Multi-Provider Support
- Desktop app
- All plans
What it does
Every agent in AgentsRoom is a real third-party coding CLI running in its own terminal, and each agent picks its own CLI: Claude Code (Anthropic), Codex CLI (OpenAI), GitHub Copilot CLI, Antigravity CLI (Google), OpenCode, Aider, Grok Build (xAI), Mistral Vibe, Kimi Code (Moonshot AI), Amp (Sourcegraph), oh-my-pi, Freebuff (Codebuff), Devin (Cognition), Cursor (Anysphere), Command Code, Muse Code (Meta), Pi (Earendil), MiniMax Code (MiniMax) and ZCode (community client), 19 providers in all. People who know the consumer brand rather than the CLI: ChatGPT is reached through Codex CLI (sign in with your ChatGPT account), Gemini through Antigravity CLI, Claude through Claude Code, Grok through Grok Build. Provider lists (pickers, Settings tabs, the lightning menu, the phone) are ordered by how many agents the community launched on each CLI over the last 30 days, Claude always first; a CLI with no launches yet, such as the two added on 2026-10-20 or Pi (2026-10-22), sits after the ranked ones. A project can run a Claude agent, a Codex agent and a Grok agent side by side on the same working tree. You can switch an agent's provider or model mid-conversation and keep the context, tune the launch flags and environment per provider or per agent, and point a CLI at a self-hosted or OpenAI-compatible backend without leaving the app. Video (1 min 20 s): Multi-Provider, shown in the real app. A Claude Code agent adds a helper and stops before its test; the same agent is switched to Codex with Agent summary, Claude writes its handoff, and Codex suggests the test, writes it and runs it.
Where to find it
- Per provider: Settings > AI providers (since 2026-10-21 the page opens on Essentials > Default provider, a picker; picking one also opens its tab). Then one tab per provider: the tab of the default provider carries a Default badge; every other tab starts with Use as default for new agents. Below it, in order: Permission mode, the provider's essentials (context window for Claude and Codex, Default reasoning effort, Codex Fast mode), the Accounts line (how many accounts, Manage accounts opens Settings > Multiple accounts; or "{provider} has no account switching" for a CLI without it), the {provider} CLI settings line (for Claude Code, Codex, Antigravity and oh-my-pi: its Open button unfolds Advanced settings and scrolls to the CLI's own settings), then a folded Advanced settings card: Claude renderer, CLI options, Environment variables, Proxy and last CLI settings, the provider's own options. That list loads from the CLI and shows folded (its first options, then Show all options); it refolds every time Settings reopen, and its own search box (from eight options) filters the whole list. With Claude accounts added in AgentsRoom, a change in CLI settings is written into every one of those accounts (both its settings.json and its .claude.json options), so all agents behave the same whatever their account; your own
~/.claudesetup, used outside AgentsRoom, is left untouched as soon as one account exists. The accounts and the rules shared by every account (auto-switch, pinned quotas) live on Settings > Multiple accounts; the rules true for every agent (MCP servers, approvals, agent mail) on Settings > Agent behavior. A footer button next to the Settings gear opens Settings > AI providers directly (hideable in Settings > Interface elements). - Per project: Project settings > Defaults > AI providers (the CLI, the account and the provider options this project's agents run on). Sidebar folders carry the same defaults for every project filed under them.
- Per agent: the Provider and Model pickers in the Add agent / Edit agent dialogs, plus the CLI options and Environment variables fields in their advanced block. In the Add agent window they are cards under the agent's avatar (Provider, Model, Reasoning, Workspace), each naming what it holds ("Claude", "Opus 4.5 · 1M context"); a click opens its panel right below. Install a CLI, in the Provider panel and in the Not installed footer of every provider picker, opens the guided installer without leaving the dialog.
- Per project: Project settings > AI providers > Environment variables, handed to every agent of that repository whatever CLI it runs on. The same section has one Default reasoning effort row per CLI that has an effort lever.
- Mid-session: the provider / model pill in the agent's header, right of its role chip (it reads the model, "opus 1M" or "gpt-6-luna"; tooltip Switch model), opens the Switch Provider window, laid out like Add agent: the agent with its current and new provider and model, the Provider, Model and Reasoning cards (the account is picked in the Provider panel and shown under the card), the context transfer modes as cards (Context transfer: Agent summary (recommended), Raw transcript, Light context, Start fresh; they show once the Provider or Model panel is folded again by clicking its card), and ⌘/Ctrl + Enter to confirm. Its header carries the video tutorial icon, which opens the Multi-Provider video (not when the same window opens as Restart Agent). The confirm button names what will happen: Summarize & Switch, Transfer & Switch, Switch & Restart or Switch and start fresh. The Usage panel (footer usage badge) also has a Temporary launch override that forces every new agent onto one provider and model: for 2 hours, until tonight, 7 days, a custom number of hours or days, or Until I stop it.
- Quick launch: click the Quick agent lightning button. Since 1.184.0 a plain click opens the provider and model menu (right-click, long press and keyboard open it too); picking a model launches a throwaway agent on it. The menu has three sections. Agent tiers comes first: Fast agent, Powerful agent and Expensive agent, each with an icon and a line naming its provider, model and reasoning effort ("Claude · Sonnet · High"); a click launches it, a right-click opens its settings (icon, provider, model, reasoning effort, saved for All projects or for this project only). Until you customise them, the three derive from your preferred provider (for Claude: Sonnet, Opus and Fable; for Codex: Luna, Sol and Sol Max). A derived tier skips the models a gateway serves for free (ids ending in
-freeor:free, such as OpenCode'sopencode/exo-free), because those promos can be retired overnight and every agent launched on them would fail; it falls back on them only when nothing else is listed. A free model you pick by hand or set on a tier yourself still works. Then Providers, the list below. Then the settings rows: One-click launch ("Click spawns the default agent, no menu"), Add or update providers (opens the Setup assistant as a window above the app since 1.187.0, so the project and its agents stay open behind it while a CLI installs) and Providers leaderboard. With One-click launch ticked, the click launches the default quick agent right away, Edit default quick agent appears under it, and only the right-click opens the menu. At the very bottom, when another of your computers holds the project, Run on picks the computer the quick agent starts on (This computer by default, for that launch only); see Remote Fleet.
How to use it
- Install the CLI you want (the setup banner in an agent terminal offers the install command if it is missing) and sign in to it once.
- In Settings > AI providers, pick the CLI in Default provider (or open its tab and click Use as default for new agents). The tab's Advanced settings card holds the flags and environment applied to every agent of that provider.
