Changelog.

New

• GPT-6.1 SolCodex sessions can use OpenAI's new GPT-6.1 Sol, including its Max and Ultra reasoning levels, and the Usage tab shows what it costs. It appears in the model picker once your Codex CLI is version 0.159 or newer.
• Claude Sonnet 5.5 costsThe Usage tab and Teams now show costs for Claude Sonnet 5.5. Teams usage already sent for Sonnet 5.5 or GPT-6.1 Sol gets its cost filled in automatically after the update.
• ChatGPT Pro 500Accounts now recognize OpenAI's new Pro 500 plan and show its name.

Improved

• Reasoning levels follow each Codex modelThe reasoning slider now offers exactly the levels each Codex model supports, so newer models like GPT-6 Astra show Max and Ultra, and future models work without an update.

Fixed

• Codex remembers Max and UltraReopening a Codex chat that used Max or Ultra reasoning showed High instead. It now shows the level the chat actually used.
• Answer Claude's questions from your phoneWhen Claude asked you a question with choices in a terminal session, the phone only showed a spinner and you had to go back to your Mac to answer. The question now appears on your phone with its choices (or a box to write your own), and your answer goes straight to Claude.
• Your phone controls Claude terminals you started in agmuxFor a Claude terminal session started in agmux, the phone's permission card showed a long code instead of the session name and Allow or Deny only said "thread not found", Stop said no terminal was running, and a message could start a hidden second copy of the conversation instead of appearing in the terminal on your Mac. Approvals, messages, Stop and model changes from your phone now reach that terminal.
• Background work keeps running when you switch awayLeaving a Claude or Grok session while it was still working in the background (a helper agent, a longer job or a command) could stop the session and cancel that work partway through. A session you've moved away from now stays running until everything it started has finished, and only then closes to save memory.
• Change counts show for Claude sessions that ask another AIWhen a Claude terminal session ran Grok to get a second opinion, its sidebar row lost its lines-added/removed count. The count now stays, and a session already affected shows it again the next time you send it a message.
• Terminal text no longer goes missingAfter the app had been open for a long time, terminal text could turn into scattered letters with big gaps. Refreshing didn't help; only restarting agmux did. The terminals now repair themselves within a few seconds, and the refresh button fixes them too.
• "Use this account" no longer gets stuck on sessionsIt kept asking you to close sessions, even ones where you had quit Grok or Codex (or it had crashed), until you restarted agmux. Quitting now frees the account right away. If Grok sessions are still open on either account, it names them and offers to stop them and switch; they pick up on the new account when you open them again. For Codex, it names the sessions to close first.
• Account switching no longer leaves an account stuckWhen a terminal couldn't start, or couldn't start on the account agmux had just moved it to, agmux kept holding that account until you restarted. "Use this account" then refused it, and reopening the session could close it again. The account is now freed right away.
• The "finished" pop-up matches the new lookThe pop-up that tells you a session finished still had the old gold style. It now matches the rest of the flat look; the frosted Glass look keeps the original.

A calmer agmux, with Focus for the threads you're working on.

This update adds an optional Focus list, a cleaner look shared with agmux.dev and your phone, fuller account controls, local models that fit your Mac, and fixes across search, tasks, Teams, and remote control.

New

• FocusAn optional group at the top of the sidebar that gathers the threads you're actively working on from every project, so you don't need to keep every project open. It lists threads that are running, plus ones active in the last 10 minutes (you can pick up to 30 in Settings). It shows 7 threads before "Show more"; right-click Focus to show more or fewer. Starting a new session from Focus asks which project it belongs to. After updating you'll be asked once whether to turn it on, with a preview of how it looks. You can also turn it on in Settings → General.

Improved

• A cleaner lookThe app matches agmux.dev and the phone app: new type, flat panels and cards, clearer labels, and calmer status colors, plus a bolder app icon that fills its tile on the Mac and on your phone. Prefer the old frosted look? Choose Settings → Appearance → Surfaces → Glass. The previous font, Geist, is still in Settings → Typography.
• Teams matches the new lookThe Teams website uses the same type, flat panels and colors as the app.
• Redesigned phone remoteThe phone and browser remote has a cleaner new look that matches agmux.dev, with a light mode that follows your system. The session list is compact, with the time beside each title. Headers stay clear of the iPhone status bar and never crowd the session name. Replies are easier to read: headings, quotes and tables look right, code blocks have a Copy button, and tool activity is grouped into tidy cards. The message box stays on one line on small phones.
• Clearer account settingsIn Settings, Agent accounts is now just Accounts, and the old Accounts tab (Git identities and connected services) is now Git & Connections.
• Simpler AccountsSettings → Accounts now has a "Your accounts" section and a section for each of your teams, so it's clear which is which. Add a team account from the team's own section. Anyone on a team can see and use its accounts without signing in to them. When a usage check is rate limited, it now says so and keeps showing the last reading. Less-used actions sit behind each account's ⋯ menu, and "Add to switching" is gone because your current login is already used first when accounts switch automatically.
• Use this accountPick any Codex or Grok account and choose Use this account to sign your CLI into it, for agmux and your terminal alike. The login it replaces stays in your accounts. A team account stays checked out to you while your CLI uses it, so your team can see it's in use. Its usage stays up to date while your CLI uses it. If an account can't be switched to, the reason shows right on that account.
• See who's using a team accountTeam accounts show who is using them right now, and whether it's their CLI. If your CLI is signed into a team account, agmux recognizes it and lists it under that team, not as your own.
• Every usage limitAccounts show each limit separately, like the 5-hour and weekly limits, and team accounts show them too. A 30-day or monthly limit is labeled by its real length. Old readings show when they were taken instead of "Status unknown", and a Claude Team plan now reads "Team plan".
• See how busy each account isAccounts show how many people on your team are using them right now, and automatic switching picks the account fewer people are using. Team owners can turn on Shared Claude accounts to see how many people use each Claude login. A Claude account's email is only shown once two or more people are using it.
• Rename accountsRename an account from its ⋯ menu. Sharing a login with a team again no longer replaces the team's name for it.
• Move accounts to a teamTeam owners and managers can move a Codex or Grok account from their personal accounts to a team from its ⋯ menu. Moving your current login shares it with the team and lists it under the team, and you stay signed in.
• Subscription tiersAccounts now show the exact tier where the provider reports it, like Max 20x for Claude and SuperGrok Heavy for Grok.
• Newer local modelsSettings → Local Models now recommends Qwen 3.8 27B, the strongest coding model you can run on your Mac, for Macs with 24 GB of memory or more. Older models it outperforms were removed from the recommendations. Models you already downloaded keep working.
• Cleaner Local Models pageSettings → Local Models is reorganized into clear sections: your Mac and setup, recommended models (pick a memory size to see other options), everything you've downloaded, HuggingFace search, and agent tools. Removing a model now asks before deleting its files, and it works in light mode too.
• Claude terminals in light modeClaude terminals now use Claude's light colors when agmux is in light mode, and switch between light and dark as soon as you change modes. If you chose a colorblind-friendly or custom theme in Claude, it's kept.
• Account usageSettings → Accounts now checks usage on its own, so a newly added account shows its limits without clicking Check usage. Team accounts have a Check usage button too, and keep showing their last reading, with when it was checked, instead of going blank. When the account you're using is nearly out (5% or less left), agmux checks your other accounts so you can see which one to switch to.
• Phone remote works with every agentDroid, Kimi, Cline, Hermes, OpenCode and Gemini terminals now show their full conversation on your phone, not just what you typed. Saved Kimi, Pi and Grok sessions from your Mac's sidebar appear on the phone too, and every agent has its own icon.
• Approve terminal requests from your phoneWhen a Claude or Kimi terminal asks for permission, Allow and Deny now appear on your phone.
• Lighter on batteryagmux uses less CPU and graphics power, especially when its window is in the background or sitting idle. Spinners and status animations pause while you're in another app, and background checks run less often or wait until you come back.
• Commit message modelGPT-6 Luna Low replaces GPT-5.6 Luna for writing commit messages, both when you pick it in Settings and in Auto.
• Phone remote polishThe phone shows when it's loading or your Mac is offline, shows errors instead of failing silently, keeps the current model visible on small screens, no longer zooms in while you type on iPhone, and signs out right away if you revoke it on your Mac. Pairing no longer hangs, and pairing links from the iPhone app now work.

Fixed

• Claude terminal replies on your phoneTables, lists, headings and code blocks in Claude terminal replies now show properly in remote control instead of running together as one long paragraph.
• Searching for file namesSearching for something with a dot in it, like `main.rs` or `package.json`, now finds matching conversations instead of missing them.
• Renaming or deleting folders in the IDEFiles you had open inside a folder now follow it when you rename the folder (so saving works), and close when you delete the folder.
• Codex reset time in UsageThe Codex usage bars now show when your limit actually resets instead of saying "resets now".
• IDE editor stuck on "Failed to load file"After opening a file that can't be shown (like an image), opening another file now shows it instead of the same error.
• Search with accented letters, Japanese text or emojiMessages containing characters like ś, せ or 🐛 are no longer garbled in search results, the words around them can be found again, and searching for words like "śledzenie" or "せんせい" now finds your conversations.
• Search finds recent work summariesThe summary of what an agent did in its latest turn can now be found in search, instead of staying missing until you started another turn.
• Session times in the command paletteRecent and other sessions in the command palette now show how long ago they were used, instead of "now" (or a time off by several hours, depending on your time zone).
• Search results match what you typedSearch no longer sometimes shows results for an earlier, shorter version of your search.
• Committing files with spaces in their namesFiles whose names contain spaces or accented letters no longer appear twice (once in quotes) in the commit and Create PR lists, and committing them no longer fails. Renamed files show their correct new name.
• Faster task creationCreating a task no longer copies Claude Code's own worktree folders into the new task, which could take a long time and use a lot of disk space.
• Deleting a task whose folder is goneDelete Task now works when the task's folder was already removed outside agmux, instead of failing with a "No such file or directory" error.
• Branch changes in a new taskThe Git panel's branch and committed views, and the committed list in Commit, now show only the task's own work, not other people's changes from the main branch, when your local main branch is out of date.
• Task pull request descriptionsThe automatic title and description for a task's pull request now describe only the task's own changes, even when your local copy of the main branch is out of date.
• Reusing a deleted task's nameCreating a task with the same name as one you deleted now explains that the branch name is taken, instead of claiming a folder already exists.
• Archiving a Gemini chat in a taskClosing a Gemini chat agent's tab in a task now stops it, instead of leaving it working in the background.
• Chat approvals no longer vanishWhen two permission requests are waiting in a chat and the first has waited over a minute, it stays on screen instead of disappearing and leaving the agent stuck.
• "Always allow" works in Codex chatCommands you chose to always allow in a Codex chat now run on their own instead of failing.
• Codex reset time in the top barWith the status line in the top bar, hovering a Codex usage limit now shows when it resets.
• Answered from your phone, gone on your MacWhen you answer a Claude chat's permission request or question on your phone, the Mac chat now clears it instead of keeping it on screen and showing an error if you click it.
• Background agents finish in Claude chatA background agent in a Claude chat now shows as finished when it's done, with its tool count as it works, instead of looking like it's still running forever.
• One "finished" alert per OpenCode replyAn OpenCode chat sends one "Agent finished" notification per reply, including a reply you start from your phone.
• OpenCode chat asks about every actionWhen OpenCode wants to run several commands at once, each permission request waits its turn on screen. A request you already answered, or one left over after you stop, no longer blocks the next one.
• Grok chat shows connected-tool resultsTools from connected servers in a Grok chat now show their output, or the error when they fail, instead of an empty result. Folder listings and background command output show up too.
• Phone shows sent messages as sentA command you typed into a Claude terminal while it was working no longer stays marked Queued on your phone after it runs, including when you pulled it back into the message box to edit it first.
• Codex access requests from your phoneWhen a Codex chat asks for extra file or network access, the phone now shows what it wants, and tapping Allow actually grants it.
• Answering OpenCode questions from your phoneWhen an OpenCode chat asks you a question, submitting or dismissing it on your phone now works instead of showing an error.
• Stopping from your phoneTapping Stop on a terminal session from your phone no longer risks closing Claude or Codex.
• Phone remote history for Claude Desktop CoworkCowork sessions from Claude Desktop now show their full history on your phone.
• Pairing code in SettingsAfter a phone pairs, Settings → Remote Control no longer keeps showing a code that has already been used.
• Local model switch promptThe "switch your local model" popup can now be closed (Not now, the X, or Esc), and Open Settings actually takes you to Settings instead of leaving the popup stuck on top. It's also readable in light mode, and it lets you switch to a model you've already downloaded without downloading it again. Setup no longer offers the retired Qwen2.5 models.
• Debug Mode and Cleanup settingsThese two Settings pages no longer show bright white outlines, and now look like the rest of Settings.
• Teams session countsTeams, and Settings → Your data, now show how many sessions each person actually started. Before, one long session was counted again for every hour it ran, and automatic reviews and helper agents counted as sessions too. Their usage still counts toward tokens, cost and active time.
• Grok usage after resumingGrok work done after reopening an earlier session is no longer occasionally left out of Teams and the Usage panel.
• Teams date rangesThe 7, 14, 30 and 90 day views no longer count an extra day in their totals that the daily chart doesn't show, so totals, active days and the comparison with the previous period cover exactly the days you picked.
• Teams peak sessions"Peak simultaneous sessions" now counts agents you run at the same time in different projects or with different agents. Before, it only counted sessions in the same project, so running several at once often showed as 1.
• Teams tool counts for ClaudeWhen Claude ran several tools at once, Teams counted only the last one, so tool calls, edits, changed lines and the error rate were too low. Every tool call is now counted.
• Teams code changes for CodexFiles changed and lines added or removed by current Codex versions now show up in Teams. Before, Codex edits were reported as no changes at all.
• Teams token compositionReasoning tokens are now shown as part of output instead of being counted a second time, so the token composition bar and its percentages add up to the token total.
• Teams history is kept when uploads pauseWhen a team's trial ends and its dashboards become read-only, its members' past activity now stays visible instead of disappearing on their next sync.
• Home and Memory in split viewClicking Home or Memory in the sidebar now opens them while split view is on, instead of doing nothing. Your split sessions are still there when you pick one again.
• Setup steps no longer jump aroundMoving between setup steps keeps the window steady instead of resizing it each time, and every step now starts at the top.
• Recommended local models answer againModels downloaded from Settings → Local Models could sit on "working" and never reply in local chats and the local terminal. They now load and answer. On Macs with 8, 12, 16 or 24 GB of memory, the models recommended for your Mac also no longer get refused for lack of memory.
• Local models remember the whole conversationLong local chats no longer quietly forget their instructions and the start of the conversation. When a chat gets long, it is summarized to make room, the way cloud models do, instead of cutting off replies.
• Faster local repliesFollow-up messages in a local chat or terminal reuse the work already done on the conversation, so they start answering in about a second instead of rereading everything. Using one local model for chat and another for the terminal no longer makes each reload every few minutes.
• Local models fit your Macagmux now works out how much memory each local model really needs for a long conversation, and sets it up to fit your Mac. When there's room, it uses the faster setup that remembers text more accurately. A model too big for your Mac is turned away with a clear message, instead of loading and slowing the whole Mac down. Settings → Local Models shows the memory each model will really use on your Mac, and warns before you download one that's too big.
• Terminals stay in light or dark modeWith Color mode set to System on a light Mac, terminal sessions briefly turned dark each time they were opened or switched to, and a Codex terminal that started during that moment could keep dark colors. Terminals now open in the right mode. Switching Color mode to System also follows your Mac's appearance again.
• Downloading a local model no longer interrupts other chatsInstalling a new local model used to cut off any chat that was answering at the time, and could leave other chats unable to continue. Chats now finish what they're doing and carry on as normal.
• New task keyboard shortcutPressing ⌘Enter in New Task no longer creates the task when the branch name is flagged as already in use or invalid, matching the disabled Create button.
• Commit & Create PR with deleted filesCommitting a task's changes before opening a pull request no longer fails with "did not match any files" when an agent had deleted a file.
• Grok terminals in folders with spacesFor projects whose folder path contains a space, reopening a Grok terminal now continues the earlier conversation instead of starting a new one, and past Grok sessions from those folders show up in the sidebar.
• OpenCode terminal context meterWhile an OpenCode terminal was working on a reply, its context meter could jump to a huge number (the whole session's token total) and then drop back. It now keeps showing the latest real reading.
• Cline terminals stop spinning after an errorWhen a Cline reply failed (for example a provider or network error), the session kept showing as working. It now settles when the reply ends.
• Crash with non-English text in terminalsagmux could quit suddenly when a Pi or OpenCode terminal sent it a prompt or file edit containing characters like accents, emoji, dashes or non-Latin text. That no longer happens.
• No leftover requests after a Grok or Gemini restartChanging effort or model in a Grok or Gemini chat now clears permission requests the old session can no longer answer, so new ones show right away.
• Hiding, pinning and deleting sessions on long-used installsWhen agmux's saved sidebar data was full, choosing Hide or Pin on a session did nothing, deleting a thread could leave it in the list, and moving threads to another project reported a failure even though they moved. These now work straight away.
• Claude terminal sessions missing from the sidebarIf you use Claude Code plugins or startup hooks, many Claude sessions you started outside agmux never appeared in the sidebar. They now show up with their first message as the title.
• Droid listed agmux twiceOn Macs that used agmux before it was renamed, Droid ran agmux's status updates twice after every message and listed them twice. It now runs them once.
• Terminal session namesUsing the arrow keys while typing in a Codex terminal no longer puts stray characters like "[D" into the session's name, and sending a prompt you recalled with the up arrow still shows that terminal as working. Text you clear with Ctrl+C in a Claude terminal no longer ends up in the name.
• Claude terminal stays "working" after you declineAfter using /clear or /resume in a Claude terminal, declining a permission request left the session showing as working. It now goes idle.
• Claude terminals on HomeClaude terminals you started in agmux now show their name on Home (including names you gave them), and opening one from Home returns to that terminal instead of starting a second copy of the session.
• Kimi sessions sort by when you used themKimi sessions in the sidebar and on Home showed no time and sank to the bottom of the list. They now show when they were last active and sort with everything else.
• Sign out before you join a teamOn the Teams website, Sign out is on the page you see when you are not on a team yet.

