Open Room

Talk to your agents without leaving your window.

Each Claude Code agent gets a name, a voice and its own folder. Ask one for something, by voice or by typing, and get on with your own work.

Open Room is in early development. We'll email you when it's ready.

Name an agent, say the job, and it comes back.

Each agent is a real Claude Code session in its own folder. You address it by name from wherever you are, without switching windows.

  1. Name an agent

    A name, a colour, a voice and a folder. Its role lives in a plain AGENT.md you can read and edit.

    One AGENT.md and one WORKLOG.md per agent

    Edit agent
    Name
    Terminal
    Colour
    Identifies the agent in the sidebar and while it is listening.
    Workspace folder
    C:\work\infraBrowse
    Terminal set up with a colour, a voice and a folder

  2. Say hey and the job

    Any name works instantly, no training. A push-to-talk hotkey or typing does the same thing.

    Custom wake words, no training

    Open Room
    BlockNew conversation
    Listening
    Nothing yet. Ask Block to do something.
    hey Block, check the build

  3. It comes back

    A native notification, or its own voice. If it has a question, it asks, and a question is never dropped.

    Speech queue, native notifications

    Open Room
    TerminalTidy the CI config
    Ready · 2 turns · $0.0186

    hey Terminal, tidy the CI config

    Two files changed in ci/. Want the diff?

    Turn complete · 2 turns · $0.0186

Run several agents at once without watching each one.

Each feature below comes from the app as it is today, not a roadmap.

  1. Address one agent from anywhere

    Say hey and its name, hold a hotkey, or type. You don't have to switch to its window, or find it.

    Wake word · hotkey · typing

    Voice input

    Enable push-to-talkHold the shortcut, say the job, let go.
    Wake wordsKeeps the microphone open and listens for hey and a name.
    Push-to-talk shortcut
    Ctrl+Shift+SpacePress once to start talking, again to send. Esc discards.

  2. It finds you when it's done or stuck

    Agents work in the background and report through a native notification or out loud in their own voice.

    Notification · spoken reply

    Show notificationsA desktop notification when this agent reports in.
    Speak aloudRead this agent's updates out loud, in its own voice.
    Voice
    AvaPreview

  3. Nothing you didn't agree to

    Every tool has three states: ask, always allow, never allow. There is no bypass mode in the UI, including from voice.

    Ask · always allow · never allow

    Permission mode
    Default: act, asking when needed

    Tools

    ReadAlways allow
    EditAsk every time
    BashAsk every time
    WebFetchNever allow

  4. It remembers yesterday

    Conversations persist across restarts, and each agent keeps a WORKLOG.md of what it did.

    Persistent sessions · WORKLOG.md

    Resumed · last active 23 hours ago

    New conversation

    Draft the release notes for 0.423 hours ago
    Fix the changelog links4 days ago

  5. You see the cost as it happens

    Model, effort and permission mode per agent. A banner shows your rate limit; a meter shows context use.

    Quota banner · context meter

    Approaching 5-hour limit · resets 2:30 PM
    ClawdAPI status
    Ready · 3 turns · $0.0412
    78% of 200K
    Opus 5· high effortEdit

Free. Open source. Your own login.

MIT licensed. It uses the Claude Code subscription you already pay for, and every turn bills that. Running five agents means five sessions on one account, and the app tells you when you hit the limit.

Open Room on GitHubMIT licence

Nothing leaves your machine.

No API keys, no telemetry, no server but Anthropic's. Speech in and out runs locally. Voice input ships off, wake words are a second opt-in, and every tool permission is yours to grant.

Keys
Uses the Claude Code login you already have. Nothing to paste.
Mic
Off by default. Wake words are a separate opt-in.
Tools
Ask, always allow, never allow. No bypass mode.
Quota
A banner shows your rate limit as it happens.

Running in a few minutes.

Open Room is in early development. We'll email you when it's ready.

Windows first. macOS is configured but not yet run on a Mac.

  1. Install

    Run the Windows installer.

  2. Sign in

    Use the Claude Code login you already have. No keys.

  3. Name your first agent

    Pick a name, a folder and a voice. Say hey and the name.

    ClawdBitTerminalBlockLoop

Good questions.

  • No. It uses the Claude Code login you already have. Each agent is a real Claude Code session, so every turn bills your existing subscription.