- When you create or edit an agent, choose its Provider; the Model picker adapts to that provider. Default leaves the choice to the CLI. For Cursor, OpenCode, Grok, Devin and GitHub Copilot the list is read from your own CLI, so it reflects what your plan actually reaches (image, speech, transcription, embedding and moderation models are left out since 2026-10-18: they cannot drive a coding agent); Refresh model list, at the foot of those pickers and as the last row of the launch menus, re-reads it without restarting the app and says what came back ("List updated: N models", "Already up to date: N models", or "Could not read the list from the CLI", which keeps the list you already had). Every picker opens with a Last used row (the models this machine launched last on that provider, most recent first, kept locally) above All models; a recent model is listed twice on purpose, and a fresh install shows no row. Next to the models, the community's habits: a flame marks the few most launched models of each provider (Popular in the community), a tile says "N users" and its tooltip "N users launched it M times in the last 30 days."; provider tiles carry "N users this month". These figures come from a public statistic read at startup and refreshed every 30 minutes, so they show whether or not you share your own usage data; offline, they are simply absent, never "0". Since 1.182.0 the OpenCode list also refreshes by itself when you run
/connectinside the OpenCode TUI or declare a provider by hand in its config, with no restart of AgentsRoom: an open picker updates in place. Devin, Grok and Cursor list what the account plan includes, nothing local changes that list. - To change provider or model on a running agent, click the pill in the terminal header. Same provider and a CLI with a live model command (Claude, Codex, Grok, Mistral Vibe): the model switches in place. Any other change: the agent is stopped, a handoff (modified files, recent activity, last message, optionally a summary written by the agent) is built, and the new CLI starts with that handoff as its first prompt. To skip the handoff entirely, pick Start fresh under Context transfer: the new CLI opens a blank conversation with no handoff message. It is preselected when the session is empty (just started, or emptied with
/clear), since there is nothing to hand over. With Agent summary (recommended) and Summarize & Switch, a Summarizing context… bar sits above the composer while the current agent writes its handoff (Current Task, Completed Work, Remaining Work, Key Decisions, Gotchas; Switch anyway skips the wait); then the same agent, same project, restarts on the new CLI with that handoff as its first message, and the new CLI asks you what to do next.
Per-agent flags: the agent's CLI options field replaces the provider-wide flags for that agent only. It is the escape hatch to pin a model the picker does not list (for example --model opus-4.8). When your flags already pin a model, AgentsRoom adds none of its own.
Environment variables: one KEY=value per line, on three levels merged key by key: machine-wide per provider (Settings > AI providers), then the whole project (Project settings > AI providers), then the single agent, each winning over the one before it. The project level is the one to use for something the whole repository needs, a gateway or a token every agent must see, without repeating it on each agent or forcing it on your other projects. This is also the only way to point a CLI at a model backend rather than another CLI: an OpenAI-compatible gateway (OpenRouter, LiteLLM, OmniRoute, a company proxy, the OpenCode Go endpoint) is wired with OPENAI_BASE_URL or ANTHROPIC_BASE_URL. A value can reference a secret from the Secret Manager as {{secret:NAME}}, resolved when the terminal is created so the token never appears in a prompt or transcript. Variables the app owns (PATH, CLAUDE_CONFIG_DIR, CODEX_HOME, AGENTSROOM_*) are refused.
Pinned models: a CLI whose catalogue is discovered from the account can list hundreds of ids (Cursor, OpenCode, Ollama). From twelve models up, each tile in the picker gets a Pin badge on hover; pin the ones you actually use and the picker shows only those plus your current selection, the rest folding behind All models (N) with its search box. The order stays the catalogue's, never the order you pinned in, and the model currently selected is always visible. Fewer than twelve models and nothing changes. Since 2026-11-12 Antigravity reads its list from agy models too: the models your Google account reaches (Gemini 4 Carbon Low / Medium / High, for example) are added after the built-in ones, the default model stays the same, and signed out of agy you keep the built-in list. Model cards have the same size whatever the provider. The launch menus (lightning button, agent card, backlog ticket) carry a Pin button on every row, whatever the length of the list: shown on hover, always visible once pinned, and pinning never launches the model nor closes the menu. Pinned models get a Pinned section at the top of the menu; Last used heads the menu only while nothing is pinned; All models follows.
Sort and filter: every model list, launch menus and dialog pickers alike, has three order buttons: Power (most capable first, the default; between two models of equal strength the newest generation comes first, so GPT-6 Sol ranks before GPT-5.6 Sol), Price (cheapest first) and Popular (most launched by the community first; while the community figures are missing, offline for example, the catalogue order is kept). Each button carries an arrow showing the direction; clicking the active button again reverses it (weakest first, most expensive first, least launched first), and picking another order opens it in its natural direction. The choice is remembered and applies to every list. In the dialog pickers the buttons show on lists of seven models or more, or as soon as the list mixes several subscriptions. When one CLI serves several subscriptions (today OpenCode: OpenCode Go, Z.AI Coding Plan, Z.AI, or a backend you connected yourself), a second row of chips, All, one chip per subscription and Other for models with no subscription prefix, narrows the list to one wallet before you pick the model. That filter is not remembered, and it never shows on a provider with a single subscription.