Agent accounts, the newest models, and agmux is now open source.

Keep working when an agent runs out of allowance by switching between your Claude, Codex and Grok accounts, pick up GPT-6, Claude Opus 5.5 and Grok 4.7, get help from inside the app, and read the full source on GitHub.

New

• GPT-6 Sol and LunaChoose the new models in Codex and OpenCode chats, tasks and the phone remote, with usage costs shown at their published prices.
• Claude Opus 5.5Select Opus 5.5 in Claude chat and see usage costs at its published price. The model menu refreshes when opened so newly installed Claude models appear without restarting agmux.
• Grok 4.7 in xAI chatNew Grok chats use Grok 4.7, xAI’s latest model, by default. Extra High reasoning is available. Grok 4.6 and 4.5 stay in the list, and a chat you already set to one of those stays on that model.
• Agent accountsSee your existing Codex and Grok login alongside added accounts in Settings → Agent accounts, with Codex plan labels and account limits on Home and in Usage. Your current login stays primary; when its allowance runs out, agmux checks Codex model compatibility before switching to another personal or team account.
• Personal Claude accountsYour existing Claude login appears automatically in Agent accounts. Add personal accounts through Claude’s browser sign-in, view their limits on Home and in Usage, and switch accounts when allowance runs out and the conversation is safe to resume. Claude accounts cannot be shared with a team.
• Direct supportSend bugs, crashes, questions and feedback from Settings, with screenshots or other attachments, an optional reply email and a confirmation number. Support is also available when the app cannot open your data.
• Data recoveryBefore an update changes saved app records, agmux keeps a backup. If those records cannot open, you can restart, contact Support or restore a saved copy while keeping the failed files.
• agmux is open sourceThe full source code is now public on GitHub under the MIT license.

Improved

• Refresh change countsRight-click a sidebar thread and choose Recalculate diff to refresh its saved change counts without reopening the conversation. Empty results stay visible, and missing measurements show as unavailable or partial instead of silently disappearing.
• Claude change counts include script editsWhen Claude edits files by running a command instead of its edit tool, those added and removed lines now appear in the sidebar counts.
• Faster first setupChoose an agent, review permissions and open your first project. Appearance and local models are optional, with installation and sign-in help available along the way.
• Task mode refreshAll supported agents and local models are available, with the same session controls as Agent mode. Phone remote control now shows tasks and their branches, and uses the selected task’s workspace for model choices.
• Simpler sidebarRemoved the Issues tab from sidebar navigation.
• Cloud command suggestions need your own keyTerminal command suggestions and Ask with OpenRouter now use your own Groq or OpenRouter key. Without one, suggestions come from a local model or stay off.

Fixed

• Older chats reopen againChats started before the previous update open in chat view again instead of showing “This session is not known to have been created in agmux.” Teams usage still counts only sessions started in agmux.
• New Claude terminals no longer disappearOn long-used installs, a newly created Claude terminal could vanish from the sidebar before you typed into it, and titles for new sessions were not kept after a restart. agmux now keeps its saved title history compact so there is always room, and a new terminal stays in the list even if saving fails. Terminals that already disappeared come back once you have sent them a message.
• Project memory in new projectsAgents started in a brand-new project get shared project memory from their very first session, not only from the second one.
• Home usage cardProvider usage rows, including Gemini, stay fully visible instead of getting cut off at the bottom.
• Codex conversationsTool results remain visible when commands run in batches or use unfamiliar formats. Expanded commands retain their full text with Show more for long output, missing or unfinished results no longer look successful, repeated patch reports count once, and more bulk file edits contribute to change counts.
• Search stabilityUnusual formatting in a conversation no longer crashes the app while preparing search results.
• Codex change countsOlder conversations show counts from their saved edits automatically, and parent chats include their subagents’ edits. More scripted edits and file copies between projects count correctly, including commands that finish after a reply. Deleted and similarly named files no longer count twice, formatting-only edits count correctly, and unavailable history is clearly marked.

New

• Team restrictionsOwners and managers can limit agents, models, reasoning effort, and chat or terminal access in an organized settings panel. Everything is allowed by default; agmux enforces explicitly configured rules and explains blocked choices.
• Debug ModeRecord recent performance in Settings and let your coding agent inspect CPU spikes and slowdowns. Captures stay on your Mac and remain available after restarting.
• App cleanupSettings → Cleanup lets you review and remove old generated thread names and disposable cached files older than 90 days, while keeping conversations, manual names, project memory, active threads, and Teams data.

Improved

• Major performance improvementsWith many sessions open, agmux stays more responsive: hidden sessions pause extra refreshes, repeated status checks are combined, and Teams sync uses less memory on large histories. Approval and completion tracking stay active.
• Subagent conversationsOpenCode now has the same subagent cards and conversation panel as Codex. Claude and Cursor child replies and tool details are more complete; Gemini and local chats show available launch details and explain when child history is unavailable.
• Commit message modelGPT-5.6 Luna replaces Codex Spark in Settings and automatic commit message generation.

Fixed

• Teams accountingCorrected duplicated Codex child usage, recovered newer response reports, and stopped signing out from giving the same desktop a second accounting identity.
• Teams coverageUnverified older history stays local, missing coverage is explained, and session activity is clearly distinguished from unique conversations.
• Codex change countsMore edits across a list of files, in linked worktrees, and from applied patch files contribute to terminal change counts. Old commands no longer leave counts blocked after a Mac restart, and new terminal sessions track when their commands can safely stop blocking later edits.
• Terminal scrollingReturning to a terminal keeps you at the latest output if you were at the bottom when you switched away.
• Steadier session titles“Continue,” approvals, and command-only follow-ups keep the task title intact. Queued and steering messages in terminal and chat retain context when updating titles. Resummarize follows the same rules, including for older saved conversations.

New

• Reconnect CodexAfter signing in again, right-click a Codex terminal in the sidebar and choose Reconnect Codex to continue the same conversation with your refreshed login.

Improved

• Many open sessionsCodex terminals in background tabs no longer process every line of output while hidden, so switching between several busy sessions stays responsive. Permission prompts in hidden Codex terminals are still noticed.

Fixed

• Sidebar status colorsCompletion pulses stay green and permission or question pulses stay amber, regardless of your accent color.
• Blank or frozen terminalsA terminal that had once been open as a split-view tab could stop updating, or show nothing, after split view was turned off. Terminals now follow what is actually on screen.
• Closing the windowClosing the main window now hides it instead of asking to quit. Click agmux in the Dock to bring it back with all sessions intact, and Quit still asks for confirmation.

Improved

• Smoother session switchingReturning to a running terminal catches up on new output without unnecessarily rebuilding its history. Unchanged background change counts no longer refresh every open session.

Fixed

• Commit button colorsCommit and push buttons, including the thread toolbar’s change badge, now match your chosen accent instead of mixing it with yellow backgrounds.
• Light mode readabilityTerminal text and input boxes stay readable, and Home project and thread names, composer selectors, settings toggles, forms, menus, editor panels, and status colors follow the selected appearance more consistently.
• Codex change countsBackground commands no longer leave later change counts stuck, including after restarting the app. Sidebar counts recognize more edits in chats and terminals (including Python edits followed by parallel checks), tool discovery no longer hides changes, and restored edits are counted once even when Codex reports them different ways.
• Codex statusChats stop showing “starting MCP” once Codex begins responding or using tools, even if another server is still starting.
• Computer-use approvalsCodex app-access prompts recover if a chat misses them, and helper-agent prompts appear in the parent chat.
• Stopping CodexStopped chats recover from a stuck “Stopping…” indicator when Codex has already ended the turn.
• Split-view tab labelsSession tabs show the correct Terminal or Chat label and model across providers, including Codex terminals, discovered Claude sessions, OpenCode chats, and local models.
• Closing the app windowClosing the main window now keeps it visible for the quit confirmation, instead of leaving agmux running with no window.
• Terminal scrollbackSwitching away and back preserves earlier output and your reading position when all new output is still available.

Gemini chat, openable subagents, and a steadier phone remote.

Chat with Gemini next to Claude and Grok, open a helper agent’s conversation beside yours, and get a long list of phone, Codex, Teams, and Cursor fixes so everyday use feels more reliable.

New

• Phone session timelineBrowse prompts and outcome summaries, then jump to a turn in web and iOS conversations.
• Subagent conversationsOpen a subagent’s conversation in a smoothly sliding panel beside your chat in Claude, Codex, Cursor and Grok, with the subagent’s own messages, readable tool names, task details and progress states styled like your main chat. Stacked Tasks and Subagents cards float below the thread toolbar and show only this conversation’s agents, with readable names and status even when launch messages are out of view. Running agents stay visible even after the parent replies; completed agents appear newest first, with an always-visible arrow to open their conversation. Codex subagent rows align with neighboring tool rows. Resumed agents show their current status, and large conversations refresh with less background work.
• Anonymous usage statsagmux can send a once-a-day anonymous ping (app version and Mac version only, plus which agent you start) so we can see how many people use the app. It’s on by default. Turn it off anytime under Settings → General → Privacy.
• Gemini chatStart a Gemini conversation from the draft composer (same style as Grok/Claude chat). Your selected model and thinking level apply when the chat starts or restarts. You can switch Plan or Chat, and choose whether it asks before running commands. The first message stays on screen and is sent once the chat is ready, instead of vanishing. Thinking text sits tight under the Thought row instead of leaving a blank gap. The New menu Gemini tile still opens the terminal. First chat asks you to sign in with Google, and signing in after a failed start retries the chat automatically. Apple Silicon only. Same chats work from the phone.

Improved

• Attach any fileUse the chat’s + button to select PDFs, documents, and other files. Images appear as previews when available; other files are added as local paths.
• Lighter background Codex sessionsKeeping several Codex conversations open does less background work while messages, questions and completion status stay up to date. OpenCode background conversations preserve the order of replies, tool activity and errors.
• Compact tool groupsClaude, Cursor and similar chats show grouped and nested tools as compact rows like Codex. Open a grouped tool’s details directly in the conversation.
• Reliable Teams trackingConversations retain their creation details across restarts. New Grok, Pi and Cline conversations keep their tracking even if their first update is missed. Opening or switching to an outside Claude conversation does not add it to team usage. Leaderboard ties keep ranked members ahead of unranked members.
• Codex questionsAnswer choices or type a reply directly above the chat composer, with separate answers for each question. Answering while Codex works keeps its ongoing work visible instead of prematurely collapsing it into a completed summary. Internal polling no longer clutters the conversation.
• Faster Codex startupPreviously loaded Codex threads appear immediately when you reopen agmux, while the list refreshes in the background.
• Quick Open providersChoose Pi, Cline, Gemini, Hermes, or Local terminal for the compose button and ⌘N, including Gemini chat.
• Codex terminal refreshRefresh the terminal layout from the thread’s top bar, just like in Claude and Grok.
• Cleaner chatsConversation scrollbars are hidden across chat views, including subagent conversations. Scrolling still works as usual.
• Phone permission modeChanging Supervised / Auto / Full access on the phone now applies to every chat (Claude, Codex, Grok, Gemini, OpenCode, Cursor), including after you’ve already sent a message, and is remembered when you reopen that chat or start a new one. Messages from other agents preserve your existing chat permissions.
• Gemini change countsGemini terminal and chat threads now show added and removed line counts in the sidebar and completion notifications.
• Model pickerThe agent list remembers which providers you used recently and shows about four at a time, so the one you want is usually at the top.
• Local terminalThe New menu’s local option now opens a Pi terminal already pointed at a model on your Mac, instead of Grok.
• Cursor chat diffsCursor chats now show green + and red − line counts in the sidebar and on each edit in the thread. Expanding an edit opens a numbered diff instead of a dump of the tool result. Restored counts exclude failed and unfinished edits and are not added again when chats reopen.
• Gemini sidebar namesGemini chats in the sidebar now show the model name only, without High / Medium / Low.
• Gemini usage limitsHome and the Usage tab now show your Gemini weekly limit (and the 5-hour limit when Google reports one), using the same Antigravity sign-in as Gemini chats.
• Gemini iconGemini chats, the sidebar, and the New menu now use Google’s colorful 2025 sparkle instead of a white star.

