSH and in hyperlink-capable tmux, keeping word-wrapped URLs clickable past their first line, linking URLs that the markdown renderer had left as plain text including ones inside code blocks, no longer highlighting unrelated links together on hover, and fixing horizontal scrolling and text selection on lines that contain a link.
• Improved tool call headers so every native and MCP tool call carries a short summary of what it is doing instead of a bare name or a raw argument blob, with the same summary naming background tasks and subagents in the activity list and the matching action phrase describing the request in permission prompts.
• Improved headroom for large toolsets by raising the per-session count on tool declarations, so heavy MCP, plugin and skill setups stop being rejected for having too many tools.
• Improved sign-in with Application Default Credentials by resolving the usage project automatically, so service-account logins work without extra configuration.
• Improved repository detection so a Git repository nested inside a larger multi-repository checkout resolves to the intended root, with submodules and worktrees covered.
• Improved headless -p runs so the agent settles a choice itself where it would otherwise ask, instead of stalling on a question nobody is there to answer.
• Improved the schedule tool by dropping the redundant Timer Cancelled notification when a one-shot timer with an early-termination condition is cancelled, and by naming both the condition and the exact sender whose message satisfied it in the step result.
• Improved file paths in tool step lines by collapsing your home directory to ~ instead of truncating the front of the path, so the interesting end of a long path stays on screen.

Fixes

• Fixed --mode being ignored in headless -p runs, where a valid value such as accept-edits or plan was never applied and an unrecognized value produced no warning at all.
• Fixed startup diagnostics being swallowed into the log file instead of reaching the terminal, including crash notices, the --conversation not-found warning, the --print and --prompt-interactive conflict, conversation load errors, and the --mode and --agent warnings.
• Fixed a character disappearing from your submitted prompt where it wrapped at the terminal's right edge, caused by the echoed prompt being indented after it had already been wrapped to the full width.
• Fixed file citations in a response losing their line numbers, so a cited range renders as path:10-25 again instead of just the file name.
• Fixed a write-in answer in a multi-select question discarding the boxes you had already ticked, and fixed a stale write-in being submitted after you went back to a predefined option.
• Fixed large background tasks crowding out the prompt by selecting each active item below the input box to a single line.
• Fixed corruption of config.json by writing user config atomically, so a crash or a concurrent writer can no longer leave a truncated file that silently breaks settings persistence.
• Fixed the CLI giving up on a slow OS keyring after one second and falling back to empty storage, which forced a re-login; it now waits five seconds, as every other keyring operation already did.
• Fixed conversation preview titles derived from the first user message by cutting them at the first line and capping them at 500 characters, so a large pasted or scripted prompt no longer produces a multi-megabyte stored title.
• Fixed a crash on Windows when resolving the conversation transcript path, by honoring the path's drive letter in the trajectory log artifact converter.
• Fixed a subagent going silent on its parent when its own message failed to send, and fixed the parent being handed an empty notice when a subagent went idle without a final response.
• Fixed the read_url_content tool leaking a connection on every call, which could eventually exhaust the machine's available ports.
• Fixed the status line reporting model usage that was always one fetch out of date, so it now matches the usage you actually have left.

To explore the full list of improvements and bug fixes, read the complete release notes on our Changelog, or visit the Download page to update to the latest version of Antigravity.

At Sonoma Raceway, Google Developer Experts built an AI Race Coach using Google Antigravity to bridge software code and vehicle telemetry.

Tapping a Pixel 10 into 100+ car sensors, local Gemma 4 delivered real-time audio coaching right in the cockpit.

Google Antigravity served as the tool that bridged the knowledge gap—allowing software engineers to translate race car physics and expert driver coaching logic without being racing experts themselves.

It’s a powerful case study in how developers can build complex systems in industries outside their core domain.

https://developers.googleblog.com/bridging-the-domain-gap-ai-race-coach-built-with-antigravity-and-gemini/

We know the transition from the AI Ultra Access for Workspace add-on is disruptive, and we're sorry for the friction. Our goal is to keep you productive. Here is the path forward:

🛠️ Keep Building Today:

Link Antigravity 2.0 & CLI to a Google Cloud project to avoid interruption. As we work on the IDE solutions we recommend running the CLI inside your favorite editor's terminal.

👉 Details:

https://antigravity.google/docs/enterprise

🚀 Coming Soon:

Native Antigravity integrations for your existing IDEs Centralized admin controls inside Gemini Enterprise

Thank you for your partnership as we build a more integrated developer experience.