Settings
defaultProvider(global, Settings > AI providers): provider pre-selected for new agents, defaultclaude. Overridable per project in Project settings > Defaults. Until you edit them in Settings > Default profiles, the Quick Agent (lightning button, Invoke an agent, the project assistant) and the QA Agent also start on it, with that CLI's own default model.providerCliOptions(global): extra CLI flags per provider id; empty = the provider's autonomous defaults.claudeCliOptionsis the legacy Claude field, still honoured.providerEnvVars(global): environment block per provider id,KEY=valuelines,{{secret:NAME}}allowed.envVars(project override, Project settings > AI providers): oneKEY=valueblock handed to every agent of the project, whatever its CLI. Wins overproviderEnvVars, loses to the agent's own block.pinnedModels(global): the models kept at the top of every picker, per provider. Empty means no shortlist and the full list is shown.reasoningEffortDefaults(global, Settings > AI providers > provider tab > Default reasoning effort; project override in Project settings > AI providers): the effort given to every agent of that provider that pins none, per provider id. Unset = the CLI keeps its own default; on a project, Default follows the global value. An agent's own effort, a ticket's effort or a launch override always win. Only the tiers the launched model offers apply, otherwise the CLI's default is used.hiddenModels(global, synced with the account: Settings > AI providers > provider tab > Models available to agents; project override in Project settings > AI providers, provider picker then the same chips, Use the global list drops the override): per provider id, the models agents may NOT use. A switched-off model disappears from every model picker and launch menu (desktop and phone) and is never launched, whatever pins it: a saved agent, a team step, a scheduled trigger, a ticket, an agent started by another agent (agents_spawn, whose reply reportsmodel-hidden) or a ticket started from the phone all launch on the first model still allowed (in picker order, never the CLI's own default, which can be the switched-off model itself). An agent or ticket left on Default is covered too when the CLI's own default is known: Antigravity (agy) runs Gemini 3.8 Flash (High) without a model, so with that model switched off an unpinned Antigravity launch starts on the first model still allowed (for example Gemini 4 Argon). A model id without its effort tier (gemini-3.8-flash, typed by an agent over MCP) counts as switched off when one of its tiers is, and prefers a tier still allowed. Other CLIs, whose default depends on the account, keep their own default. The saved choice is not rewritten: switch the model on again and the agent goes back to it. A console already running is not restarted. The agent editor and the team editor flag such a stale pin and name the model the next launch uses instead. The last model of a provider cannot be switched off. A/modeltyped inside the CLI itself is not blocked. The phone filters its pickers but has no screen to edit the list. MCP:settings_set hiddenModels(global and project). Related,defaultModels(global, synced with the account: Settings > AI providers > provider tab > Default model; project override in Project settings > AI providers, provider picker then Default model, synced too: choosing Default there follows the global setting again): per provider id, the model an agent, a team step, a scheduled trigger, a ticket or anagents_spawnleft on Default launches on. Empty (the default) keeps today's behaviour: the CLI picks its own model. The saved agent stays on Default, so clearing the setting gives it the CLI's choice back; a model pinned on the agent or the ticket always wins. A default model that is switched off in Models available to agents launches the first model still allowed. Useful when a CLI's own default is not the model you want (Antigravity runs Gemini 3.8 Flash without a model). MCP:settings_set defaultModels(for example{"antigravity": "gemini-4-argon-medium"}). MCP project scope:settings_set scope=project defaultModels. The phone has no screen to edit it, but agents and tickets started from the phone launch through the desktop and use it.modelSortOrder(global): order of every model list,strength(Power, default),price(Price) orpopular(Popular).modelEquivalenceRef(global, synced with your account; the Equivalents in selector under the order buttons of every model list, or a click on any "≈ Opus" chip; a searchable list with the providers' logos, limited to the CLIs installed on this computer and never the provider you are looking at): the provider the equivalences are written in.null= Auto (the provider you run the most agents on, never the one you are looking at, so a Claude + Codex user reads Codex models in Claude terms; on your default provider Auto shows no comparison, since its models are already your yardstick), a provider id = that provider (applies even on your default provider),off= No comparison.modelSortReversed(global):trueflips the direction ofmodelSortOrder(the arrow next to the order button); reset tofalsewhen the order changes.recentModels(global): the Last used row of every picker, per provider parent id, 5 model ids at most, rewritten at every launch. Machine-local, never sent anywhere.agentTiers(global, synced with your account, overridable per project; right-click a tier in the lightning menu, or Settings > Defaults for the account and Project settings > Agents for a project override): the Fast agent, Powerful agent and Expensive agent shortcuts,{ fast?, powerful?, expensive? }, each{ provider, model, effort?, icon? }withicona lucide icon name. A tier you never customised is absent and derives from the preferred provider (Fast = its standard model, Powerful = the complex one, Expensive = the flagship). A project override replaces a tier as a whole, never field by field. The phone's quick agent sheet shows the same three tiers and launches them through this desktop; they are configured on the desktop only.quickAgentClickLaunchesDefault(global, ticked as One-click launch in the lightning menu):true= a plain click on the lightning button spawns the default quick agent with no menu. Off by default; the right-click always opens the menu.agentPermissionMode(global): per provider,autonomous(default: the CLI runs every action without asking) ormanual(the CLI asks before risky actions). In the Non-dev mode the first-launch question is not asked: assistants stay autonomous unless Manual is chosen here. Applies to new sessions.preAcceptBypassWarning(global, synced with your account; Settings > AI providers > Claude tab, under Autonomous mode when it is selected: Accept the Bypass warning for me):true(default) = at an autonomous launch, AgentsRoom accepts Claude Code's "WARNING: Claude Code running in Bypass Permissions mode" for you before the CLI starts, so a first launch never stops on it.false= Claude asks again at its next launch and you answer in the terminal (arrow to Yes, I accept, then Enter). Nothing is written for an agent in Manual mode, for an SSH launch, or once Claude already recorded your consent. Applies to the next launch, desktop and phone alike.skipClaudeEmailCheck(global, synced with your account; Settings > AI providers > Claude tab, under Autonomous mode when it is selected: Skip Claude's account email check, since 2026-11-06):true(default) = at an autonomous launch, AgentsRoom turns off Claude Code's account email check for that session only (a--settingslayer on the launch line; nothing is written to your Claude settings, your other plugins stay loaded), so an unattended agent never stops on "This call's input holds your account email address".false= Claude asks again and the agent waits for your answer. Not applied to an agent in Manual mode, to a restricted run (a webhook trigger acting on an outside payload keeps the check), or when the agent's CLI options already pass their own--settings(Claude reads only one). Applies to the next launch, desktop and phone alike.codexFastMode(global, overridable per project and per agent):inherit,onorofffor Codex Fast mode.claudeRenderer(global, overridable per agent):classicscrollable pane orfullscreenalternate-screen TUI for Claude Code.cliPresenceOverrides(global): your "I already have it" answers to the CLI-missing banner, per provider.lastLaunchOverride(global): the last temporary launch override, pre-filled next time, never re-applied by itself. Itsdurationis2h,today,week,untilStopped(no end) orcustom; a custom duration is stored in hours (customHours),customUnit(hoursordays) only says how the field shows it.
Agent tools (MCP)
settings_get/settings_set: read or change the keys above (scope: "global", orscope: "project"for the project override of the default provider and Codex Fast mode, orscope: "agent"for one saved agent's provider, model, CLI options and environment).agents_save: create or update a saved agent with a givenproviderandmodel;agents_spawnandbacklog_create/backlog_updateaccept the same executor overrides (provider,model,reasoningEffort) for a throwaway agent or a ticket.agents_spawnalso takesaccountProfileIdsince 1.184.0 (an id from the provider's profiles setting, or"system"for the CLI's own login; an id of another provider is dropped and reported). The confirmation card shown in the requesting agent's terminal has a Runs on block (Provider, Model, Account) pre-filled with what the agent asked for: you can change all three before the agent is created, "for a review by another model or on another account for example", and the answer reports what it actually runs on.capabilities_get(topic: "teams"): lists the provider ids a team node may use.