Fixed

• Codex completion statusFinished chats now clear the header’s running badge, including conversations that used subagents.
• Remember Codex effortNew chats keep your last effort selection, including Medium, even if you changed it before sending a message.
• Cursor context meterRemoved misleading context-full warnings based on total tokens processed. Cursor chats hide the meter when current context usage is unavailable.
• Codex permission alertsComputer-use permission prompts now show an amber pulse and notification, including in background terminals.
• Codex Stop and follow-upStop now shows “Stopping…” until Codex finishes cancelling. Your next message waits in the queue, avoiding “turn is already active” errors and conflicting activity indicators. Failed cancellations show an error and keep Stop available to retry.
• Agent token totalsIncludes Claude helper-agent work, corrects duplicated usage in copied conversations and cached input, and reads Gemini’s reported tokens. Recorded Teams usage stays available if an agent removes its old logs.
• Clearer cost estimatesCorrected model prices and cache charges. Teams labels and explains partial estimates in plain text when an agent leaves out needed usage details or a price is unavailable.
• Fairer leaderboardsThinking tokens are no longer counted twice, and incomplete cost estimates no longer rank as free usage.
• Remote conversationsMessage inputs stay fully readable when opening a session, and internal Claude task notifications no longer appear as queued messages.
• Teams provider breakdownAgents you use occasionally stay visible by name instead of disappearing into “Other”. Claude conversations created in agmux also recover their missing history, while outside conversations remain excluded.
• Unexpected quits on macOS 26/27A glitch in the system's hover popover could close agmux without warning. The app now shrugs it off and keeps running.
• Teams provider trackingCodex chats and other agents now contribute to team activity. Past retained history syncs automatically, and large updates preserve earlier usage totals even when the Mac’s clock is ahead.
• Codex terminal questionsQuestions now show a notification and an amber sidebar pulse until you reply.
• Session timelinesCodex turns now describe the outcome, saved chats can recover missing timelines, and jumping handles repeated prompts and partial histories more accurately.
• Teams agent policiesTeam owners can choose from every app provider when setting allowed agents.
• Codex completion and promptsFinished background terminals show the sidebar pulse, and repeated chat updates no longer duplicate your prompt.
• Codex terminal spinnerAdding a newline with Option+Enter no longer starts the sidebar spinner. It waits until Codex actually starts the submitted prompt.
• Sidebar change countsVerified Python and shell edits contribute to totals, including fast Codex edits and commands that keep running tests. Codex chat and terminal counts update while other conversations are open, and recorded patch changes restore from saved history without counting the same edit twice. Ambiguous overlapping edits are skipped to avoid crediting the wrong agent, including long-running commands and commands whose file targets cannot be determined.
• Codex terminal timelineSaved turns now appear in the session timeline, with jumps to their prompts in terminal scrollback.
• Codex chat scrollingChat stays with new messages and tool updates, including when you switch back from another app. Jump to latest moves straight to the bottom instead of sometimes stopping partway. Scrolling up or opening tool details keeps your reading position; sending a new prompt takes you back to the bottom.
• Unexpected crashesFixed crashes caused by damaged saved chat-title data or Claude settings, malformed agent questions or Gemini conversation data, and non-English startup diagnostics. OpenCode now reports a missing executable instead of abruptly closing its connection.
• Codex chat stuck on WorkingAfter Codex launches helper agents, the chat no longer stays on Working (and “May be unresponsive”) after the reply is actually finished. Missed messages now appear from the session file without quitting the app, even if extra command rows were already on screen. A follow-up you queued while it was working is sent when that turn finishes, instead of sitting in the queue.
• Codex hidden commandsWhen Codex runs several commands or lookups inside one code step, those now show as the usual command and tool rows (collapsed in a group when there are a few), so you can see what it actually did. Repeated calls keep their separate results. Finished commands stay Ran when you reopen a chat, instead of spinning as Running. A command that keeps running in the background (like a local preview server) is labeled Serving or Running with the actual command, instead of a blank Ran row.
• Phone message deliveryQueued messages wait for the previous turn, including when reopening a terminal. Drafts and photos stay with the right conversation when you switch chats, and a new chat keeps its first message if you navigate away. Failed or uncertain sends remain available to review and retry. Creating chats from two phones no longer mixes up their first messages. New phone chats also work with Macs still running 4.0.2.
• Phone questions and approvalsChoose an answer or write your own when an agent asks a question. Pending prompts return after reconnecting, prompts already answered on the Mac clear, answering one session no longer dismisses a different session's prompt, and tapping outside the sheet no longer drops the request.
• Remote settings tabOpening Settings → Remote no longer briefly disconnects your phone.
• Phone chat controlsClaude and Codex model choices now come from the Mac. Claude and Grok effort changes reach the running chat, and Codex fast mode is applied when sending.
• Phone tool outputOpenCode replies and tool results appear on the phone, Cursor edits show the actual diff, and Codex results stay attached to their tools. Failed edits show the error instead of changes that never happened. Expanded tools stay open as new output arrives, and diff lines resembling file headers remain visible.
• Large phone historiesLong conversations and large session lists load without getting stuck. Exceptionally large individual messages clearly indicate when the full content needs to be viewed on your Mac.
• More phone terminalsExisting Cline, Hermes, and Droid terminals appear in remote control. Pi conversations include saved messages, thinking, and tools. Codex terminals keep their own controls when another Codex chat is open.
• iOS remote updatesThe native app now builds from the current remote interface instead of an older website checkout.
• Phone Allow on a finished terminal promptTapping Allow or Deny on the phone after you already answered on the Mac (or after the prompt is gone) no longer types Y or N into whatever is running in that terminal. If the terminal is unavailable, the request stays open for retry.
• Revoking a phone disconnects it right awayTurning off a paired phone, or disabling remote, now drops that phone even if it was idle in the background.
• Cursor model nameCursor chats now show Fable 5.1 instead of “Fable 5 1”.
• Cursor last modelStarting a new Cursor chat remembers the last model you picked, instead of jumping back to Composer.
• Teams Knowledge disclosureThe “Before you share” notice on the Knowledge tab now shows as a real panel with an accept button, instead of a block of raw HTML.
• Gemini chat sign-inThe first Gemini chat no longer sits on “Starting session…” while Google sign-in opens in a browser. The chat now says to finish signing in there.
• Gemini context ringGemini chats now show how much of the context window a turn used, instead of staying at 0%.
• Gemini stopThe Stop button in a Gemini chat actually cancels the reply instead of only clearing the spinner.
• Gemini tool rowsFile search and read tools in Gemini chats show as Search / Read instead of “Tool Running find_file”. File reads clear the spinner once the file is opened, instead of spinning for the rest of the reply.
• Gemini model pickerIn a Gemini chat, the model menu only lists Gemini models (not Claude, Codex, or Cursor).
• Gemini permission promptsWhen Gemini asks you to allow a command or file change, the sidebar now pulses amber and the approval toast appears, even if you’re looking at a different chat.
• Renaming a sidebar chat switches to itRenaming a chat in the sidebar (for example a Grok terminal) no longer jumps you into that chat while you’re typing.
• Grok terminal clicks miss buttonsClicking Cancel, Send now, or pop-out on a subagent in a Grok terminal should register instead of doing nothing, including after switching away and back.
• New Grok chat from the phone shows no messagesStarting a Grok chat on your phone no longer leaves the Mac on “Starting session…” with an empty thread. Your first message shows up right away.
• Phone photos in Grok chatPictures you send from the phone now show up as pictures Grok can see, instead of a file it tries to read as text.
• Grok queued follow-upsMessages you queue while Grok is working now send when it finishes, instead of sometimes vanishing.
• Grok chats reopenReopening a Grok chat more reliably shows the conversation and continues it, instead of a blank or forgetful thread.
• Phone-started Grok chat on the MacOpening a Grok chat you started on your phone no longer leaves the Mac stuck on “Starting session…” while Grok is already working.
• Claude terminal starts faster after switching chatsSwitching away from a Claude terminal and back no longer restarts Claude from scratch every time.
• Chat toolbarThe bar above a chat is less busy: Commit and your editor are icons.
• Chat Local and branchGemini and Cursor chats keep the Local and branch pickers under the composer, like Codex.
• Gemini and Cursor “Starting session…”Opening a Gemini or Cursor chat no longer stays stuck on “Starting session…” after the chat is actually ready.
• Phone file editsEdited files on remote now show green + and red − counts on the row. Tapping the file opens a numbered diff (additions in green, removals in red) instead of a gold-tinted dump.
• Grok “finished” toast while still workingWhen a helper agent inside a Grok terminal finishes, you no longer get a “finished” toast if the main Grok agent is still going.
• Codex Stop vanished while still workingWhen Codex launches helper agents, the Stop button and spinner stay until Codex itself is done, not when one helper finishes.
• Phone Codex listHelper agents Codex launches no longer show up as extra chats with the same name as the original conversation.
• Phone Claude history after compactWhen Claude shortens a long terminal chat, the phone no longer shows the hidden “session continued” dump as a message.
• Codex chat after StopStopping a Codex chat and sending another message no longer fails with a “thread not found” error.
• Edit and copy on a promptHovering your own message in Gemini, Claude, Grok, or Cursor chat no longer stacks the edit and copy buttons on top of each other.
• Hide and info on a tool rowHovering a helper-agent or tool row in chat no longer stacks Hide, the info button, and the expand arrow on top of each other. Result / Show on neighboring rows stay lined up.
• Chat text lines up with the message boxIn Cursor, Claude, Gemini, and Grok chats, replies and tool lines now sit on the same left edge as Codex — inside the message box, not hanging past it.
• Cursor context usageCursor chats no longer show more of the context window used than the model actually holds (for example 100% with a used count bigger than the window). The ring now matches the real conversation size.

New

• More terminal agents in NewThe New menu’s terminal chips now sit on two rows of five: Claude, Codex, Pi, OpenCode, and Grok; then Local, Kimi, Cline, Gemini, and Hermes. Each opens in this project’s folder, keeps working/done indicators, names the thread from your first prompt, and can resume the same conversation when you reopen it. The sidebar shows the model and +/− line counts when files change, like the other terminals. Existing Droid chats still open. If macOS blocks one of these CLIs from starting, the error now says so instead of a blank message. Gemini now opens Google’s Antigravity terminal, because Google sign-in on the old Gemini terminal no longer works for personal accounts.

Improved

• Claude and Codex model lists stay currentThe model picker now reads the models your Claude and Codex apps already know about, so new ones show up without waiting for an agmux update.
• New menuThe extra “Plain shell” row is gone. Start a terminal with the Terminal row or an agent chip.
• Home lists more recent projectsOn a tall window, Home shows more of your recent projects instead of leaving a blank gap under usage.
• Cooler when idle or with several chats runningThe app uses less of the computer’s processor and graphics when you’re in another app, when a terminal is sitting still, or when extra chats are working in the background while you look at a different one.
• Time spent on models and providersUsage and Teams now show how long agents were actually working on each model and provider, next to tokens. Idle time is left out, same as the active-hours chart.
• Usage and Teams token countsCodex cache hits and Claude replies that stream in pieces are counted the same way other usage tools do, so the totals should look right. Estimated cost follows each provider’s list prices, including long-context and 1-hour cache, and duplicate sub-agent / forked-session copies are not counted twice.
• Teams only counts agmux sessionsTeam usage no longer includes Claude, Codex, or Grok work you did in those apps or in a regular terminal. Only sessions you started in agmux are counted. A sync after this update drops the extra history without clearing the in-app numbers.
• Phone remote tool rowsOn remote, file edits stay visible with the actual diff and line numbers (and +/− counts), instead of collapsing into a group or showing “file updated.” Expanding a group of reads or commands still lets you open each tool’s output.

Fixed

• Grok looks scrambled when switching chatsSwitching from one Grok terminal to another no longer leaves the screen garbled so you have to hit Refresh.
• Cline working indicator and chat namesCline chats now show the working spinner while a reply is in progress, and the sidebar names the thread from what you typed.
• App quitting while a window is loadingOn the latest macOS, the app could close itself while a chat or the window was loading. That should no longer happen.
• Grok terminal white stripGrok’s conversation no longer shows a bright white bar along the right edge. Other terminals were not affected.

Improved

• Local models stay lighter in long chatsOn smaller Macs, a long local chat now keeps its working memory within the size Settings already showed, so it is less likely to fill the machine as the conversation grows.

Fixed

• Grok project memory on a new MacOpening a Grok terminal now connects project memory right away, instead of skipping it until you trusted the folder. Grok should use the memory tools instead of rewriting the memory files by hand.

Cowork, on-device models, and a look that matches the site.

4.0 is a big step: Cowork for everyday work chats (including ones you started in Claude and ChatGPT), AI models that run only on your Mac, Cursor sign-in, Team Knowledge, a beta program, and gold-and-black styling. Remote, Teams, and many session bugs are also cleaned up.

New

• Cowork modeA briefcase button next to Task opens Cowork, a place for everyday work chats instead of coding terminals, with a short “Opening Cowork” screen so the switch never looks frozen. Cowork keeps its own list of folders: it starts empty, you add the ones you care about, and your Claude Cowork and ChatGPT Work chats for those folders — including chats started in the Claude and ChatGPT desktop apps — appear there with their earlier history. Refreshing picks up new desktop chats, New always starts a chat in a folder you added, and hiding or deleting a chat keeps it gone.
• Cowork folders and file previewsRename a folder in Cowork by double-clicking it (or right-click → Rename) and drag folders into the order you want; nothing moves on your Mac. Clicking a file name inside a Cowork chat opens just that file — markdown in the reading preview — with the file tree closed for more room, and the file-explorer button still opens the tree when you want it.
• ChatGPT Work and Grok Cowork chatsChatGPT Work is a Codex chat tuned for everyday work, with the same connected tools and plugins Codex already uses on your Mac. Grok Cowork is a regular Grok chat with the same everyday-work focus.
• Beta programApply at agmux.dev/beta. If you’re approved, you can download upcoming builds others don’t get, and paste a tester token under Settings → About → Updates so the app can install later betas for you.
• Grok 4.6 in xAI chatNew Grok chats use Grok 4.6, xAI’s latest model, by default, and Extra High reasoning is available. Grok 4.5 stays in the list, and a chat you set to 4.5 stays on 4.5.
• Chat with AI models that run on your own MacStart a new chat and pick a local model from the model list. Nothing you type, and nothing the model reads, leaves your computer, and it keeps working with no internet connection. New local chats respect your Local or Worktree choice and accept photos on the first message. The New menu also has a local option alongside Claude, Codex, and the rest, which opens the familiar terminal already pointed at one of your installed models.
• Local Models page in SettingsBrowse models suited to your Mac, see how much space and memory each one needs, and download or remove them with one click. The one-time setup local models need sits at the top of the page with an Install button, so you can get ready before downloading anything. Recommended Speed, Balanced, and Quality picks are listed for every Mac size from 8 GB to 256 GB — switch the memory tier to see the picks for any machine, not only your own — and models that cannot read or change files are left out, so anything you can pick can genuinely help you code. You can paste an Exa key so local models can search the web, and turn native tool calling on or off.
• Local models manage their own memoryagmux loads a model when you start using it, unloads it once it goes idle, and shuts everything down when you quit, so several models can be available without filling up your Mac’s memory. To free memory sooner, use the Eject button in the chat composer, or right-click a local chat or terminal in the sidebar and choose Eject model.
• Sign in to Cursor from SettingsOn Accounts, one click opens Cursor sign-in in your browser. After that, Cursor chat can use every model on your plan (including Ultra) without pasting an API key.
• Team Knowledge (Teams plan)Paid teams (or an active trial) can keep shared decisions and short session summaries on teams.agmux.dev; free teams get analytics only. Owners turn it on after accepting the privacy note, and sharing anything — or letting agents read it — always requires accepting that note first. Members can add decisions and promote session notes, and the Memory tab can share a decision or summary straight to the team. When the owner allows it, agents can look up official team decisions, never full conversations. Owners can export Knowledge from the web.
• PR Leaderboard (Teams plan)Weekly cost and GitHub pull-request ranking is part of the Teams plan (or trial), not Free. Owners connect a GitHub App and pick repositories in Settings.
• Help tab on TeamsEvery team on teams.agmux.dev now has a Help tab covering setup, what is collected, roles, each page, budgets, leaderboard, knowledge, billing, and common problems, in plain language.

Improved

• Yellow and black look to match the websiteThe default Midnight theme uses the same gold accent and dark surfaces as agmux.dev. The other Appearance themes (Forest, Indigo, Golden, and the rest) keep their own colors but now look cleaner in both dark and light mode — softer tints, readable text, and accents that stay clear on light backgrounds.
• Home, Usage, and Issues look cleanerThose screens use the same quieter cards and headers as the rest of the app.
• Cleaner terminal panelThe slide-up terminal drops the yellow-tinted wash across its top bar. Tabs, the folder path, and the buttons now sit on one calm strip that matches the terminal below, with gold kept for the small dot that shows a shell is alive.
• Cleaner chat scrollingThe thin scrollbar along the right edge of Cursor and Claude chat is gone; you can still scroll with the trackpad or mouse wheel.
• Cursor chat controls and model namesCursor chats show Chat/Plan, model, and Supervised / Auto / Full tool modes, and list friendly model names like Composer 2.5 and Sonnet 4.6 Thinking instead of raw codes. Thinking is Off or On (or Low / High) instead of true/false, and only appears when the model supports it. Tool rows for shell, edit, and search render like other chats, and replies no longer leave stray status labels (like “FINISHED”) under the answer.
• Faster, clearer app updatesUpdate checks start as soon as you open the app, the update prompt sits larger in the bottom-right corner, and it shows again every time you open the app if an update is still waiting. Settings → About has an Automatic updates switch to download and install on open without a click.
• Richer first-run setupOnboarding now covers project memory, whether new chats start with full permissions, phone remote, finish and approval alerts, keep-awake while agents run, automatic updates, and Claude vs Codex chat-or-terminal defaults — not only theme and fonts. Existing installs that already finished setup get a short “new options” pass for the new choices.
• Clearer, safer project memoryMemory shows who wrote each entry and a compact health summary. Editing a fact can mark older entries as replaced, so agents stop treating stale notes as truth.
• Memory rules are set by agents, not a review queueA rule agents must follow is marked by the agent that verified it, rather than by you clicking through a queue, and important stars stay sparse attention markers. The Memory screen shows both counts without a confirm step and warns when either piles up. Clean memories clears important flags and archives replaced entries and resolved issues; binding rules stay, and nothing is permanently deleted.
• More reliable shared memoryWhen several chats save memory at once, their changes no longer overwrite each other. Memory saves recover after an interruption, and agmux warns you if its local readable copy needs repair.
• Photos on RemoteOn your phone, the + button next to the message box attaches images (only after this app update — older Macs hide the button so photos don’t vanish). Chat sessions receive them like on the Mac; terminal sessions get a file path on your computer so the agent can open the image. The button stays put when the phone reconnects while your Mac is still online, and the temporary “sending” bubble clears once the photo lands in the conversation.
• More agents on remotePhone remote can see and control OpenCode, Cursor, Kimi, and local-model sessions as well as Claude, Codex, and Grok. You can start a new Cursor chat from the phone (pick any model on your Cursor plan, send messages, and stop a turn) without opening it on the Mac first, and changing OpenCode or Cursor models from the phone works.
• Phone remote at remote.agmux.devThe phone remote site is now at remote.agmux.dev. Older links to agmux.dev/remote still work, so you don’t need to re-pair if you already use remote.
• Connect this Mac from the remote websiteOn a Mac browser at remote.agmux.dev, tap Connect this Mac to open the app and pair that browser automatically, with no code typing. For safety, if Remote control is switched off, the app takes you to Settings to turn it on yourself — a web page can never quietly switch on phone access to your Mac.
• Phone alert when an agent is waiting on youIf you leave the remote phone app in the background, it can show a system notification when an agent needs approval or an answer, so you can jump back in without watching the screen.
• Team agent policyTeam owners can set which coding agents members may start and a default permission mode (Teams → Settings → Agent policy). Linked Macs pick that up on sync.
• Teams: how long agents wait on peopleTeam dashboards can show how many approvals happened and how long agents spent blocked waiting for a human (counts only — never what you typed).
• Codex usage shows weekly correctlyWhen Codex only has a weekly limit (the 5-hour one is off), the home and usage cards show it as Weekly instead of mislabeling it as 5-hour.
• Kimi model name and context meterKimi terminal sessions now show the active model (for example K2.7 Coding) and how full the context window is in the sidebar and top bar, same as the other agents.
• Grok chat first message looks cleanThe long project-memory instructions Grok needs on the first turn no longer fill the yellow user bubble; you only see what you typed.
• Task mode catches upYou can start Grok or OpenCode as a terminal agent on a task (not only as chat), status badges only show “running” when an agent is actually working, and empty Task mode has a clear New Task button (⌘N).
• Quieter background sessionsGrok sessions you leave open release their heavy helpers when you switch away, and unused Codex project connections stop hanging around, so the app uses less memory with many projects open.
• Window size rememberedIf you resize or move the app window, that size and place come back the next time you open agmux.
• App data folder renamed to match agmuxYour projects, settings, and history move automatically from the old private folder name into `.agmux` on first open after update. Nothing to do by hand.
• Built-in browser removedThe Browser button in the chat top bar and the side browser panel are gone. Web links still open in your normal system browser.
• No “Rec” badges on model listsThe model picker no longer marks some Claude, OpenCode, or Grok models as recommended; every model is listed the same way.
• Settings cleanupRemoved the Agentic terminal control; Issues dispatch instructions live on their own Issues page; chat summaries are local-only, and older Qwen2.5 models must be replaced with Qwen3 or Phi-4. Claude Auto mode is labeled as available on every plan, and Settings pages use tighter padding.
• Friendlier tips and settings wordingHome-screen tips and several Settings descriptions drop engineering jargon so the product reads more clearly for everyday use.