Providers
- CLI settings of oh-my-pi: Settings > AI providers > oh-my-pi > CLI settings edits its own configuration with no YAML to touch: skills on or off, ignored skills, memory backend, model roles, the servers of its MCP file, and every other setting, sorted by section with a search field. Model roles are text fields with your models as suggestions, so anything omp accepts can be written (
provider/model:highfor a thinking level,@smolto reuse another role), for every role including image and speech. Lists and records are edited as JSON in a full-width field. An option with a fixed list of values keeps its menu, because omp refuses anything else; its Default entry, or emptying a field, brings back omp's default. omp validates each value, its reason appears under the row it refuses; a copy of the file is kept next to it before the first change. The setting's name and explanation follow the language of the app: they come from omp in English and are machine translated on display, with omp's real key kept under each row. Signed out, over the monthly translation limit, or in a language the translation service does not cover (Dutch, Thai, Ukrainian), they stay in English. Tokens and API keys are never shown: most are set as environment variables under Advanced settings > Environment variables of the same tab. - Live model switch (sent into the running session, no restart): Claude, GitHub Copilot, Grok Build and Devin with
/model, oh-my-pi with/switch(for this session only, since 2026-10-22: its/modeltook no model name and the request reached the agent as a message). Kimi opens its own model picker pre-selected on the requested alias. Codex, Mistral Vibe and Cursor only open a selector on/model, so AgentsRoom restarts them on the new model, like everyone else, with a handoff (Mistral Vibe since 2026-10-22: the selector stayed open and swallowed the next message). Claude's/modelalso saves the model as the default of the whole Claude account, and Kimi's once confirmed: in a team run, a step that changes model on those two restarts its console instead, keeping the conversation. - Session resume after a restart: Claude, Codex, Grok Build, Antigravity, Cursor, Kimi Code, oh-my-pi, Pi (since 2026-10-22) and, since 2026-10-21, Command Code and Muse Code, and since 2026-11-06 MiniMax Code and ZCode, reopen the same conversation; Copilot, OpenCode, Aider, Mistral Vibe, Amp, Freebuff and Devin start a new one. Command Code saves its conversation at its first answer and Muse Code at launch: an agent closed before that has nothing to reopen. If the saved conversation is gone, the agent simply starts a new one. oh-my-pi (since 2026-09-25) keeps its conversation in its own sessions folder, so it also shows in
omp statsand inomp --resumefrom your own terminal; a session started in another folder (the agent moved to a worktree) is not resumed, and a/newtyped inside oh-my-pi is not followed (a restart reopens the conversation from before it). - Readable transcript (the Conversation view, the phone's reading mode, Read Aloud of the last answer): Claude, Codex, Grok Build, Mistral Vibe (since 2026-09-15), oh-my-pi (since 2026-09-25), Antigravity (since 2026-09-30) and Pi (since 2026-10-22). The other CLIs only expose the messages you sent.
- Full-screen interfaces: Codex is always launched in its classic inline rendering (since 2026-09-26: Codex 0.157 switched to a full-screen transcript by default, which hid the console from the status detection and from the phone, and put a "FULLSCREEN MODE" banner over the top of the console). Grok Build and Mistral Vibe always paint their own screen, so a long turn shows nothing in the terminal until it ends. Grok can be relaunched in a plain-scrollback mode; Mistral Vibe has none, and its progress is followed in the Conversation view instead (Conversation View).
- System prompt: passed as a flag for Claude, Codex, oh-my-pi, Pi; MiniMax Code takes it at each launch and ZCode as its first message; the others read it from project files (
AGENTS.md,CLAUDE.md,.cursor/rules). - Reasoning effort: exposed for the CLIs that have a lever (Claude, Codex, Grok, Kimi, oh-my-pi, Pi and, since 2026-09-26, Antigravity among them); it is a launch-time setting, so changing it restarts the agent. On oh-my-pi it is its
--thinkinglevel, from Low to Max; a level the model does not offer is lowered by oh-my-pi to the closest one it has, and a--thinkingyou put in the agent's CLI options wins. On Codex the tiers follow the model: GPT-6 Astra and GPT-6 Sol go up to Ultra, GPT-6 Luna stops at Max. On Antigravity the tiers are Low, Medium, High and Max (no Very high); a tier Antigravity does not know, such as a Very high inherited from a ticket set up for another CLI, is left out so the agent still starts, and an--effortin the agent's CLI options wins. Since 2026-09-28 each of these CLIs can have a Default reasoning effort, machine-wide in its provider tab or per project: every agent that has no effort of its own starts on it, and the Default chip of the Add / Edit agent dialogs shows the inherited tier. Changing the default moves every agent that inherits it at its next launch; it is never written onto the agents. - Autonomous mode flags exist for Claude, Codex, Antigravity, Grok, Mistral Vibe, Kimi Code, OpenCode, Amp, oh-my-pi, Copilot, Devin, Cursor, Command Code, Muse Code and ZCode; MiniMax Code has no flag, so AgentsRoom writes the mode into its config file; Aider always auto-confirms; Freebuff has no permission flag; Pi never asks for a tool approval by design, so Manual changes nothing on it. OpenCode gets
--autoin Autonomous mode (it approves what you did not explicitly deny) and keeps its own prompts in Manual: it used to be left on its prompts in both modes, which stopped every unattended launch (backlog ticket, team step, scheduled run) on an approval nobody was there to give. The flag of each CLI: Agent Permissions (Autonomous or Manual). - Models: Amp and Freebuff choose the model server-side (single default entry); oh-my-pi and OpenCode accept any
backend/modelid typed by hand, in Add agent and in a backlog ticket alike; oh-my-pi lists the models of the accounts and API keys you connected with/logininside it (read fromomp models), and since 2026-09-26 its picker has one chip per account over the list (Anthropic, ChatGPT, Z.ai...) to keep only that account's models, on the desktop and the phone; Mistral Vibe offers Medium 3.5 first, then every model declared undermodelsin your own Vibe configuration (~/.vibe/config.toml, or$VIBE_HOME/config.toml), the same list as/modelinside Vibe (since 2026-10-08, it used to show Medium 3.5 only); a block added to that file shows up in the picker without restarting the app, the chosen alias is passed again at every relaunch (also for an agent started from the phone or a backlog ticket), and Default leaves Vibe on the model last chosen with/model; Cursor lists roughly 190 models from your plan at pick time; OpenCode also surfaces the flat-rate OpenCode Go models (opencode-go/...) and the Z.AI Coding Plan models (zai-coding-plan/...) as soon as you have run/connectinside OpenCode. The same model can then appear once per plan: the picker describes each entry by its plan name (OpenCode Go, Z.AI Coding Plan), and the session pill, the folded engine summary and a ticket's executor line readglm-5.3 · Z.AI Coding Plan; a backend AgentsRoom does not know keeps its raw name. Codex lists the GPT-6 family (GPT-6 Astra, GPT-6 Sol, GPT-6 Luna, since 1.187.0, and GPT-6.1 Sol since 2026-09-30, on desktop and phone) next to the older GPT-5.6 tiers. Since 2026-10-02 it also lists Daybreak Blue (defensive cybersecurity work) and Daybreak Red (authorized offensive security research), OpenAI's cybersecurity models, which need Daybreak access on your OpenAI account (an account without it gets the CLI's own refusal at launch); the chosen model is saved on the agent and passed again at every relaunch, so you do not re-run/model. Grok Build lists Grok 4.7 only since 2026-10-02 (the CLI and the server dropped Grok 4.6 and 4.5). Since 2026-11-11 a Codex agent still saved on a model Codex can no longer launch (GPT-5, GPT-5 mini, nano or pro, o3, o3-pro, o4-mini, codex-mini-latest, gpt-5-codex) starts on GPT-5.5 instead, so it keeps working after OpenAI retires those models on 2026-12-11; a model you force with--modelin your Codex CLI options is left alone. Since 2026-10-01 a new Codex agent is pre-set on GPT-6.1 Sol, the same default as the Codex CLI itself (it was GPT-5.6 Sol); agents you already created keep the model they were saved with. GitHub Copilot shows a curated list (Claude Sonnet 5, GPT-6 Astra, Claude Opus 5, Claude Haiku 5.5 since 2026-10-10, Claude Haiku 4.5, Auto among them) plus every other id the installed CLI accepts, read from that CLI, because GitHub adds models to the seat between two AgentsRoom releases. - OpenCode 2: the pinned model is no longer passed as a
--modelflag (OpenCode 2 refuses it on its root command) but through the CLI's own configuration, for that launch only; a--modelyou typed in the agent's CLI options is honoured the same way. Each agent also gets its own private server (--standalone) when the installed CLI supports the flag, so every agent sees its MCP servers, its status plugin and its model instead of only the first one started. A task given at launch is submitted by AgentsRoom once the TUI is up and the terminal quiet, because OpenCode 2 only pre-fills its input box with it. OpenCode 1 users see no difference. - Codex note: the AgentsRoom MCP tool calls Codex prints in its terminal (
AgentsRoom-MCP.memory_save({...}), sometimes 50 lines of JSON) are folded to(…)in the desktop terminal so the useful output stays readable. The full arguments remain in the Conversation view (Conversation View). Codex only, always on, no setting. - MiniMax Code and ZCode (since 2026-11-06) run natively, without going through OpenCode. ZCode is a community client (
zcode-app-cli), not published or endorsed by Z.ai. Install and sign-in: MiniMax Code with its install script or npm, thenmcode login --region global(mcode loginfor a mainland China account); ZCode withnpm install -g zcode-app-cli@latest(Node.js 22.19 or newer), then/loginin its interface or an API key. Neither has a reasoning effort lever, a quota gauge, a readable transcript or per-agent change attribution. MiniMax Code has no permission flag in its interface, so AgentsRoom writes the Permission mode into its own config file before launch: the mode then applies to that MiniMax account on this computer,mcodeoutside AgentsRoom included, and between agents the last launch wins. With Leave this project's .mcp.json untouched on, it has no AgentsRoom tools. ZCode gets its model as a/modeltyped before the first message, so an agent started from the phone stays on the account's default model. On an SSH project, neither gets a config written on the host.
Mobile
Present. The New agent sheet has a collapsible Engine block (provider, variants, model, 1M toggle, Reasoning effort, account), pre-selected on the desktop default provider. Since 2026-09-26 the reasoning effort is chosen there at creation, with the tiers the chosen model offers, and the collapsed header shows the tier next to the model. Its model picker has the same LAST USED and ALL MODELS sections as the desktop, with the flame and "{{count}} users" hints; the recent row merges the choices made from the phone with the launches recorded by the desktop for your account. The same Power / Price / Popular order buttons and subscription chips appear on lists of seven models or more or on a list that mixes subscriptions; the phone keeps its own order choice, separate from the desktop's. Pinned models are on the phone too: a PINNED section heads the list once something is pinned (and "Last used" steps aside while it does, like the desktop menus), a long-press on a pill pins it and the badge on a pinned pill removes the pin; pins and last-used models are saved to your account, so a pin taken on the laptop shows on the phone and the other way round. Long-press an agent, or tap the spec strip in the console header, to open the Switch Provider sheet. It is laid out like the New agent sheet and the desktop window: the agent at the top with its current provider and model and, once you pick another one, the new pair next to it; then the Provider, Model and Reasoning cards (provider with variants and account, model with version and 1M window), each opening its panel; then the context transfer as two cards, Capture context summary and Start fresh (neither selected = light context, the default); the switch is executed by the desktop through the encrypted relay, and the desktop console announces it first (Model switch requested from your phone: applying {provider} · {model}…) so a model change is never a surprise mid-run. The spec strip shows which CLI and model each agent runs on. Not on the phone: editing environment variables at any level and provider-wide CLI options (both desktop only). An agent launched from the phone still receives the project's environment variables, and the phone's model lists are short enough that there is nothing to fold. The one-time Agent permissions question (Autonomous or Manual, asked on the first automatic launch of a provider) also opens as a sheet on the phone since 2026-09-26: before that, an agent launched from the phone did not start until someone answered on the computer.
The Usage panel of the Activity tab has the Temporary launch override card of the computer you are connected to, since 2026-10-27: it shows the running override (provider, model, time left, account, scope) with Stop override, and a tap opens the same window as the desktop (provider and model, reasoning effort, account, scope, duration including Until I stop it and a custom number of hours or days) with Apply override. The phone acts on that computer only: each machine keeps its own active override.
The quick agent sheet (lightning button of the room) opens on the three Agent tiers (Fast, Powerful, Expensive): one tap launches the tier on your computer, with the model, effort and project override the desktop resolved. Tiers are configured on the desktop (right-click a tier); the phone shows an icon outside the three defaults as a lightning bolt. Since 2026-11-11 every provider picker of the phone (Switch provider, New agent, ticket, trigger, specialist) keeps the CLIs found on the computer (or on the project's SSH host) at the top and folds the others under Not installed, closed by default; a tap unfolds it. The current provider always stays at the top, and with an older desktop that sends no list, everything stays at the top as before.
Limits
- Every provider is available on every plan, Free included: there is no provider paywall and no Pro badge in the picker.
- AgentsRoom never bundles a CLI: each provider must be installed and signed in on the machine that runs the agents. A CLI installed after AgentsRoom started may need an app restart to be detected on Windows.