Fixed

• Light mode is readable again on Teams, updates, and the file editorSettings → Teams headings, the update prompt, editor tabs, the file filter, and ask/approval banners no longer go white-on-white when Appearance is Light. The update prompt also stays put when you click back into the app.
• “Manual download needed” no longer pops up over and overClosing that update notice keeps it closed for the rest of the session, and a failed background update check no longer claims you need a full reinstall from the website.
• Automatic updates recover from a failed installIf an auto-update fails, you see an error and can retry instead of a stuck “installing…” state.
• What’s New no longer covers WelcomeOn a first install, and when existing installs see the short “new options” pass, the What’s New sheet and notification prompt wait until you finish setup.
• Finish and approval alerts honor setupTurning off “Notify when an agent finishes” or “Notify when approval is needed” in first-run setup (or Settings → Notifications) actually stops those Mac notifications.
• App no longer quits when dragging a file onto a terminalDragging a file from Finder into a Claude (or other) terminal session no longer closes the whole app mid-drag. You can drop the file to paste its path into the prompt as before.
• New Codex terminal sessions open cleanlyStarting a fresh Codex terminal no longer fails with a “failed to resume session” error, so it loads as usual and you can type right away.
• Chats no longer appear twice in the sidebarStarting a Grok chat, or a Claude or Codex chat from remote.agmux.dev, shows one entry on the Mac instead of the chat plus a matching terminal session with a messy model name.
• Home no longer lists the same Codex chat twiceCodex chats show once on the home Recents list, matching the sidebar.
• Idle terminals no longer sit in ActiveThe top-strip Active group only includes chats that are actually working, not every open terminal.
• Cursor chats behave like every other chatCursor sessions show the Cursor app icon instead of a plain letter “C”, spin the same working indicator in the sidebar while answering (and clear it when the turn ends), and show how full the context window is in the top bar after a turn.
• Cursor remembers Local vs WorktreeThe Local / Worktree control under a new Cursor chat no longer snaps back to Worktree every time you open the composer. Pick Local once and new Cursor drafts stay on Local until you change it.
• OpenCode model and context in the sidebar and top barOpenCode terminal sessions now show which model is running under the session name and a context-window meter in the top bar, same as Claude, Kimi, and Grok, instead of only “Terminal · now” with an empty row.
• Grok chats stay on Grok modelsThe model menu in a Grok chat lists Grok models instead of Claude and Codex, and a new Grok draft no longer starts on a Cursor or Claude model.
• Grok chats keep their reasoning after reopenPrior thinking is shown again when you reopen a Grok chat, not only the reply text.
• Grok terminal no longer double-notifies when doneWhen a Grok terminal session finishes in the background, you get one completion notice instead of two stacked ones.
• Grok helpers stay out of the sidebarReviewers, writers, and other helpers Grok launches in the background, and one-off commands, no longer appear as extra chats in the project list on desktop or phone.
• Jumping to an older prompt in GrokPicking a past turn from the Timeline works with short prompts and can be repeated in the same session, and if a turn really is gone the terminal returns you to where you were instead of stopping part-way up.
• Task mode keeps working agents and your last taskOpening Task mode no longer marks a live Grok or OpenCode terminal as failed, and your last selected task comes back even if you have more than one project.
• “Worktree” from a project’s New menuChoosing Worktree switches to Task mode and opens a new task for that project (its own branch for parallel work), instead of a dialog that often left the chat hard to find or never set up the branch at all.
• Hidden task folders rediscovered with custom settingsTasks you made with a custom folder location or “branch first” layout show up again when you reopen Task mode.
• New Task keeps the error on screenIf the task folder is created but the agent fails to start, the dialog stays open with the error so you can fix it instead of closing silently.
• Extra Codex agents on multi-folder tasksStarting another Codex agent on a task that spans several folders no longer fails to open.
• Commit works when design docs are part of the changeCreating a commit no longer fails with a red “git add failed / paths are ignored” error when a tracked file under an ignored folder (like `docs`) is included. Those files stage and commit as usual.
• Phone send shows the Mac as workingStarting a Cursor or Claude chat from your phone keeps the desktop sidebar spinner going until the turn actually finishes.
• Phone Cursor chats follow Supervised / Auto / FullNew Cursor chats from your phone use the permission you pick, not unrestricted tools.
• Codex chats started from your phone stay cleanStarting a Codex chat from remote no longer dumps project instructions into the conversation.
• Remote session status matches the MacA small spinner shows while an agent is working, an amber pulse means it needs your attention, and a green pulse means it finished and you have not looked yet. Green clears as soon as you open the session, including when you are already sitting in the chat.
• Chat titles match between phone and MacChats started from your phone get a normal short title on the Mac sidebar and on remote after the first message, instead of staying “New Grok Chat” or showing a temporary file path when a photo was attached. Grok chats are named while they are still working, sessions that were only named on the Mac now show that name on the phone, and renaming a session updates the title at the top of the open chat right away.
• Remote model picker on web and phoneChoosing a different provider (Claude, Codex, Grok, OpenCode, Cursor) in the new-chat model menu opens that provider’s models instead of closing the menu. OpenCode and Cursor load the same full catalog from your Mac as the desktop app (with search), Grok no longer lists Composer, the OpenCode logo appears correctly, and a fresh chat defaults to Claude Opus 5.
• New chat from a project on RemoteEach project in the remote session list has a + next to the count; tap it to open a draft chat already set to that project. The folder picker lists projects in the same order as the Mac sidebar, not by when each was created.
• Tables in replies on your phoneTables in agent replies render as real tables on the phone (with bold and code inside cells), instead of one long line of pipes.
• Remote terminal no longer shows blank user lines for control keysClearing the prompt or interrupting a run does not leave empty yellow bubbles in the phone timeline.
• Phone remote matches the app’s colorsPhone remote uses the same gold accent and black backdrop as the Mac app and website.
• Remote control after site updatesIf the remote service is briefly updated while your phone is connected, the phone and Mac reconnect on their own and keep your pairing.
• Teams upload keeps going after a Codex approvalApproving a Codex tool no longer blocks that Mac’s Teams upload.
• Linking a Mac asks you to confirmLinking a Mac to a team from the browser now asks for confirmation first, so a link cannot be triggered silently.
• Project memory tools on other people’s MacsShared project memory (the tools agents use to remember decisions) starts correctly when you install agmux from the app package, including Codex and Grok terminals — not only on a developer machine that still has the source folder.
• Safer file toolsDeleting or renaming files cannot reach outside your project folder, image attach only reads allowed paths, and @-mention file browsing stays inside the project too.

Improved

• Teams leaderboard drill-downClick a person on the PR leaderboard to open their metrics for that same week (model mix, token composition, cache hit) so you can see why TOK/pt or $/pt is high.
• Teams site on phonesOn a narrow screen the top nav wraps instead of clipping, wide tables scroll sideways, and the PR leaderboard shows as cards instead of a cut-off table.

Fixed

• Grok timeline jumpClicking an older prompt in the Session timeline scrolls the Grok terminal back to that turn (Grok’s full-screen UI has no terminal scrollback, so jump now searches and scrolls inside Grok itself).
• App freezes during long chatsThe window should stay responsive while agents run many tool steps; session timeline no longer overloads the app when updating the turn badge.
• Blank frozen window after long sessionsIf the built-in browser view stops responding, the app reloads it automatically so you do not have to quit and reopen.
• Sluggish app after very large chat history cleanupOn launch, agmux reclaims wasted space left behind after old oversized logs were removed so the app stays snappy.

More accurate costs, Kimi Code, smarter titles, and a more reliable phone remote.

This release tightens Usage and Teams spend numbers, adds Kimi as a terminal provider, refreshes chat titles as you keep talking, and fixes remote loading hangs on your phone.

New

• Kimi CodeStart a Kimi terminal chat from the provider picker (replaces the older Droid option). Existing chats still open normally.
• Teams PR leaderboardOn teams.agmux.dev, owners can turn on a leaderboard that ranks people by merged pull requests and usage after linking a GitHub organization.
• On-device model for titlesChat titles and short summaries use a small model that runs on your Mac (one-time download, no API key). First launch asks you to install it; you can upgrade later to newer options like Qwen3.

Improved

• Up-to-date cost estimates for new modelsWhen a model isn’t in agmux’s built-in price list, estimated cost uses live public rates (OpenRouter) instead of guessing. If no price is known, cost shows as zero rather than a made-up number.
• Cursor chat model listStarting a Cursor chat shows every model available on your Cursor plan (not only Composer), so you can pick the same models Cursor itself offers for your account.
• Teams member profile photosTeam people lists show each person’s GitHub or Google profile picture instead of only colored initials.
• Smarter chat titles as you keep talkingAfter each new message in a long chat, the sidebar title updates from the whole conversation (with more weight on what you just asked). Short “go ahead” / “do it” replies no longer rename the chat. Right-click a chat and choose Resummarize anytime to refresh the title.
• Teams leaderboard tableMember names line up cleanly (no more right-aligned handles), GitHub-style logins show as readable names, and you can click any column header to sort by merged PRs, tokens, tokens per PR, $/pt, and more.
• Teams leaderboard for NenuWhen a team links the NenuAI GitHub organization, ranks use complexity points from each pull request’s Project Size (XS through XXL) instead of only small/medium/large by lines changed.
• Teams groups and manager viewsOn teams.agmux.dev, owners can create groups and limit what each manager sees to specific groups or people.
• Select text in terminalsHold Shift while dragging, or use Shift+arrow keys, to select and copy text in agent terminals—even when the agent is using full-screen or mouse mode.
• Terminal links open in your browserWeb links in agent terminals open in your system browser (not the in-app side panel).

Fixed

• Session timeline jumpClicking a turn in the Session timeline scrolls the chat or terminal back to that prompt (with a brief highlight in chat). Works for Claude, Grok, Codex, Kimi, and OpenCode sessions.
• Usage and Teams cost estimatesEstimated spend for Codex, Grok, and Claude models is much more accurate. Codex no longer double-counts cached prompt tokens (which inflated costs by several times), Grok now uses real token and cache numbers from sessions instead of rough guesses, and model rates match current public pricing (including Fable, Sonnet 5 intro pricing, and the July GPT-5.6 Terra/Luna cuts). The same numbers feed the Usage tab, Settings → Your Data, and Teams.
• Composer removed from xAI chatNew Grok chats no longer offer Composer as a model under xAI. Composer stays available under Cursor chat.
• Grok slash-skill chat titlesStarting a Grok terminal chat with only a slash skill (like `/checkagentsdk` or `/commit`) now gets a real sidebar title instead of staying “New Grok Thread” or showing the raw command.
• Search for hyphenated wordsSearching for phrases like “multi-prompt titles” in global search (⌘⇧F) finds matching chats again. Hyphenated terms were treated as “exclude this word,” so results came back empty.
• Sessions stuck on “Loading… / Fetching history from your Mac” on your phoneOpening a session from your phone now shows the conversation right away instead of hanging on the loading screen. Your Mac was re-sending the whole session list and every open conversation every couple of seconds even when nothing had changed, which clogged the connection; it now sends updates only when something actually changes, and it stays quiet entirely when no phone is connected. This also cuts the data agmux uploads in the background to a small fraction of what it was.
• Remote Control stuck on “Online” after the connection diedIf the link to your phone silently broke, your Mac kept reporting Online while nothing actually reached the phone until you restarted the app. It now notices within about a minute and reconnects on its own.
• What's New logoAfter an update, the release notes popup shows the agmux app icon instead of a generic X mark.
• Multi-Agent option removedStarting a new chat no longer lists Multi-Agent in the provider picker. Use Claude, Codex, Grok, or another provider as usual.
• App freeze with blank windowsLong Grok/Claude terminal sessions no longer fill the local database with huge terminal scrapes, which could make the app hang and leave windows blank. Old oversized scrapes are cleaned up on launch.
• Thread name summariesAuto-generated chat titles work again with the on-device model (they were failing with “No title returned”).

Richer Teams dashboards and a more reliable Grok terminal.

Managers get budgets, tool breakdowns, and exports; usage charts fill from real agent activity; Grok terminal refresh and scroll behave as expected.

New

• Teams budgets and exportsManagers can set a monthly spend budget with a month-end projection, download usage as a spreadsheet, and review a simple log of team changes (joins, budget edits, exports).
• Tool activity on TeamsSee work broken down by kind (terminal, edits, reads, search, web, helpers) plus how many files changed and lines added or removed — counts only, never your code or paths.

Improved

• Teams charts show real activityUsage now comes from your actual Claude, Codex, and Grok sessions so dashboards fill in as you work instead of staying empty.
• Friendlier Teams labelsProviders and models show as readable names (Claude Code, Claude Opus 4.6, Grok Code Fast) instead of raw codes.
• Grok cost estimatesUsage and Teams cost figures for Grok use real rates instead of showing $0.

Fixed

• Grok terminal refreshThe refresh control next to the branch name redraws the Grok terminal again (including in split view).
• Smoother Grok trackpad scrollScrolling a Grok conversation moves line by line instead of jumping by whole pages.
• Teams Sync nowSyncing from Settings always rechecks and tells you whether something uploaded or you were already up to date.

Multi-Agent chats, Teams usage, and a built-in browser.

Start Multi-Agent from a new chat, share hourly team usage under Settings → Teams, and open pages beside your work. Grok terminals scroll with the wheel and stay correct after /clear.

New

• Multi-Agent chatsWhen starting a new chat, pick Multi-Agent and choose a main model (Claude, Codex, Grok, and others). You talk to that main agent like a normal chat; it can bring other agents in behind the scenes to help.
• agmux TeamsUnder Settings → Teams, sign in and join a team. See simple team usage in the app and on teams.agmux.dev. Only hourly summaries are shared — never your prompts, code, or file paths — and you review what is shared before joining.
• Built-in browserOpen a browser from the chat top bar to view pages beside your work (back, forward, reload, or open in your system browser).

Improved

• Multi-Agent model pickerMain models show the right logos, and effort and permission controls match a normal chat for that model.
• Grok terminal scrollingTrackpad and mouse wheel scroll the Grok conversation instead of moving the cursor in the prompt.

Fixed

• Grok after /clearStarting a fresh Grok session with `/clear` no longer gets stuck on the old session when you return to the tab.

Control agents from your phone — plus timeline, orchestrator, and shared project memory.

Run Claude, Codex, and Grok from your phone while the Mac keeps working. On the desktop: jump past turns with Timeline, watch every active agent from Orchestrator, share memory across chats, and dispatch GitHub issues to a new agent.

New

• Phone remote controlTurn on Remote control under Settings to manage Claude, Codex, and Grok chats and terminals from your phone at agmux.dev/remote. Pair with a short code (QR link), see your sessions, read the conversation, send messages, stop a run, and approve or deny tool requests. While remote is on, the Mac can stay awake with the lid closed so agents keep going when you’re away. Turn remote off anytime to disconnect phones (even if the Mac was offline when you turned it off).
• Session timelineOn any agent chat or terminal, open Timeline in the top bar to see past turns in this session (what you asked, a short summary of what the agent did, done/running status, and how long ago). Click a turn to jump back to that point in the conversation.
• Issues tabBrowse open GitHub issues for your project (uses your existing GitHub login from the terminal). Pick an issue and Dispatch to agent to start a new chat on it, with the issue title and description filled in. Add extra repos to track if you like.
• Orchestrator tabSee every active agent as a tile: who’s working, who needs your OK, last action, and approve/deny without hunting through the sidebar. The full agent list stays in the sidebar while you’re there. Message or launch agents from the composer at the bottom, including pasting or dropping images into the prompt (same as normal chat).
• Horizontal agent tabsIn Settings → Appearance, switch Agent tabs to Horizontal for a browser-style top bar: project pills, sessions in a strip (active ones first), compact + / compose to start new agents, and full-width chat. Vertical (sidebar) stays the default. Includes Running (live sessions across projects) and Your Threads (opened sessions stay until you close them with × — like multi-view tabs for the top bar).
• Smarter chat searchSearch finds text inside past messages and notes (not just chat titles), with short previews of the match.
• Claude Cowork chatWhen starting a new Claude chat, toggle Code / Cowork on the composer. Cowork is for everyday work (docs, files, research) instead of coding, still using your Claude subscription.
• Shared project memoryEvery chat and terminal in a project shares one memory (Claude, Grok, Codex, and others — including terminal mode). Agents can save decisions and facts so the next model or terminal session still knows them. Browse everything in the sidebar Memory tab (replaces the old Agents tab): durable memory, session history, and archived items now have separate views, with clear labels showing whether an entry was recorded by you, an agent, or an automatic summary. You can edit, archive, restore, resolve, and reopen entries there. Agents can search past memory and sessions, open a summary, then pull a short slice of the chat log if they need more — without pasting whole histories. If an agent forgets to write a session note, a local model can fill a short handoff (only then — never overwrites what the agent wrote). New sessions can get a tiny recent session list (titles only; optional under Settings → Behavior). Turn project memory off anytime under Settings → Behavior → Project memory (on by default).
• Update project path & move threadsRight-click a project to Update project path… when the folder moved on disk (the sidebar name follows the folder, and chats that still used the old path come along). Or Move all threads… to another project if you reorganized folders.