- A stale model id costs more than a missing one: on Grok an unknown id does not fail, the CLI runs unpinned and drops the reasoning effort. The pickers are refreshed from the CLI for the providers that can list their models; only OpenCode is re-read on its own when its credentials change. On OpenCode 2, where
/connectwrites to OpenCode's own database, that automatic refresh needs thesqlite3command on the machine (usually absent on Windows): without it, click Refresh model list after a/connect. - Folder trust: Kimi Code and Antigravity remember a trusted folder by its exact path and have no flag to skip their "trust this folder" question, so trusting the project did not cover a ticket's new worktree and the agent stopped on that question. Since 2026-09-26 launching one of them from AgentsRoom counts as the trust decision: the launch folder is recorded as trusted in that CLI's own settings before it starts, on the desktop and from the phone alike. If that recording fails, the CLI simply asks as before.
- Claude Code email check (Claude Code 2.1.317 and later): before a shell or web tool call whose text contains the email address of the Claude account the agent is signed in to, and that may reach the network (
curl,wget, anhttp(s)://address,ssh,scp,rsync...), Claude Code asks "Do you want to proceed?" under "Hook tool.check requires confirmation for this command: This call's input holds your account email address... [plugin:cc-plugin-privacy]". Claude's own bypass flag does not skip it, so since 2026-11-06 AgentsRoom turns that check off at every autonomous Claude launch (skipClaudeEmailCheck, on by default). Where it still shows (Manual mode, a restricted run, the setting switched off, a console started before the update), AgentsRoom shows it like any permission prompt (Needs you, Allow / Deny from the phone's lock screen) and never answers it for you. - Amp runs a task given at launch in one-shot mode and drops back to the shell when it finishes; Freebuff and Kimi receive the task typed into their TUI after start. An agent of these two started from the phone begins without its task.
- GitHub Copilot CLI (since 2026-10-08): the agent's role, persona and AgentsRoom briefing are loaded as Copilot custom instructions when the console starts, so they cost no extra request and stay in place for the whole session; a task given at launch opens an interactive session that stays open after the first answer. AgentsRoom also lets Copilot read its own role and briefing folders without an approval prompt. If you set
COPILOT_CUSTOM_INSTRUCTIONS_DIRSyourself, your folders are still read, after the agent's own. - Changing provider on a live agent is a restart with a handoff, not a shared memory: a long conversation arrives as a summary. Agent summary needs quota left on the current CLI (the current agent writes the summary); out of credits, use Raw transcript or Light context. The model picked in the window is the catalogue's: on a Codex account whose plan does not include it (GPT-6.1 Sol on a free ChatGPT plan), the agent starts on the closest model the account lists (GPT-6 Luna), so pick that model yourself to see in the window what will run.
- MCP tools reach every provider except Aider, which has no MCP support.
Common questions
- Claude stops on "Bypass Permissions mode" with "1. No, exit / 2. Yes, I accept", or closes right after my first message. Why? Claude Code shows that warning the first time it runs in autonomous mode, and "No, exit" is the selected row, so an Enter quits. Since 2026-10-31 AgentsRoom accepts it for you before the launch (Settings > AI providers > Claude > Accept the Bypass warning for me, on by default). If you switched that off, answer in the terminal: arrow down to Yes, I accept, then Enter.
- My agent stopped on "Hook tool.check requires confirmation... This call's input holds your account email address". How do I make it go through on its own? The question comes from Claude Code itself (2.1.317 and later), not from AgentsRoom, and Claude's bypass flag does not skip it. Since 2026-11-06 AgentsRoom turns that check off at every autonomous Claude launch: update AgentsRoom and restart the agent (a console started before keeps the check until it restarts). Check that Settings > AI providers > Claude tab > Skip Claude's account email check is on, and that the agent is in Autonomous mode: in Manual mode you are there to answer, so the check stays. To keep the check everywhere, switch that setting off. A test or demo account that signs in with your own address is still better kept in the vault (Commands & connections > Secrets) and handed to the agent as an environment variable (Project settings > AI providers > Environment variables,
DEMO_EMAIL={{secret:NAME}}): the password then never shows in the console. - My agents left on Default always start on a model I do not want (Antigravity on Gemini 3.8 Flash). Can I choose the model they use? Yes. Settings > AI providers > the provider's tab > Default model: pick the model. Every agent, team step, trigger or ticket of that provider left on Default now launches on it, on every project (a project can choose its own in Project settings > AI providers). A model pinned on the agent or the ticket still wins. Choose Default again to let the CLI pick (details: Default Model per CLI). Switching a model off in Models available to agents also keeps Default launches off it. Can I stop my agents from ever using a model I do not want? Yes. Settings > AI providers > the provider's tab > Models available to agents: switch off every model you do not want. They disappear from every model picker on the computer and the phone, and nothing launches on them any more: an agent, a team step or a scheduled trigger that was set to one of them before starts on the first model still allowed, and so does an agent asked to start another agent on one. A project can have its own list (Project settings > AI providers).
- I switched a model off but a saved agent or a team still starts on it. Why? In earlier versions a model switched off only left the pickers: an agent, a team step or a scheduled trigger saved on it kept launching on it. Now it never launches: the next start uses the first model still allowed, and the agent editor (model picker) and the team editor (issue list) flag the stale choice and name the model used instead. Pick another model there to make it explicit; the saved choice comes back on its own if you switch the model on again. A console that was already running keeps its model until it restarts.
- The agent header says "Opus": which version is it? The header shows the version the CLI announced (for example Opus 5.5) when the agent is on the
opusalias, and the tooltip of that pill shows the reasoning effort the agent really started with. The Claude picker also offers pinned versions, Fable 5.2 among them since 2026-11-10 (desktop and phone). A pinned version that needs a newer CLI than the one installed is refused with a message naming the version to install ("Fable 5.2 needs a newer Claude CLI: version 2.1.290 or later"), instead of the header announcing a switch the CLI never made: update the CLI, then switch again. - A ticket assigned to a role (not to an agent) ran on Antigravity, or another CLI I did not pick. Why? Fixed on 2026-10-26 (reported by Robert): it used to take the CLI of the project's first agent. A ticket assigned to a role that no agent of the project covers now runs on the project's Default provider (Project settings > Defaults, or its folder's), else on the CLI all the project's saved agents share, else on your global Default provider; the automatic triage sizes its model for that same CLI. Same family, fixed on 2026-11-16 (reported by Allen): with Codex as your default provider, the project assistant, the Git Expert, a quick agent or the QA Bot could still open Claude. The Quick Agent and QA Agent profiles you never edited now start on your Default provider (the project's or folder's one when set), and the agents AgentsRoom adds on its own no longer copy the project assistant's CLI. An agent already created keeps its CLI: switch it from its settings. A profile you saved yourself in Settings > Default profiles is kept as is.