Improved

• Richer first-time setupSetup walks through visual themes, fonts, vertical vs horizontal session tabs, what ⌘N / compose opens, which AI drafts commit messages, notifications, chat vs terminal defaults, and an optional local AI model download. Re-run anytime from Settings → About → Run setup. After updates that add new setup choices, a short refresh may appear once.
• Tasks on the right of chatTo-do lists and plans (Claude, Cowork, Codex, Grok, OpenCode, and local models) show as a floating checklist on the right of the chat, with progress and a collapse chip — no longer a bar above the message box. Chat text and tool lines stay lined up with the left edge of the message box.
• View commit after you pushWhen a commit finishes, use View commit (bottom left of the success dialog) to open that commit on GitHub.
• Click a notification to open that chatWhen an agent finishes or needs your OK while you’re in another app, clicking the macOS notification brings you straight to that conversation. You can also click a row in the in-app notification list to jump there.
• Codex says when tools are still startingIf Codex is still connecting its tools at the start of a turn, the status line shows starting MCP (with the tool name) instead of a silent “thinking.”
• Finish toast clears when you open the chatOpening a finished agent’s chat (from the sidebar, a tab, or split view) dismisses its “Finished” toast so it doesn’t linger.

Fixed

• Grok helper workers stay out of the sidebarWhen Grok Terminal spins up short helper workers for a task, those no longer show up as extra chats in the project list (and the list rows stay lined up).
• Grok token history in UsagePast Grok chats (including terminal ones) now show up in the token charts and totals, not only the SuperGrok credit bar.
• Selected items readable in light modeThe highlighted Settings tab and the selected chat in the sidebar no longer use pale text on a pale background.
• Smarter Grok approval pingsWhile Grok’s Auto mode is still deciding (classifying), the sidebar no longer flashes amber early. You only get the amber “needs approval” pulse and Mac notification when the Yes/No menu is actually on screen — not for silent auto-approves, and not during the wait. After you approve or deny, the amber clears right away (not only when the command finishes).
• Codex terminal model labelWhen you switch models inside a Codex Terminal session, the sidebar and top bar update to the new model instead of staying on the old name.
• Codex terminal context meterThe top bar shows how full the context window is for Codex Terminal sessions (same as Codex chat), and it updates as the session continues.

A major refresh — new models, glass UI, and agents that keep working with the lid closed.

agmux 3.0 brings GPT-5.6 and Grok 4.5, a unified glass chat experience for every agent, SuperGrok usage on Home, and the option to keep agents running when your MacBook is shut.

New

• GPT-5.6 modelsCodex and OpenCode now include Sol (default), Terra, and Luna. Sol and Terra also offer Max and Ultra reasoning levels.
• Grok 4.5New default Grok model. Existing threads keep their old model names.
• Codex Auto ReviewA new permission option that lets Codex auto-review approval requests for you (alongside Default and Full Permissions).
• Grok usage on Home & UsageSee SuperGrok credit use (how much is left and when it resets), plus Grok session history in the same charts as Claude and Codex.
• Show only running threadsRight-click a project in the sidebar to hide idle history and keep only threads that are working, need your attention, or are open.
• Keep running with the lid closedUnder Settings → General, allow agents to keep going with the MacBook lid shut (works on power and battery). Uses Touch ID once to set up; normal sleep returns when the agent finishes or you quit the app.

Improved

• Fresh glass look throughoutChat, Home, Usage, Settings, the sidebar, and the slide-out terminal share the same soft emerald frosted-glass style: translucent panels, quiet dividers, and a calmer overall surface.
• Cleaner chat for every agentClaude, Grok, OpenCode, and Codex chats now share the same layout: plain agent replies (not speech bubbles), short one-line tool rows you can expand for details, and finished turns that fold into a simple “Thought for …” summary. Click to expand the full step-by-step again.
• Smarter message composerModel, reasoning effort, permissions, and send sit on one row. Effort is a Faster ↔ Smarter slider (same control everywhere), the context meter is a clear ring, and menus match the glass look.
• Smoother chatReplies stream more naturally, expansions animate instead of snapping, and tables/code blocks match the glass style (with a copy button on code).
• Smarter commit messagesThe commit dialog drafts a message automatically (and you can pick which model does it under Settings → General).
• Clearer agent-done soundThe default finish chime is a soft double-hit instead of a low hum.
• Lighter on memory when terminals sit unusedIdle background terminals free memory after a couple of minutes and come back when you open them again. Grok also releases heavy helper tools it started, so long sessions use less RAM.
• Simpler SettingsRemoved tabs that didn’t do anything. Models is now Summaries; terminal history length lives under General. Quick Open can start any agent type, not just a few.
• Keep-awake while waiting on youThe Mac stays awake not only while the agent is working, but also while it’s waiting for your approval.
• Tidier project menuRemoved the unused “Edit conventions” item from the project right-click menu.

Fixed

• Settings and sidebar look right in light modeText, cards, and +/− line counts stay readable on a bright theme.
• Codex model list is cleanerRetired models are gone; GPT-5.3 Codex no longer appears twice.
• Context meter stays accurate for GrokUpdates while the agent is working, and no longer sticks after you compact a long chat.
• Codex sub-agents read clearlyLaunching helpers shows one stable “Launched … Agent” line (not a clutter of internal steps), live and when you reopen the chat.
• Codex Terminal starts cleanlyNo more frozen “Fitting terminal…”, sudden window crash, “session not found,” or an extra Chat thread appearing in the sidebar.
• Codex working spinner clears on timeStops soon after the agent finishes, and clears right away if you press Escape. Stray red “/” badge next to the title is gone.
• Grok (and similar) terminals are more reliableNo endless “Fitting terminal…,” no garbled layout when you return to a resting session, correct +/− counts in the sidebar and completion toast, and the working spinner clears when you press Escape.
• Grok asks for attention properlyWhen Grok asks you a question, the sidebar pulses amber and you get a Mac notification.
• Terminal refresh button works againClicking refresh next to the branch name resizes the terminal instead of dragging the window.
• Reopening a terminal doesn’t reset its time to “now”Last activity stays honest when you open an existing Grok, Droid, or OpenCode terminal.
• Home shows friendly model namese.g. Grok 4.5 instead of a raw code like `grok-4.5`.
• In-app updates after the renameAuto-update works again after the Xanom → agmux name change. If an update still can’t install, a banner points you to agmux.dev.

Now called agmux — leaner on your CPU and friendlier to drag-and-drop.

This release renames Xanom to agmux and lands a round of performance work: running many terminal sessions at once no longer heats up your machine, off-screen terminals stop repainting, plus fixes for file drops and terminal offloading.

New

• New name: agmuxXanom is now agmux, with a refreshed logo and app icons across macOS, Windows, iOS, and Android.

Improved

• Much lower CPU with many terminals openRunning several Claude terminal sessions at once no longer heats up the machine. Idle terminal sessions no longer wake the CPU on a timer, background sessions stop polling git and decoding output they aren't showing, and only the visible session refreshes its branch/status — so CPU stays flat as you open more sessions instead of climbing with each one.
• Home screen no longer keeps the GPU busyA background Codex or shell terminal that was still streaming would keep repainting off-screen while you were on the Home screen (or any other tab), holding CPU/GPU usage up. Off-screen terminals now pause rendering whenever they aren't the visible view.

Fixed

• Dropping a non-image file now pastes its pathDragging a PDF, document, or any non-image file from Finder into a chat composer or terminal now inserts its full path (quoted when it contains spaces); previously only images were handled and other files did nothing. Images continue to attach as before.
• Terminals no longer unload while an agent is waiting on youA backgrounded Claude session with a pending permission prompt, or a shell terminal running a full-screen interactive agent, now stays loaded instead of being offloaded after inactivity, so the question isn't torn down out from under you.

Buttery-smooth chat, instant launch.

This release is a focused performance pass — chat streaming, the sidebar, and app startup are all noticeably snappier — plus a couple of Codex protocol fixes and a fresh app icon.

New

• Claude Sonnet 5Added to the model picker with proper display-name formatting.

Improved

• Smoother chat streamingMarkdown rendering is now memoized and streaming updates are batched, cutting redraw overhead during long responses across Claude, Codex, and OpenCode sessions.
• Snappier sidebar and project listZustand store subscriptions are scoped more precisely and background polling pauses while the window is hidden, so the sidebar re-renders far less during heavy activity.
• Faster app launchThe settings dialog and CodeMirror language modes now load on demand instead of bundling upfront, shrinking the startup bundle.
• Codex sidebar loads instantlyThe thread list no longer waits on reading every session's model from disk; model labels backfill in the background as they resolve, and app startup no longer blocks on this fetch.
• Refreshed app iconNew icon artwork across macOS, Windows, iOS, and Android.

Fixed

• Codex thread rename, compaction, and rate-limit status work againUpdated to match a Codex app-server protocol rename that had silently broken these actions.
• Model labels in the thread header show proper namesClaude and Codex model slugs are now prettified (e.g. "claude-sonnet-5" → "Sonnet 5") instead of raw slugs.
• Streaming updates no longer stallFixed a timing edge case where coalesced streaming deltas could be delayed past their flush deadline.

Cursor joins the lineup — a new agent, a new flagship Claude model, and clearer multi-agent views.

Xanom 2.1.3 adds Cursor as a fully supported coding agent, introduces Anthropic's new Claude Fable 5 model, and makes collaborating agents and Codex sessions easier to follow and more reliable.

New

• Cursor agent supportRun Cursor as a first-class coding agent in both chat and task modes, with the same worktree isolation as your other agents.
• Claude Fable 5Pick Anthropic's most capable model, with a 1M-token context window, right from the model dropdown.
• Subagent threadsAgents spawned during a run now show up as their own threads, so you can follow exactly what each collaborating agent is doing.

Improved

• Clearer multi-agent viewCollaborating agents are visualized more clearly, making it easier to see who's working on what.
• Smoother Codex sessionsCodex handles concurrent work and conversation history more reliably.

Fixed

• Cleaner task shutdownTask agents now stop properly when a task ends or is archived, so stray background processes no longer linger.

Claude Opus 4.8, a faster Codex, and a smoother long-running app.

This release adds Anthropic's new Opus 4.8 flagship model, brings the Priority service tier and smarter project defaults to Codex, and clears up resource buildup that could slow the app down over multi-day sessions.

New

• Claude Opus 4.8 (1M context)Anthropic's latest flagship is now selectable in the model picker, with a full 1M-token context window.

Improved

• XHigh & Max reasoning on Opus 4.8The highest extended-thinking effort levels now work with Opus 4.8 (and still work on 4.7).
• Faster Codex responsesCodex now runs on the Priority service tier for quicker turnarounds.
• Codex inherits your project defaultsNew Codex chats automatically pick up the workspace's configured model and reasoning effort, while still honoring any model or effort you set manually.
• Codex remembers fast modeYour fast-mode preference now persists across new chats.
• Cleaner input fieldsLighter borders and improved focus states across the chat inputs.

Fixed

• Long inline code wrapsLong monospace snippets no longer overflow or stretch the message column.
• No more process buildup over long sessionsStopped agents are now fully cleaned up, so the app no longer accumulates leftover "zombie" processes that dragged down system performance during multi-day sessions.
• Reliable Claude session detectionFixed a race when switching threads quickly that could leave orphaned file watchers running in the background.

Stability and CLI-compatibility fixes — Xanom now survives long-running sessions and works with the latest Codex and Claude CLIs.

A maintenance release focused on subprocess hygiene, macOS input fidelity, updated agent CLI flags, and a smoother chat-composer experience for Claude.

New

• Grok per-thread chat sessionsEach Grok SDK thread now runs its own `grok agent stdio` process, so per-thread model, reasoning effort, and permission settings actually stick. Filesystem reads outside the workspace are gated by an approval prompt, and chat history survives an app restart.
• AskUserQuestion pickerWhen Claude asks a multiple-choice question in an SDK chat, you now get a real interactive option picker instead of a text input.
• Complex-script rendering tipIf a Claude PTY chat is rendering Tamil, Devanagari, Arabic, or another complex script and the glyphs misalign, Xanom now shows a one-time amber tip suggesting you switch that thread to SDK chat mode (which shapes correctly).

Improved

• Picking Opus from the chat composer always opens a real chatThe new-chat composer no longer silently drops you into the \`claude\` CLI in an xterm; Claude submissions from the composer always create an SDK chat thread.
• Approval toast and commit dialog polishSmaller pre-release fixes to the approval banner, commit-message flow, and Claude SDK session view.

Fixed

• Holding Shift no longer scrambles capital letters on macOSXanom now disables the macOS press-and-hold accent popup for its own bundle, so holding Shift to type capitals in xterm-backed Claude / Codex sessions no longer duplicates or reorders keystrokes. (The fix activates on the *next* launch after install.)
• Codex spawns work again on the latest CLIThe removed \`--full-auto\` flag is replaced with the equivalent \`--sandbox workspace-write --ask-for-approval never\`, so Codex sessions no longer exit before the TUI renders. Warp shell-out (\`codex exec\`) is updated to match.
• Zombie subprocess pile-up on long uptimeAfter ~2 days of uptime, timed-out CLI calls (thread-name summarization, Cmd+K Ask AI, commit-message generation, task spawns) could accumulate 600+ zombie processes and drive load average past 300. Subprocesses are now reliably killed and reaped on timeout.
• DMG sidecar packagingThe release DMG now bundles the sidecar Node dependencies correctly.
• Claude Agent SDK bumped to v0.3.145 and OpenCode SDK bumped to v1.15.6 — picks up upstream fixes from both providers.

Meet Grok — Xanom now manages a fifth coding agent.

v2.1.0 brings Grok in as a first-class agent alongside Claude, Codex, Droid, and OpenCode, plus a home screen that loads instantly, a commit dialog that remembers where you left off, and a round of stability fixes for running multiple agent sessions at once.

New

• Grok agent supportStart a thread with Grok and chat just like you do with Claude, including session discovery, resume-after-restart, model selection, and live context-usage tracking.
• Installer setup instructionsThe macOS installer window now shows the exact one-line command to clear Apple's quarantine flag, so the app launches cleanly the first time.

Improved

• Commit dialog remembers your progressClose the commit dialog mid-message or mid-push, do something else, and reopen it to find your subject, body, and progress exactly where you left them.
• Instant home screenDiscovered Claude, Codex, and Droid sessions now appear immediately from cache instead of flashing empty and slowly refilling.
• Lower CPU with multiple terminalsHidden terminal tabs stop rendering while their pane is inactive, cutting background CPU when several terminals are open.

Fixed

• Concurrent Codex chats no longer cross-talkRunning two Codex chats in the same workspace no longer leaks one chat's output into the other.
• Stuck approval promptsApproval toasts that were already handled elsewhere no longer linger, and you can now dismiss a toast manually with an X button.
• Grok terminal colorsGrok's terminal UI now renders in full color instead of a washed-out 16-color palette.
• Stale session rowsRemoving a thread no longer leaves outdated Grok session entries behind in the sidebar.
• Claude terminal startupFixed Claude's terminal-mode startup when running with bypass-permissions enabled (#73).

Lighter on your Mac, easier on your eyes, and a smarter top bar.

This release brings major performance improvements when Xanom is running in the background, a fully theme-aware light mode, and a brand-new status row in the top bar that surfaces usage, quotas, and pacing at a glance.

New

• Top bar status rowA second row under the breadcrumb shows live clock, context-usage %, Claude session/weekly quota bars, and pacing deltas. Hidden in split-pane mode.
• Per-provider lock iconThe bypass-permissions toggle now works for every provider: Claude (`--dangerously-skip-permissions`), Codex (`--full-auto`), OpenCode (`permissionMode`), and MLX (auto-approve). Tooltip names the flag per provider.
• Worktree branch chipTop bar highlights the worktree branch in green when a thread is running inside a worktree.
• Move status line to top barNew Settings toggle. When enabled, quota/pace/context info renders in the top bar and the Claude CLI's built-in statusline is suppressed so info isn't duplicated in the PTY.
• Bulk delete archived threadsHover the Archived panel to reveal a trash icon that wipes every archived thread across projects after a confirm.
• Drag-and-drop file paths into the terminalDragging files (non-images) from Finder now pastes quoted paths into the prompt instead of being silently swallowed.

Improved

• Lower idle CPU when Xanom is backgroundedGit-status and Claude-session polling now pause while the window is hidden and resume with an immediate refresh on focus. IPC coalescing matches the display refresh rate, and xterm cursor blink is suppressed when terminals are off-screen. Big win for laptops running multiple Claude terminals.
• Light-mode coverageSidebar hover/active states, tab bars, the commit dialog, git sidebar, terminal panel chrome, settings dialog theme previews, and confirmation modals all now adapt correctly in Light/Auto theme — no more white-on-white surfaces.
• Top bar pacing deltas now visibleFixed a frontend bug where the pace delta (e.g. "+5%" or "-3%") next to your quota was silently invisible. The colored delta now shows correctly.
• Pace delta readabilityPace deltas are now wrapped in parens ("34% (-6%)") so the two percentages don't read as one number.
• Claude Agent SDK updatedBumped to v0.2.138 for Claude Code v2.1.132–138 parity. OpenCode SDK bumped to v1.14.46 with HTTP query fixes, MCP tool discovery fix, and a Plan Mode security fix.
• Approval banner for Grep / GlobTools with a `path` field no longer get mis-rendered as file operations in the approval banner.