- How do I connect my Codex, Cursor or Claude account? Add an account has a crown. Sign in with the CLI's own login once (
codex login,cursor-agent login,claudethen/login) and every agent of that provider uses it, on every plan, Free included. Add an account is for extra accounts: one on Free and Plus, unlimited on Pro (the crown shows once the free one is used): see Multiple Accounts. - Why doesn't the model I added to Mistral Vibe show up? Since 2026-10-08 the picker reads the
modelsblocks of your Vibe configuration (~/.vibe/config.toml); a model declared only in a project's.vibe/config.tomlis not listed, type its alias in the free-text model field under the model buttons (Add agent, Edit agent) instead. An agent pinned on Medium 3.5 relaunches on Medium 3.5 even if you switched with/modelinside Vibe: pick the model in AgentsRoom, or choose Default to keep Vibe's own choice. - My GitHub Copilot agent says it has no instructions, or its tab falls back to the shell after one answer. Fixed on 2026-10-08: update AgentsRoom and relaunch the agent. A saved Copilot agent now starts with its persona loaded as custom instructions (ask it "who are you?" to check), and an agent started with a task stays in an interactive Copilot session after its first answer, so you can reply in the same tab.
- Can the temporary launch override stay on until I turn it off? Yes, since 2026-10-26: in the Temporary launch override window, pick Until I stop it under Duration. The Usage panel then reads "until you stop it" and only Stop override ends it. Custom duration takes hours (up to 720) or days (up to 365): type 6 and pick Days instead of 144 hours. Can I stop or set the launch override from my phone? Yes: Activity tab > Usage > Temporary launch override. Stop override ends the running one; a tap on the card sets a new one. It applies to the computer the phone is connected to, not to your other computers.
- Can I run two providers in the same project? Yes, provider and model are per agent. One agent on Claude Opus, one on Codex, one on Antigravity's free tier, all in the same room.
- Does switching provider lose the conversation? No. AgentsRoom builds a handoff (files touched, activity, last message, optionally an agent-written summary) and the new CLI starts from it. Same-provider model changes on Claude, Codex, Grok and Mistral are instant with no restart.
- How do I hand a task from Claude Code to Codex without explaining it again? Click the model pill in the agent's header to open Switch Provider, pick Codex and a Codex model, keep Agent summary (recommended) under Context transfer, then Summarize & Switch. Claude writes a handoff of what it did and what is left, and the same agent restarts on Codex with that handoff as its first message. Is there a video that shows it? Yes, a Claude Code agent handed to Codex mid-task, in the real app: https://youtu.be/M__kQ5YSThk
- Can I change the model without any handoff? Yes. In the Switch Provider window, choose Start fresh under Context transfer: the agent restarts on the new model with a blank conversation and no handoff message. When the session is empty (just started, or after
/clear) this option is already selected. On the phone, tap the Start fresh card in the Switch Provider sheet. - My oh-my-pi agent starts on an Anthropic model although I only use my own providers in omp. Fixed on 2026-10-09: the oh-my-pi model list now starts with Default, and an agent with no model chosen is launched without a model so omp applies the default of its own configuration. A model you pick explicitly is still passed as is. Older agents left on a catalogue model your omp does not list switch to Default once the model list has been read from omp.
- My oh-my-pi terminal shows only its last few lines, with nothing to scroll back to, after I switch agents or reopen AgentsRoom. Fixed on 2026-10-12. oh-my-pi writes each message once, then only redraws the bottom of its screen, so after a while (above all with a spinner or a timer running) what AgentsRoom kept to redraw the pane no longer contained the conversation. Every time the pane is shown again, AgentsRoom now asks oh-my-pi to redraw itself, and with its default setting (
tui.resizeScrollback: rebuild) it rewrites the whole conversation, scrollback included. Withappendorpreserveit only redraws the visible screen; the Conversation view always has the full history. - The end of a long Claude Code answer comes back garbled in the terminal after the pane or the window changed width: fragments of wrapped lines, a section printed twice, words missing. Fixed on 2026-10-19 on macOS and Linux, and on 2026-11-06 on Windows. Claude Code only repaints the end of its conversation at the new width and cannot touch what already scrolled up, so the copy at the old width stayed above it. A side panel that opens, a split, or a tiling window manager making room for another window is enough to trigger it, even if you never touched the pane. Once the width has settled, AgentsRoom now has Claude redraw the whole conversation at the new width, once. Blocks that were displayed narrower may keep a few blank lines under them: that is how Claude lays them out. The Conversation view always shows the full answer.
- A Claude Code terminal stays almost blank, with only a few fragments of the session: the same number twice, pieces of the logo. Fixed on 2026-11-06. Claude Code only sends what changed on the screen. A console rebuilt during a long session (coming back to its tab, a reloaded or detached window) could only replay the latest small updates once the older output had left its buffer. On Windows nothing repaired it, and on every system a redraw made while the agent was working could leave the status bar at the top of an empty pane. AgentsRoom now has Claude redraw the whole conversation in these cases. On Windows this needs the console component AgentsRoom ships: if an antivirus removed it, resizing the pane brings the screen back.
- The model I want is not in the picker. Pick Default and let the CLI decide, or add
--model <id>in the agent's CLI options. For Cursor, Grok, Devin and GitHub Copilot, click Refresh model list under the picker: it is re-read from your CLI on the spot. For OpenCode, run/connectinside OpenCode: the list updates by itself, and the same button reloads it after a CLI update. - My OpenCode agents pinned on a model fail with
--model is not supported. OpenCode 2 dropped that flag on its main command. Fixed in 1.181.0: AgentsRoom now hands the model to OpenCode through its configuration, and a--modelin your CLI options goes through the same channel. Update AgentsRoom; only Default started before. - My OpenCode agents show no status, no changed files and no token count since OpenCode updated itself, and the terminal looks wrong in dark mode. OpenCode 2.0.14 changed its plugin contract, so the AgentsRoom integration no longer loaded. Fixed in 1.188.0, dark theme included: update AgentsRoom and restart the agents.
- I have OpenCode Go and a Z.AI Coding Plan: which
glm-5.3is which? The plan name travels with the model everywhere (picker description, session pill, ticket line), and the subscription chips over the list keep only one plan's models. - Can I choose the reasoning effort when I create an agent from my phone? Yes: in the New agent sheet, open Engine, pick the provider and the model, then a tier under Reasoning effort. The row only shows for the CLIs that have an effort lever, its tiers follow the model, and Default leaves the CLI on its own. The desktop must be up to date too: an older one creates the agent without the tier.
- My Kimi or Antigravity agent started in a ticket's worktree and waited on "Trust this folder?". Fixed on 2026-09-26: AgentsRoom now records the worktree as trusted in that CLI's settings before launching it. Update AgentsRoom and relaunch the agent.