Fixed

• Codex no longer crashes when resuming a missing sessionIf a thread's saved Codex session was pruned (e.g. by `codex` upgrade or manual cleanup), the CLI used to exit instantly with "No saved session found". Xanom now checks for the session on disk and spawns fresh when it's missing.
• Settings dialog opens cleanly in Light modeThe theme preview swatches no longer render pitch-black against light backgrounds.
• Auto-approve state no longer drifts on thread switchSwitching between threads now keeps the auto-approve toggle in sync.
• Terminal no longer flashes loading screen on rapid session switchFixed an xterm re-init flash caused by an unstable `isResume` prop.

Run AI coding agents fully on-device with Apple's MLX — no cloud, no API keys.

This release adds a complete local-model chat experience built on MLX. Pick a Qwen coder model, let Xanom install the runtime for you, and chat with an agent that can read, write, edit, and run commands on your machine — all without leaving your Mac.

New

• MLX local chatBrand new provider that runs coding models entirely on your Apple Silicon Mac. Streaming responses, tool use, file edits, and bash execution work just like the cloud agents.
• Local Models settings panelBrowse curated MLX models in Settings → Local Models, see which fit your hardware tier, and download them with live progress.
• Auto-bootstrap installerFirst time you open MLX chat, Xanom detects Python, creates a venv, and installs `mlx-lm` for you with a progress banner. No terminal commands required.
• Model auto-discoveryXanom finds models you've already downloaded via LM Studio or Hugging Face and lists them alongside the curated set.
• Eject button on local modelsOne-click way to free RAM by stopping the local server from the model dropdown.
• Sticky todo barActive task todos stay pinned at the top of TaskView while you scroll through the conversation.
• TaskView terminal toggleOpen and hide an integrated terminal panel inside any task without leaving the workspace.

Improved

• Hardware-aware tuningXanom now picks a kv-cache size, max tokens, and concurrency settings that match your Mac's chip tier so local models stay fast.
• Native tool calling for MLXQwen 2.5-Coder uses the model's native tool-call format for cleaner, faster tool execution. Qwen 3-Coder automatically falls back to the XML protocol that works best for it.
• Prettier model namesThe provider/model dropdown shows readable names like "Qwen 2.5 Coder 32B" instead of raw Hugging Face IDs, while preserving tags like A3B.
• Local/Worktree branch indicator for MLXMLX threads now show the same workspace + branch chip as the cloud providers.
• Provider dropdown groupingMLX models are grouped by source (curated, LM Studio, Hugging Face) so you can find the one you want quickly.

Fixed

• MLX chat history now persistsConversations survive app restarts instead of resetting to empty.
• Model-mismatch HF lookupsFixed a bug where a stale model arg caused MLX to re-download files that were already on disk.
• Local server lifecycleCleaner shutdown, orphan cleanup on startup, and graceful kill instead of leaked `mlx_lm.server` processes.
• Image drops on MLX threadsNow show a clear "not supported" message instead of silently failing.
• Approval banner reliabilityFixed a race where approval prompts could be missed for long-running tool calls.
• Sidebar project stateCollapsed/expanded state in agent mode now persists across app restarts.
• Tab titles with ANSI codesStripped escape sequences so tab labels show the actual session name instead of color codes.

Multi-tab terminals, smarter sidebar badges, and a fix for the lock-icon crash on fresh Claude sessions.

This release brings tabbed shells to IDE mode, a configurable worktree folder layout, friendlier pace labels, and a handful of fixes that quietly remove papercuts you've been seeing in the sidebar and on long-running OpenCode turns.

New

• Tabbed terminals in IDE modeThe bottom terminal panel now supports multiple shells side-by-side. Click the + to spawn another shell in the same working directory, switch with a click, close individual tabs, or close them all to dismiss the panel.
• Branch-first worktree layoutNew Settings toggle that organizes task worktrees as `<root>/<branch>/<repo>` instead of `<root>/<repo>/<branch>`. Useful when you tend to think branch-first and want sibling repos grouped under one feature.
• Commit button in task worktree headerEach task worktree now has a dedicated Commit button in the header that opens a slimmed-down commit dialog (no Create PR — that lives in its own button).
• Per-file diff counts in Codex sessionsCodex `apply_patch` blocks now render with a colored unified-diff style and show a `+N / -M` summary per file, matching what you already see for Claude sessions.

Improved

• Sidebar diff stats stay freshThe `+lines / -lines` badge next to each Claude session now updates live as the agent edits files. Sessions whose first scan landed before any tool-use diffs were on disk no longer get stuck at `0+ / 0-` until you click them.
• Model badge shows on cold startClaude sessions now display their model in the sidebar even when the most recent transcript chunk doesn't contain an assistant turn (long tool result, mid-tool-call, etc.). No more opening a session just to learn what model it's on.
• OpenCode handles long agent turnsLong-running OpenCode SDK turns no longer surface a misleading "timed out" error after 30 seconds. The Stop button stays visible for the full turn and turn-end is signaled cleanly when the agent finishes.
• Friendlier pace labelsUsage/pace badges now read "Behind pace by X%", "On track", "Ahead of pace by X%", or "Well over pace by X%" with a sensible decimal for small deltas.
• Polished New Task dialogSurfaces, borders, and the error banner now respect light/dark mode and feel more consistent with the rest of the UI.

Fixed

• Lock-icon no longer crashes a fresh Claude sessionToggling the permission-skipping (lock) icon on a brand-new Claude terminal session before sending any message previously crashed the TUI with `No conversation found with session ID: …`. Xanom now verifies the underlying transcript actually has a real turn before passing `--resume`, and otherwise starts cleanly.
• OpenCode threads receive their first-message attachmentsImage attachments handed off from the draft chat to a freshly-created OpenCode SDK thread are now forwarded in the OpenCode-native shape, so the agent sees them on the very first turn instead of silently losing them.
• Effort auto-downgrades when leaving Opus 4.7XHigh effort is restricted to Opus 4.7. Switching to any other model now automatically downgrades effort instead of leaving the session in an invalid state.
• Files.persisted events surfaced from OpenCodeFile-write events from the OpenCode SDK bridge now reach the frontend, so the UI can react when the agent actually persists files to disk.

Task Mode is now available on every machine.

A small follow-up to v2.0.2 that unlocks Task Mode for all users and picks up the latest Claude Agent SDK parity patch.

New

• Task Mode for everyoneThe worktree-isolated task view (Cmd+Shift+T) is no longer gated to a single development machine. Toggle into task mode and run agents in their own git branches with the dedicated review sidebar.

Improved

• Claude Agent SDKUpdated to 0.2.123 for parity with Claude Code 2.1.123.

Codex gets smarter, Claude chat feels lighter, and a handful of papercuts disappear.

This release expands Codex with custom slash commands, attached images, persistent auto-approval rules, and an "Extra High" reasoning effort. Claude SDK chat is more compact and unified visually, multiview tab restoration is fixed, and dialogs/toasts are properly themed across the app.

New

• Codex slash commandsType `/` in a Codex chat to run built-in commands or your own custom prompts from `~/.codex/prompts/`.
• Image attachments in Codex chatDrop or paste images straight into a draft Codex thread; they're sent with your first message.
• Codex auto-approval rulesApprove once and Xanom can remember the decision for that command pattern, so trusted tool calls stop interrupting you.
• "Extra High" reasoning effort for CodexA new `xhigh` option for the deepest reasoning pass when you really want Codex to think.
• MCP tool blockA dedicated, keyboard-accessible block for MCP tool calls in chat, with proper screen-reader support.

Improved

• Tighter Claude SDK chat layoutReduced spacing around assistant messages, thinking blocks, and todo-list tool blocks so more of the conversation fits on screen.
• Unified user message bubblesUser bubbles are now the same font size in Codex and Claude SDK chats.
• Theme-aware approval toasts and dialogsApproval toast text and buttons, the Quit Xanom confirmation, and the tool detail modal now respect the dark theme correctly.
• Cleaner ThinkingBlock stylingThinking blocks now match the visual style of the other tool blocks.
• Faster CI release buildsApple Silicon and Intel builds now run in parallel on self-hosted runners, so updates ship sooner.
• Claude Agent SDK updatedBumped to 0.2.122 (parity with Claude Code 2.1.122).

Fixed

• Multiview Claude tabs resume correctlyRestored Claude tabs in split view no longer fail with "no conversation found" after restarting the app.
• Quit Xanom dialog no longer freezesThe Cmd+Q confirmation now properly transitions into the shutdown spinner instead of locking up.
• Codex auto-approval failures fall through cleanlyWhen Codex can't auto-approve internally, it now prompts you manually instead of silently hanging on the request.
• Codex image attachments clear after sendPending images and references are properly cleaned up after the first message ships.
• Tool detail modal background darkenedNo more washed-out modal background in the chat interface.

Polish pass — terminal redraw, light-mode visuals, and SDK chat reliability.

Small but high-impact fixes for everyday flows: switching back to a terminal tab now repaints reliably, light mode no longer shows dark bands above editor tabs, and Claude SDK sessions resume against the right transcript more often.

Improved

• Terminal redraw on tab switchSwitching back to a terminal that was hidden no longer leaves a black panel until you resize the window; the redraw runs every frame for ~200ms so the paint always catches WKWebView's compositor.
• Manual terminal refresh now triggers a real SIGWINCHHitting refresh on a stuck terminal wiggles the PTY size by one row and back, which forces TUI apps (Claude Code, vim, etc.) to repaint and clears any rendering artifacts.
• xhigh effort levelThe Claude effort selector now persists and restores `xhigh` alongside `low`, `medium`, `high`, and `max`.
• Distinct accents for Skill, MCP, and other toolsTool-use blocks now use different left-border colors so Skill calls (purple), MCP tools (cyan), and uncategorized tools (zinc) are easier to spot at a glance.
• Light-mode editor and file treeRemoved a dark band above editor tabs and re-skinned the file tree filter input so they fit cleanly into a light workspace.
• Stable input bar spacingIn SDK chat, the gap between the last message and the input bar no longer shifts as turns start and stop.

Fixed

• Thinking content no longer drops mid-turnStreamed-text and streamed-thinking states are now tracked independently, so models that stream text live but emit the thinking block only in the final assistant message (extended-thinking summary mode, certain effort levels) keep their reasoning visible.
• Claude SDK session resume binds to the right transcriptWhen the SDK reports a logical session ID that doesn't match the actual JSONL transcript filename, Xanom reconciles it from the first few user prompts and persists the corrected ID, so chat history loads cleanly on resume.
• More accurate Codex pace trackingPace status now flips to behind/ahead on a smaller threshold and the weekly window stops dampening short bursts, so the pace label tracks current usage more closely.

Xanom 2.0 — every coding agent, one home, one workflow.

This release brings OpenCode to Xanom as a full chat experience alongside Claude, introduces a redesigned Task mode with isolated worktrees and built-in pull-request flow, and unifies every provider's sessions on a single home screen. Plus dozens of quality-of-life upgrades to usage tracking, the input bar, and the sidebar.

New

• OpenCode chat is hereTalk to OpenCode the same way you talk to Claude: file mentions, image drops, permission prompts, model picker, and live token + cost stats. No terminal required.
• Sign in to OpenCode providers from inside XanomAdd API keys or run OAuth flows for any OpenCode provider without leaving the app.
• Browse every OpenCode model in one dropdownIncluding providers you haven't connected yet, sorted by recents and grouped logically. Fuzzy-search across long lists.
• Task modeA dedicated workspace for running agents inside isolated git worktrees, with a built-in review sidebar, status pills, and per-task agent tabs.
• One-click pull requests from a taskCreate PRs straight from the task view, with auto-generated titles and descriptions.
• Unified home screenEvery provider's recent sessions (Claude, Codex, OpenCode, Droid) show up in one deduplicated list with pretty model names.
• Live usage overlaySee your current token burn, cost, and pacing in real time as the agent works.
• Auto permission modeLet Claude SDK chats run hands-free without approving every tool call.
• Keep-awakeXanom keeps your Mac from sleeping while an agent is actively running, so long jobs finish even if you walk away.
• Live diff stats in the sidebarSee lines added/removed update in real time as the agent edits files.
• Resumable local-model downloadsIf a local LLM download gets interrupted, it picks up where it left off with a visible progress bar.
• Did-you-know tipsFriendly tips that surface lesser-known features as you work.
• Provider shown in the thread top barAt-a-glance confirmation of which agent you're chatting with.

Improved

• Redesigned input barCleaner layout with a live context ring, shared composer dropdown across providers, and consistent file-mention + image handling.
• Revamped approval promptsThe approval toast now shows the full command on hover, with cleaner styling and proper routing for OpenCode.
• Git sidebar overhaulReworked branches, diffs, and PR creation; branch matching is now case-insensitive.
• Better model names everywhereNo more cryptic IDs in dropdowns, tabs, or session lists.
• Smarter session resumeReopening a Claude task resumes the existing session instead of starting fresh.
• Project memory in Task modeXanom now remembers the last project you used and your worktree state across restarts.
• Faster, calmer home screenDuplicate sessions across providers are now collapsed into a single entry.
• Tighter Codex chatUpdated session lifecycle and event handling for smoother streaming.
• Better commit dialog and thread top barRefreshed visuals with clearer state indicators.

Fixed

• Phantom sessions are goneFixed an issue where deleted sessions could reappear in the sidebar.
• Stale diff stats no longer lingerNumbers in the sidebar always reflect the actual state of your worktree.
• Approval state stays in syncNo more leftover "approval pending" indicators after you've already responded.
• Task-mode model picker remembers your choiceThe model you select now persists across restarts.
• Drafts auto-submit reliablyFixed a race that could swallow the first message in a new task.
• Toast notifications no longer fire mid-promptCompletion toasts wait until the agent is actually done.
• OpenCode tasks route through the SDK viewDrafts targeting OpenCode now open in the right chat experience.
• Bundled OpenCode binaryXanom auto-detects and ships its own OpenCode runtime, so you no longer need to install it separately.
• Hono security patchUpdated the OpenCode SDK to pull in a security fix for the Hono web framework.
• Markdown, terminal links, and command-block rendering glitchesLots of small visual fixes across chat output.

New Providers

• OpenCode supportLaunch and manage OpenCode sessions alongside your other agents, with full terminal integration.
• Ollama Cloud supportConnect to Ollama Cloud models directly from the provider picker; cloud models now load eagerly so they appear instantly on hover.

Ollama Improvements

• New Ollama Terminal buttonOne-click new terminal with a centered model picker so you can jump into a local model in seconds.

Claude Code

• Terminal-only Claude CodeClaude Code now runs exclusively in the terminal view for a snappier, more consistent experience.
• Five-level effort systemNew `xhigh` reasoning mode joins low / medium / high / max, giving you finer control over how hard Claude thinks on tough problems.

Terminal Rendering Fixes

• Fixed garbled glyphs on Retina displaysSwapped the terminal renderer to Canvas to resolve the WebKit DPR bug that caused fuzzy or scrambled text.
• No more duplicate-frame flashesTerminal snapshot rehydration now uses offset-based dedup, so switching threads no longer flickers.

UI Polish & Accessibility

• Ctrl+Enter to submitAdded the keyboard shortcut for the task creation dialog so you can fly through the flow.
• Better notification panelThe notifications history panel now has proper dialog semantics, focus management, and keyboard navigation.
• Terminal errors surface in notificationsFailed terminal spawns now show up in the notification panel instead of disappearing silently.

Other

• Safer archivalReplaced the native browser confirm with Tauri's dialog for a cleaner, more reliable archive confirmation.