- Can I put all my Kimi (or Codex, Claude...) agents on Low effort at once? Yes, since 2026-09-28: Settings > AI providers > that provider's tab > Default reasoning effort, or the same row in Project settings > AI providers for one project only. Every agent without its own effort follows it at its next launch; an agent with its own effort keeps it.
- Every agent I launch from the OpenCode Fast tier dies with "Model exo-free has been deprecated". Fixed on 2026-10-10: a tier you never customised no longer lands on a free gateway promo model, it picks a regular model of your OpenCode list. If you pinned the free model on the tier yourself, right-click Fast agent and choose another model.
- Free OpenCode models showed a
$$$price tier. Fixed on 2026-09-27: a model whose id ends in-freeor:free, and OpenCode's freebig-picklegateway, now sit on the cheapest tier ($). - Which Codex model is the equivalent of Opus? Every model list shows it: under each model, "≈ Opus" (or "≈ GPT-6.1 Sol" on a Claude list) names the model of the same capability tier in your reference provider, picked in Equivalents in under the order buttons (Auto by default). The tiers are one scale for every provider, set by AgentsRoom per model family: Fable / GPT-6 Astra (frontier), Opus / GPT-6.1 Sol, Sonnet / GPT-6 Luna, Haiku / Codex Spark. When the reference has no model of that tier, the nearest is shown with
>(more capable than its strongest) or<(lighter than its lightest). Routers (Auto, Default, Adaptive) and families AgentsRoom does not know yet show no equivalent rather than a guess. A model superseded by a newer generation of the same line in the list (GPT-6 Sol next to GPT-6.1 Sol, Opus 4.8 next to Opus 5.5) is marked Older version; it stays selectable. Since 2026-10-05 the$badges follow the same scale, so the same Opus shows the same badge on Claude, Copilot or Devin. - Can I run Command Code or Muse Code (Meta) agents? Yes, since 2026-10-20, on macOS, Windows and Linux: install the CLI (
npm install -g command-code@latestfor Command Code, Meta's installer for Muse Code), sign in once (command-code login,muse login), then pick it as the agent's Provider. Both take a model and a reasoning effort. Since 2026-10-21 both reopen their conversation after a restart, and their quota appears in the Usage panel (Command Code: 5-hour, weekly and monthly plan bars; Muse Code: current window and weekly bars), with the same alerts and quota rules as the other providers. Each reads the account the CLI is signed in to, so there is no automatic switch to a second account. - Can I sort the models by price? Yes: click Price above any model list. The order is remembered for every list; Power brings back the default. Click Price again to see the most expensive models first.
- Is Pi the same thing as oh-my-pi? No. Pi (pi.dev, by Earendil, command
pi) is the original minimal coding agent; oh-my-pi (commandomp) is a fork of it that has grown its own features, config folder (~/.ompinstead of~/.pi) and launch options. AgentsRoom supports both as two separate providers (Pi since 2026-10-22), on macOS, Windows and Linux. - How do I run a Pi agent? Install Pi (
curl -fsSL https://pi.dev/install.sh | shon macOS and Linux,irm https://pi.dev/install.ps1 | iexin Windows PowerShell), runpionce and/loginto connect a provider (or set an API key such asANTHROPIC_API_KEY), then pick Pi as the agent's Provider. The model list is read frompi --list-models(the backends you signed in to), with one chip per backend; Default keeps Pi's own default model. The reasoning effort is Pi's--thinkinglevel, Low to Max. A Pi agent gets the AgentsRoom tools (backlog, memory, browser...) through an AgentsRoom extension loaded at launch, so nothing is written into your repository and the project does not need to be trusted in Pi; Pi lists them on demand with itstool_search. Its conversation reopens after a restart, its transcript feeds the Conversation view, and its file edits show in the Changes tab. Not yet: the Usage panel (Pi has no quota of its own, it uses your providers' accounts). - How do I use OpenRouter / LiteLLM / a company proxy? Environment variables, not flags: set
OPENAI_BASE_URLorANTHROPIC_BASE_URL(and the key, ideally as{{secret:NAME}}) in the provider tab or on the agent. - Why did my agent start in autonomous mode? It is the default for every provider that has a bypass flag. Set Permission mode to Manual in the provider tab, or paste your own flags in CLI options.
- A ticket assigned to a Generic Agent, or an agent spawned by role, started as "codex / opus". Fixed after 1.173.0. A role's default model is a Claude alias; it is now translated to the matching tier of the provider a new agent of the project gets (the project's Default provider when you set one, else the provider of the project's first agent you added yourself, else your Default provider; since 2026-11-16 the project assistant and throwaway agents no longer count), exactly like the lightning button does, instead of handing a Claude model name to another CLI.
- Clicking the lightning button opens a menu instead of launching an agent. That is the 1.184.0 behaviour: nobody found the menu behind the right-click. Pick a model to launch, or tick One-click launch in that menu to get the old direct launch back (the right-click still opens the menu).
- Why does a model tile say "12 users" although I never shared anything? The hint is a community statistic, published for everyone; sharing only decides whether your own launches count in it. Your Last used row is local and never leaves the machine.
- Codex asks for approval on every command. Check Settings > AI providers > Codex tab > Advanced settings > CLI options: custom flags such as
--sandbox workspace-write -a neverkeep the sandbox on; leaving the field on Auto applies the bypass. - A "FULLSCREEN MODE" banner covers the top of my Codex console and the phone shows nothing. Codex 0.157 turned on a full-screen transcript by default. Fixed on 2026-09-26: AgentsRoom now always starts Codex in its classic rendering. Update AgentsRoom and restart the Codex agents.
- How do I get a cheap, a strong and a very expensive agent without browsing the model list? Click the lightning button: Agent tiers is the top section of the menu. Right-click a tier to change its icon, provider, model or reasoning effort, for every project or only for the current one. The same three rows are in Settings > Defaults (all projects) and Project settings > Agents (this project only, with a Use global button to drop the override).
- The lightning menu no longer has Edit default QA agent. It moved out of that menu: the default QA agent is still edited in Settings > Defaults and in the project settings.
Related
- Ollama and Local Models: Ollama and the other catalogue providers, where the model is typed by hand.
- Multiple Accounts: several logins per provider, pinned per project, folder or agent.
- CLI Doctor: what happens when a CLI refuses to start.
- AgentsRoom MCP: how the AgentsRoom tools reach each CLI.
- Adaptive Mode: letting AgentsRoom pick a cheaper model for the task.
- Secret Manager: the secrets an environment variable can reference.
- MCP Servers: your own MCP servers, handed to every CLI at launch.
- Codex Data Storage: the disk the Codex tab lets you reclaim.
- Default Reasoning Effort: one thinking level per CLI, inherited by every agent.