Collapsible Sidebar Rail

• Icon-only sidebar when collapsedpressing ⌘B now shrinks the sidebar to a slim icon rail instead of hiding it completely, so you can still switch tabs and expand it without losing your place

Scroll to Bottom in SDK Chat

• Quick-scroll buttona "scroll to bottom" button now appears in Claude SDK chat when you scroll up, letting you jump back to the latest output instantly

SDK Session Stability

• Messages no longer lost on multiview reopenif you sent a message right after reopening a split-view pane, it was silently dropped; now it's buffered and automatically retried
• Slash commands no longer crash inactive sessionstyping a slash command when the SDK session isn't ready now shows an error message instead of crashing
• Session cleanup always runsclosing a thread mid-turn no longer leaves the session running in the background

Terminal Reliability

• No more missing output before exitterminal output that arrived just before a process exited could occasionally be lost; now all output is guaranteed to arrive before the exit notification

Bug Fixes

• Deleting Claude sessions with trailing-slash pathsproject paths ending in `/` no longer cause session deletion to silently fail
• Archived threads restore correctlyunarchiving a thread now works reliably even if the sidebar hasn't fully loaded yet
• Droid sessions stay visible on spawn failureif starting a Droid session fails, it no longer vanishes from the sidebar
• Command palette keyboard guardarrow keys in an empty command palette no longer cause unexpected behavior
• Tab bar alignmentleft padding is now correct when the sidebar is collapsed

Under the Hood

• Claude Agent SDK updated to v0.2.104picks up the latest stability fixes and improvements from the SDK

Slash Commands in SDK Mode

• Completely overhauled slash command handlingcommands like /compact, /clear, and custom commands now route through a dedicated path instead of fragile regex detection, eliminating crashes and message disappearance
• Slash commands work immediatelysending a slash command before the session is fully started now works reliably instead of hanging or crashing
• Concurrent command protectionrapidly sending multiple slash commands no longer causes race conditions

Clickable File Paths in Terminal

• Cmd-click to open filesfile paths printed in the terminal are now clickable links that open directly in the built-in editor

Better Chat Rendering

• Newlines display correctlyASCII art, plain-text lists, and multi-line content from agents now render with proper line breaks instead of collapsing into a single line

Smarter Approval Notifications

• Tool name in notification titleapproval notifications now show which tool needs approval (e.g. "Xanom — Bash Approval") so you can triage without switching windows
• Dangerous command warningsnotifications flag potentially dangerous commands so you know when to pay extra attention

Subagent Reliability

• Stop button no longer gets stuckwhen agents use sub-agents (like explore or search), the stop button stays responsive and the session doesn't stall mid-stream

Thread Management

• Archive and delete persist across restartsdeleting a Claude session now properly removes the transcript file so it doesn't reappear on next launch

Terminal & IDE Improvements

• Live font size changeschanging terminal font size no longer causes a full terminal rebuild; the resize is instant and seamless
• All font sizes availablethe quick appearance popover now shows every terminal font size option instead of every other one
• Tab switching fixedswitching between terminal tabs in IDE mode now properly redraws the canvas instead of showing a blank pane
• Quieter session handlingno more false "stream ended" error banners when you intentionally stop a session

Two New AI Providers: Droid & OpenCode

• Droid supportYou can now run Droid sessions directly inside Xanom. Your model selection is remembered across threads, sessions auto-launch when you switch back to them, and loading spinners keep you informed while things start up.
• OpenCode supportOpenCode is now available as a provider. Create threads with OpenCode just like you would with Claude Code or Codex.
• Automatic session discoveryDroid sessions you've started outside Xanom are detected and surfaced in your project sidebar, so you can pick up where you left off.

Command Palette

• Cmd+K to search everythingA new command palette lets you quickly jump to any project, thread, or action with fuzzy search. Type a few letters and go.

Redesigned Terminal Engine

• Faster, smoother terminalThe terminal rendering engine has been completely rebuilt. Scrolling is smoother, text is crisper, and switching between tabs no longer freezes the screen.
• Instant tab switchingWhen you switch back to a terminal tab, your scrollback is restored immediately instead of replaying line by line.
• Better font handlingChanging your terminal font or font size now updates cleanly without garbled text or needing to restart the session.

Archived Threads

• Archive old threadsYou can now archive threads you're done with to keep your sidebar clean. Archived threads are accessible from a dedicated panel, and you can unarchive them anytime.

Notification History

• Never miss a notificationA new notification history panel tracks all your agent notifications with unread indicators, so you can review what happened while you were away.

Quick Appearance Switcher

• Fast theme and font switchingA new popover lets you quickly change your theme and font without opening the full settings dialog.

Bug Fixes

• Fixed memory leaksSwitching between threads no longer leaves old sessions running in the background. Agent subprocesses (MCP servers, language servers, etc.) are now properly cleaned up when you navigate away.
• Fixed SDK session resumeResuming a Claude SDK chat session now works reliably, even when the session ID stored internally didn't match the one on disk.
• Fixed terminal font size changesChanging font size now properly updates the terminal subprocess dimensions, so text reflows correctly.
• Fixed Droid terminal panelsThe terminal and git sidebar toggles in Droid sessions now work as expected.
• Fixed Droid spawn loopA bug that caused Droid sessions to endlessly retry spawning on failure has been resolved.
• Fixed hook deliveryHook payloads for Claude and Droid relay scripts are now delivered reliably.
• Fixed terminal rehydration artifactsLeftover sentinel markers no longer appear when restoring a terminal session.
• Fixed OpenCode plugin pathsFile paths with special characters are now encoded correctly in OpenCode's relay plugin.
• Fixed Homebrew cask publishingAuto-update releases now publish to the Homebrew tap correctly on CI.

IDE Mode

• Full IDE layoutSwitch to IDE mode with `Cmd+Shift+.` for a four-panel workspace: file tree, code editor, integrated terminal, and AI chat — all in one window
• Tabbed terminalsOpen multiple terminal sessions side-by-side within IDE mode, each with its own ghostty-web instance
• Integrated AI chatChat with Claude directly in the IDE sidebar, with full SDK session support, history dropdown, and thread creation
• Resizable panelsDrag to resize any of the four IDE zones to fit your workflow

Quick Open

• Cmd+P file searchInstantly find and open any file in your project with fuzzy matching, just like VS Code

Git Integration

• File tree status indicatorsSee which files are modified, added, or untracked at a glance with color-coded icons in the file tree
• Right-click context menuOpen, reveal in Finder, copy path, or perform actions on files directly from the file tree

Smart File Tracking

• Auto-open on AI editWhen the AI agent edits a file, it automatically opens in your editor with an animated badge so you never miss a change

Thread Management

• Rename threadsRight-click any conversation thread to give it a custom name that persists across sessions

Reliability Improvements

• SDK session recoveryChat sessions now properly detect and handle new vs. resumed threads, eliminating "could not resume session" errors
• Fixed usage trackingToken counts and cost tracking now update reliably
• Scroll and accessibility fixesResolved nested scroll issues in editor tabs and improved screen reader support

Undo File Changes

• Rewind any turnclick "Undo changes" on any assistant message to revert all file modifications from that turn, with clear loading and success/error feedback

Improved Tool Visibility

• Better command previewstool executions now show cleaner previews with truncated content for large file writes, so you can see what's happening without information overload
• Dedicated thinking displayClaude's reasoning process is now shown in a distinct, collapsible block so you can follow along or hide it
• Files changed cardseach turn now shows a summary card of which files were modified

Smarter Approval Dialogs

• Approval banner for tool usetool approval requests now appear as a prominent banner at the bottom of the chat instead of inline, making them impossible to miss
• User input questionswhen Claude asks a question, it now uses the same banner UI for a more consistent, polished experience
• Better fallback parsingapproval dialogs now gracefully handle edge cases, showing file paths and action types even when details are incomplete

Codex Enhancements

• Fast mode toggleenable or disable fast mode for Codex sessions directly from the draft view
• Effort and permission settingseffort level and permission mode choices now persist between sessions

SDK Session Improvements

• Slash command support/compact and /clear commands now work properly in SDK chat sessions with automatic session recovery
• Background agent trackingsub-agents spawned during a session are now tracked and displayed in the chat
• Token usage displaysee detailed input/output/cache token counts and context window consumption during conversations
• Compacting indicatora visual indicator now appears when the SDK is compacting context to free up space
• Session persistencePTY sessions now survive app restarts so your sidebar threads stay intact

Bug Fixes

• Fixed animation consistencyall Codex message items now animate in consistently
• Fixed tab managementimproved tab display and pane cleanup in split view

Undo File Changes in SDK Chat

• Rewind button on turn summariesAfter Claude edits your files in SDK chat mode, you can now click "Undo" on the turn summary to revert all file changes back to how they were before that turn

Background Agent Tracking

• Live status for background agentsWhen Claude dispatches agents in the background, you now see running/completed/failed badges with progress info instead of a silent spinner
• Agent group headers update in real timeThe collapsible agent group shows how many background agents are still running

Smarter Approval Prompts

• Human-readable approval detailsTool approval banners now show the actual file name and path (for file edits) or the command being run (for bash), instead of raw JSON
• Graceful fallback for broken payloadsEven if approval data is truncated, Xanom extracts what it can and shows a friendly summary

Codex Effort & Permission Controls

• Reasoning effort selectorChoose between Low, Medium, and High reasoning effort for Codex sessions directly from the draft view
• Permission mode toggleSwitch between default and full-access permission modes when starting a Codex session

Codex Stability Improvements

• Server disconnection recoveryIf the Codex server crashes or disconnects, Xanom now detects it immediately, clears the spinner, and shows a clear error instead of hanging forever
• 5-minute safety timeoutStuck Codex turns that receive no response for 5 minutes are automatically cleared so you're never left waiting on a frozen session

Files Changed Cards

• Per-turn file change summarySDK chat now shows a dedicated card listing all files that were created or modified during each turn, making it easy to see what changed at a glance

Tab Switching Fixes

• Sidebar syncs with active tabSwitching between tabs in split view now properly updates the file browser to show the correct project context

Fun Working Indicators

• 150+ playful status verbsThe SDK thinking spinner now cycles through a much larger variety of fun verbs like "Clauding", "Percolating", "Moonwalking", and "Flambéing" while your agent works

Xanom 1.0

• The big one. Xanom graduates to v1.0 — a complete AI coding agent manager for macOS with two powerful ways to interact with your agents.

Claude Agent SDK Mode

• Chat with Claude using the structured SDKa new interaction mode that gives you rich, structured conversations alongside the classic terminal experience
• See tool calls as they happenwatch Claude read files, edit code, run commands, and search your codebase with dedicated visual renderers for each tool type
• Approve or reject tool use inlineapproval banners show exactly what Claude wants to do, with options to allow once, always, or for the whole project
• Pending approval queuewhen multiple approvals stack up, see a count badge and work through them one by one

Setup Wizard

• Guided first-run experiencea 4-step setup wizard walks you through configuring Xanom when you first launch the app
• Re-run anytimefind the "Re-run Setup Wizard" button in Settings > General if you need it again

@ File Mentions

• Reference files directly in your promptstype @ in the input bar to get autocomplete suggestions from your project directory
• Browse your project treethe file mention popup lets you navigate folders and select files to include as context

Thread Forking

• Fork any conversationbranch off from any point in a thread to explore a different direction without losing the original
• Preserved historyforked threads keep the full conversation up to the fork point

Git Worktree Support

• Isolated workspaces per threadeach agent thread can work in its own git worktree, keeping changes separate until you're ready to merge

Draft Composer

• Compose before you commitwrite and refine your initial prompt in a draft view before spinning up an agent session
• Pick your model upfrontchoose provider and model from the draft view so the thread starts exactly how you want

Smarter Tool Rendering

• Inline todo listsClaude's task lists now render inline with live status indicators (pending, in-progress, completed) instead of hidden behind a click
• Skill names visible at a glancesee which skill Claude is invoking without expanding the tool block
• Grouped tool activityconsecutive tool calls are visually grouped with loading spinners and colored status indicators
• Plan follow-up promptswhen Claude completes a plan, get a follow-up banner to continue the conversation

Paginated Chat History

• Infinite scrollbacklong conversations no longer hit a wall; scroll up to load earlier messages automatically with cursor-based pagination

Improved Session Naming

• Smarter thread titlesslash commands get instant deterministic names instead of waiting for LLM summarization
• More accurate summariestightened prompts prevent overly embellished or generic thread names

Context Ring

• Better token usage displaythe context ring now shows more accurate token counts for your conversation

UI & Quality of Life

• Light mode fixesimproved text visibility and contrast across the app in light mode
• Image attachmentsattach images to your prompts for visual context
• Model selection dropdownquickly switch models from the input bar with a clean dropdown UI

Bug Fixes

• Fixed stuck sidebar spinnerthe processing indicator no longer gets stuck when an SDK session completes
• Fixed chat scrollback issuesresolved a bug where long conversations would jump or fail to scroll properly
• Fixed race conditions in session file watchingpartial JSONL lines no longer cause parsing errors
• Cleaner process cleanupfailed SDK sessions now properly clean up orphaned child processes
• Sidecar error handlingunhandled rejections and exit conditions are now caught and surfaced properly

Streamlined Providers

• Focused on Claude Code and Codexremoved the experimental Ollama integration to streamline the app around the two production-ready providers

Light Mode Overhaul

• Dramatically improved light mode visibilitytext, icons, buttons, and borders across the entire app are now properly readable in light mode
• Settings dialog fixedsection labels, slider tracks, and segmented buttons are now clearly visible against the glass overlay
• Chat view fixeduser message bubbles, Claude's name, thinking indicators, tool use blocks, and approval banners all render with proper contrast
• Chat input text fixedtyped text is now dark and readable in light mode instead of invisible white-on-white
• Top bar fixedTerminal/Chat/Split tabs, action buttons, and IDE selector are now visible in light mode
• Sidebar elements fixeddrag handles, terminal session icons, usage dashboard, and MCP server paths are no longer invisible

Terminal Light Theme

• Shell terminals now use a light color schemethe standalone terminal panel and general terminal views automatically switch to a light background with dark text when in light mode
• Claude Code terminal stays darksince Claude Code's TUI manages its own colors, its terminal keeps the dark theme for proper rendering

Better Text Contrast

• Improved muted text readabilitydarkened the muted and tertiary text colors in light mode so labels and descriptions stand out against glass surfaces
• Context ring adapts to themethe token usage ring in the status bar now uses theme-aware colors instead of hardcoded values

Live Tab Status Indicators

• Pane tabs now show real-time statusan amber pinging dot appears when a session needs your attention (e.g. tool approval), a blue spinner shows when the agent is actively working, and an emerald dot signals when a task is complete
• Tab underlines match statusthe active tab's accent bar changes color to match the current state so you can spot what needs action at a glance

Input Field No Longer Clears During Approvals

• Your typed text is preserved when a permission dialog pops up — previously, if you were typing while an approval banner appeared, your input would be lost

Image Attachment Fix

• Messages with attached images now display correctlyfixed a bug where sending an image with your message could cause duplicate or missing chat bubbles

Smarter Unread Badges in Split View

• Sessions you're actively viewing no longer get marked as unreadin split view, only sessions outside your focused pane will show unread indicators, reducing unnecessary visual noise

Ollama Agent Support

• Run local AI models directly in XanomOllama is now a first-class provider alongside Claude Code and Codex. Create Ollama threads from the sidebar, pick from your installed models, and chat with full tool-use support (file read/write/edit, bash, search).
• Agentic tool loopOllama models can read, write, and edit files in your project with an approval flow before each action, just like Claude Code.
• Auto-starting sessionsOllama threads start automatically when selected, no extra clicks needed.
• Git sidebar and terminalOllama threads get the same ThreadTopBar with commit, open-in-IDE, and git sidebar controls.

Chat Reliability Improvements

• Fixed messages sometimes not appearingAdded a polling fallback so chat messages always show up, even when macOS file events are missed.
• Slash command messages now visibleSending a slash command no longer causes your message to disappear from the conversation.
• Better tool approval bannersSubagent permissions and Agent/Task tool uses now show clearer descriptions in the approval prompt.

New Fonts

• Zed Sans and Zed MonoTwo new font options available in Settings > Typography. Clean, modern fonts designed for code editors.

Settings & Permissions

• Image read whitelistNew toggle in Settings to automatically allow Claude Code to read images you drag into Xanom, so you no longer need to manually approve each image read.

Visual Polish

• Redesigned thinking spinnerThe processing indicator is now a smooth 13-arm starburst animation, replacing the old cross spinner.
• Improved dark theme stylingRefined CSS with better scrollbar styling and glass effects.

Security Hardening

• Path traversal protectionFile operations now validate paths stay within the project boundary.
• Race condition fixesFixed potential issues with concurrent file deletion and model locking.
• Command injection preventionHardened shell command construction for search tools.

Parallel Agent Display

• Grouped agent viewwhen Claude launches multiple agents in parallel, they now appear as a single grouped block instead of deeply nested individual items
• Live progress trackingthe group header shows how many agents are running vs completed, with spinner and checkmark indicators
• Agent type badgeseach agent displays its type (e.g. code-reviewer, Explore) so you can tell them apart at a glance

Session Stability Fix

• Fixed stuck "Processing..." indicatorresolved a race condition where the processing spinner would get stuck indefinitely after sending a message, requiring you to restart the session
• More reliable session statesession state now syncs correctly even when Claude's internal session ID is mapped after hook events have already fired

Usage Dashboard

• Session cost trackingsee how much each coding session costs across Claude Code and Codex, broken down by model
• Pace monitoringtrack your daily and weekly spending pace with visual indicators
• Model breakdownview token usage and costs per model in a clear dashboard

Codex Interactive Mode

• Full Codex supportrun OpenAI Codex sessions directly in Xanom with interactive mode
• Keyboard shortcutsnavigate and control Codex sessions with keyboard shortcuts just like Claude Code

Terminal Clickable File Paths

• Click to open filesfile paths mentioned in terminal output are now clickable links that open directly in your editor
• Smart path resolutionrelative paths are resolved against your project directory automatically

MCP Server Management

• Improved server controlsbetter UI for managing MCP (Model Context Protocol) servers in settings
• Server status visibilityclearer indication of which MCP servers are running

Global Keyboard Shortcuts

• Quick navigationnew keyboard shortcuts for switching between agents, terminals, and views
• Multi-view tabsimproved tab management in split view with keyboard support

Bug Fixes

• Fixed loading spinner flickerthe loading indicator no longer flickers between tool calls during Claude Code sessions
• Fixed file path linksterminal file path detection is more reliable across different output formats
• Improved thread view switchingsmoother transitions when switching between chat and terminal views

Light Mode

• Switch between Dark, Light, and System (follows your macOS appearance) in Settings → Appearance → Color Mode
• All glass themes adapt automatically — surfaces, text, borders, and scrollbars adjust for comfortable daytime use

Search Across Sessions

• New search dialog (Cmd+K) lets you search across all your sessions by name or prompt content
• Results are ranked by relevance and take you straight to the matching session

Real Context & Cost Tracking

• The context ring now shows real token usage from Claude's API — input/output tokens, cache reads/writes, session cost, and turn count
• Before the first API response, falls back to an estimate marked with ~

Custom Notification Sounds

• Choose your notification sound in Settings → Behavior
• Pick from the bundled Xanom sound, macOS system sounds (Basso, Glass, Ping, etc.), or go silent

Better Session Ordering

• Sessions in the sidebar now sort by when you last sent a prompt, keeping active sessions at the top

Terminal Improvements

• Font size setting now respected in all terminal panels
• Clipboard paste uses Tauri's native clipboard API, fixing paste issues in the macOS WebView
• Auto-refresh no longer interrupts scrollback — scrolling up won't snap you back to the bottom
• Reduced refresh frequency (4s instead of 1s) for less CPU usage when idle
• Debug logging stripped from production builds for cleaner console output

Smoother Chat Scrolling

• Auto-scroll no longer fights with Virtuoso's built-in scroll when Claude is actively working, eliminating jitter

Session Naming Fix

• Slash commands with arguments (like `/release 0.4.10`) now get proper session names

Vibrancy Update

• Window vibrancy uses the Sidebar material for a cleaner frosted-glass look matching native macOS apps

Git Account Validation

• Email addresses are now validated when adding git accounts in Settings

What's New in v0.4.9

Browser-Style Navigation

Navigate between sessions like browser tabs — new Back and Forward buttons in the sidebar toolbar let you quickly jump between recently viewed sessions without scrolling through the sidebar.

Redesigned Sidebar Navigation

The sidebar tabs have been reorganized into a clean vertical list with a dedicated Home button to return to the welcome screen. Switching to the Agents tab now intelligently restores your last viewed session instead of dropping you on an empty screen.

Right-Click Context Menus

Right-click any session or thread in the sidebar to access quick actions:

• Renameedit the name inline
• Archivearchive threads you're done with
• Deleteremove threads entirely
• Hidetemporarily hide Claude or Codex sessions you don't want to see

Smoother Sidebar Animations

Collapsing and expanding the sidebar now has a smooth slide animation instead of an instant snap. The expand button also fades in gracefully.

Fixed Paste in Terminals

Pasting text via right-click or keyboard now works reliably across all terminal types. Previously, ghostty-web could mangle pasted text into garbled characters (em-dashes, random symbols). Also added support for pasting via the system context menu in Tauri's WebView.

More Accurate Context Usage Ring

The context usage indicator now correctly includes both input and output tokens, giving you a more accurate picture of how much context window is being used.

Fixed Loading Bar Disappearing

The loading bar no longer vanishes during long sessions. Previously, it could disappear after context compaction or between batched operations — it now correctly reappears whenever the agent resumes work.

Fixed Approval Queue Not Advancing

Tool approval prompts now reliably advance to the next queued item after you approve one, regardless of whether hooks are active.

Fixed Session Data Cross-Contamination

Resolved a bug where background sessions could accidentally pick up data from other sessions, causing the wrong content to appear in the wrong session view.

Fixed Duplicate Content Blocks

Messages with multiple content blocks (e.g. thinking + text in the same response) no longer disappear or flicker due to UUID collisions.

Open Existing Folder Shortcut

The "Open Existing" button on the home screen now opens a native folder picker directly, instead of routing through the new project dialog.

Settings Improvements

• Updated cloud modelsAdded DeepSeek R1 70B and Qwen QwQ 32B to Groq model options; removed deprecated models
• System infoThe About page now shows your app version, Tauri version, and platform
• Better layoutFont selectors and settings controls are now properly right-aligned
• Keyboard shortcutsUpdated description to clarify context-dependent behavior

Performance

• Reduced console logging overhead throughout the app
• Usage panel now refreshes every 10 minutes instead of 5
• Drag handles in the sidebar no longer cause layout shifts on hover

What's New in v0.4.8

Faster Terminal Responsiveness

The terminal display now refreshes much more frequently (every 4 seconds vs every 30 seconds), so layout changes and content updates appear almost instantly. This refresh is also smart — it only runs when the terminal is actually visible on screen, so background sessions don't waste any CPU.

Fixed Spurious Notifications

Fixed a bug where Xanom would sometimes show "Claude needs your attention" notifications even when Claude had already finished working. Empty status pings from Claude Code are now correctly recognized as idle signals instead of being misclassified as requiring your attention.

Cleaner Notification Behavior After Task Completion

Removed a workaround that was suppressing notifications too broadly after Claude stopped working. Notifications are now handled correctly at the source — you'll get notified when Claude genuinely needs your input, and won't be bothered when it doesn't.

What's New in v0.4.7

Full Permissions Mode

You can now run Claude with full permissions directly from Xanom — no more manually restarting with flags.

• Lock button in the top barclick the lock icon to toggle full permissions mode. Xanom will automatically restart the session with `--dangerously-skip-permissions` and accept the bypass prompt for you.
• Permission selector in Chat viewthe permission dropdown in the chat input bar is now wired to the same control, so you can switch modes from either place.
• A loading overlay appears while the session restarts, with automatic recovery if the restart gets stuck.

Multiline Input in Terminal

Shift+Enter now inserts a newline in the terminal input instead of submitting the prompt. This lets you compose multiline messages before sending them to Claude.

Fewer Unwanted Notifications

Fixed a bug where a delayed "Claude needs your attention" notification would pop up 30–60 seconds after the agent had already finished working. Since you already receive a "Claude Finished" notification, this duplicate has been suppressed.

Chat Experience

• Chat now starts at the bottomwhen you open a session, you see the latest messages immediately instead of having to scroll down
• Auto-scroll fixedchat properly follows new messages as Claude works, including the thinking spinner at the bottom
• Scroll-to-bottom button now reliably jumps to the very end of the conversation
• No more phantom permission popupsauto-approved commands no longer trigger unnecessary approval prompts in chat view

Approval Notifications

• Smarter approval toastsmultiple pending approvals from the same session are now collapsed into a single toast with a count badge (e.g. "Action Required (+2)") instead of flooding your screen
• Notification throttlingnon-critical notifications are now rate-limited so you don't get spammed with OS alerts while Claude is working

Codex Integration

• Full permissions mode now workstoggling "Full Permissions" in Codex sessions actually sends the setting to the backend now (previously it only changed the UI toggle without effect)

New Fonts

• 5 new font options in Settings → Typography:
• UI fonts: SF Pro, New York
• Monospace fonts: SF Mono, Menlo, Source Code Pro

Safety & Security

• Expanded command warningsXanom now flags 11 additional dangerous command patterns before you approve them, including disk tools, filesystem formatters, world-writable permissions, git clean, redirects to sensitive directories, fork bombs, and more

Bug Fixes

• Fixed settings layout where label text would get squeezed when controls wrapped to a new line
• Fixed terminal getting stuck on "Receiving session data..." during new session creation
• Fixed stale processing state that could cause the spinner to persist after Claude finished
• Fixed session state machine handling duplicate session-start events mid-session

What's New in v0.4.5

Smoother, faster chat scrolling

Long conversations now use virtualized rendering — only the messages on screen are drawn, so even sessions with hundreds of messages scroll smoothly without lag.

Rename sessions right from the sidebar

Click the pencil icon next to any Claude session to rename it inline. No more hunting through menus — just click, type, and press Enter.

Control Claude's reasoning effort

A new selector in the input bar lets you choose how hard Claude thinks before responding: low, medium, high, or max. Useful for quick answers vs. deep problem-solving.

Collapsible usage panel

The token usage panel in the sidebar now collapses on hover/focus, giving you more room when you need it and staying out of the way when you don't.

Better commit messages

Commit message generation is now faster and produces higher-quality summaries by sending actual diff content to the LLM instead of just file names.

Image attachments in chat

You can now see image attachments displayed directly in the conversation for both Claude and Codex sessions.

Codex timeline view

Codex sessions now display all events — messages, tool calls, and file changes — in a single chronological timeline, making it easier to follow what happened.

Fixes

• Approval notifications now correctly navigate to the right session, even after switching between projects
• The amber "needs attention" dot no longer disappears while waiting for tool approval
• Codex sessions show their names immediately in the sidebar instead of loading as blank
• Terminals stay mounted when switching between split-view tabs — no more reloading
• "Compacting conversation" status now shows in the chat instead of appearing stuck

What's New in v0.4.4

Threads now show their names immediately

Previously, new threads could appear as unnamed placeholders in the sidebar — especially when switching between sessions quickly. Thread names now appear right away and stay in sync, so you always know which conversation is which.

Terminal text renders correctly on first load

Fixed an issue where terminal text could appear misaligned or incorrectly sized when first opening a session. Terminals now wait for fonts to fully load before rendering, so text looks crisp from the start.

Cleaner sidebar

Removed the redundant settings icon from project headers to reduce visual clutter. The sidebar feels a bit tidier now.

What's New in v0.4.3

Codex Integration

• Sign in to CodexLog into your Codex account directly from Settings. Your browser opens, login completes automatically, and your account status appears in the app
• MCP Server StatusSee which MCP servers are connected to Codex with a live status indicator
• Approval RulesSet persistent approval rules so Codex remembers which tools you've allowed
• Rate LimitsView your Codex rate limits and usage at a glance
• Collaboration ModesSwitch between autonomous, semi-autonomous, and manual modes from the session UI

Smarter Sessions

• Session State TrackingSessions now show clear status indicators (running, processing, waiting for input, stopped) so you always know what's happening
• Auto Thread NamingNew sessions are automatically named based on your first message
• Processing & Unread BadgesThe sidebar shows live processing dots and unread message counts

Redesigned Settings

• Better organizationSettings are now grouped into logical categories: Accounts, Appearance, Typography, Editor, Terminal, Models, MCP Servers, Notifications, Shortcuts, and About
• Dedicated Accounts pageCodex login and Git accounts are in their own section

Approval Handling

• Faster promptsApproval banners appear with smart debouncing to avoid false flashes
• Subagent approvalsWhen Claude dispatches subagents that need approval, the prompt surfaces in the parent session
• Keyboard shortcutsPress Enter or a number key to quickly respond to approval prompts

Terminal Improvements

• Faster startupTerminals now appear in ~500ms (down from ~3 seconds)
• Image paste & dropPaste or drag-and-drop images directly into terminal sessions
• Scroll preservationTerminal scroll position is preserved when switching between sessions
• Font selectionChoose your preferred terminal font in Settings

Split View Polish

• Cleaner toolbarBranch name and IDE launcher are hidden in split mode to save space
• Compact controlsCommit button and view mode switcher go icon-only in split panes

MCP Server Management

• Add & remove serversManage MCP server connections from a dedicated Settings panel
• Live connection statusSee the state of each configured server

What's New Dialog

• Now automaticRelease notes are fetched live when you update, so you'll always see what changed

Fixes

• Corrected model size displayLocal AI model download now shows the accurate size (~1.1 GB)
• Cleaner chat messagesSystem tags are filtered from the conversation view
• File ExplorerHidden files are no longer filtered out
• Usage panelTime formatting now uses proper rounding and labels

What's New in v0.4.2

Dynamic Release Notes

• What's New dialog is now fully automaticversion and release notes are fetched from GitHub at runtime, no more hardcoded content to update each release
• Fallback handlingif the GitHub API is unreachable, shows a generic "Bug fixes and improvements" message

Fixes

• What's New dialog was stuck on v0.3.7the dialog never showed for v0.4.0 or v0.4.1 because the version was hardcoded and never updated
• CSP updatedadded `api.github.com` to connect-src for release notes fetching
• Process kill and App.tsx improvementsmisc local fixes included

What's New in v0.4.1

Codex Login Now Works

• OAuth login flow fixedclicking "Log in to Codex" in Settings now correctly opens your browser, detects when login completes, and shows your account status
• Authentication check before sessionsXanom now verifies you're logged in before starting a new Codex session, with a clear error message if not

Redesigned Settings

• Better organizationSettings are now grouped into logical categories: Accounts, Appearance, Typography, Editor, Terminal, Models, MCP Servers, Notifications, Shortcuts, and About
• Dedicated Accounts pageCodex login status and Git accounts are in their own section for easy access

Split View Polish

• Cleaner toolbar in split modebranch name and IDE launcher are hidden to save space
• Icon-only controlscommit button and view mode switcher go compact in split panes

Fixes

• Corrected model size displaylocal AI model download prompt now shows the accurate size (~1.1 GB instead of ~230 MB)
• Cleaner chat messagesIDE-related system tags are now filtered from the conversation view

What's New in v0.4.0

This is a major update focused on Codex integration, smarter session management, and a more responsive approval workflow.

---

Codex Integration

• Codex Account & LoginSign into your Codex account directly from Settings. Your login state and account info are displayed in the General settings page.
• MCP Server StatusSee which MCP servers are connected to Codex, with a refresh button to check live status.
• Approval RulesConfigure persistent approval rules so Codex remembers which tools you've allowed, reducing repeated prompts.
• Rate LimitsView your Codex account rate limits and current usage at a glance.
• Collaboration ModesSwitch between Codex collaboration modes (autonomous, semi-autonomous, manual) from the session UI.
• New JSON-RPC MethodsBackend support for fork, compact, steer, and collaboration mode operations.

Smarter Session Lifecycle

• Session State MachineSessions now follow a well-defined lifecycle (idle → spawning → running → processing → awaiting approval → stopped) with smooth UI transitions and no invalid state flickers.
• Auto Thread NamingNew sessions are automatically named based on your first message — no more "Untitled Session" clutter.
• Processing & Unread IndicatorsThe sidebar shows live processing dots and unread message badges so you always know which sessions need attention.

Approval Handling

• Faster Approval DisplayApproval prompts appear with a 2-second debounce to avoid false flashes on fast tool runs.
• Subagent Approval SynthesisWhen Claude dispatches subagents that need approval, the prompt is surfaced to you in the parent session.
• Instant Rejection ClearingDenying a tool clears the approval banner immediately.
• Keyboard ShortcutsPress Enter or a number key to quickly respond to approval prompts.

Terminal Improvements

• Faster Terminal StartupTerminals now appear in ~500ms (down from ~3 seconds) with smarter idle detection.
• Image Paste & DropPaste or drag-and-drop images directly into terminal sessions.
• Scroll PreservationTerminal scroll position is preserved when switching between sessions.
• Font SelectionChoose your preferred terminal font in Settings, including Hack font support.
• Post-Reveal Alt-Screen DetectionBetter handling of full-screen terminal apps (vim, htop, etc.) that launch after session start.

MCP Server Management

• Add & Remove MCP ServersManage MCP server connections from Settings with a dedicated configuration panel.
• Live Server StatusSee connection state for each configured MCP server.

Settings & UI

• Notification Sound SettingToggle notification sounds on or off.
• Default Model ConfigurationSet your preferred default model in Settings.
• Expanded Settings DialogReorganized settings with dedicated sections for MCP, fonts, notifications, and model preferences.
• Agent Type LabelsTool use blocks now show the agent type (e.g., "Dispatched Explore Agent") for better context.
• Usage Panel FixTime formatting now uses proper rounding and singular/plural labels.
• File ExplorerHidden files are no longer filtered out; drag-and-drop disabled to prevent accidental moves.

---

Full Changelog: https://github.com/neel-xanom/xanom/compare/v0.3.8...v0.4.0

What's New in v0.3.8

Terminal in Agents Tab

• New "New Terminal" option in the project `+` dropdown — creates a terminal session directly in the agents list alongside Claude and Codex sessions
• Terminal sessions include the full Warp-style input bar with git info, cwd tracking, and alt-screen detection
• Right-click context menu to close terminal sessions

Split View Improvements

• Split view toggle in sidebar header — quick toggle without opening Settings
• Fixed duplicate tabssplitting a tab now moves it to the new pane instead of duplicating
• Unsplit buttoncollapse back to single pane when split (preserves focused pane's tabs)
• Terminal sessions fully supported in split view mode

Other

• Terminal dot color (amber) in split view tab bar
• Terminal entity deduplication in split view store

See the assets below to download and install this version.

What's New in v0.3.6

✨ New Features

• UI polish — refined themes, updated settings, and component cleanup

🐛 Bug Fixes

• Force dark mode regardless of system settings
• Rewrite updater URLs to point to public xanom-releases repo
• Resolve terminal stuck on "Loading terminal engine" in production builds

🔧 Improvements

• Parallelize CI builds for aarch64 and x86_64

See the assets below to download and install this version.

See the assets below to download and install this version.

See the assets below to download and install this version.

See the assets below to download and install this version.

What's New in v0.3.1

✨ New Features

• Warp-style terminal input with CWD display, git branch info, and inline directory explorer
• Groq LLM provider support (llama-3.3-70b via Groq API/proxy)
• Context Ring — live token usage and cost visualization for Claude sessions
• Directory Explorer for browsing and navigating files directly from terminal
• Desktop notifications when background tasks complete
• Content Security Policy enabled for webview security hardening

🔧 Improvements

• OpenRouter provider refinements
• Updater endpoint moved to dedicated releases repository
• Release workflow configuration updates
• Various component and store refinements across frontend and backend

Updater artifacts for Xanom v0.2.0