Cross-device task overview
See tasks from multiple computers in one view, organized by agent and project, including work started directly at the computer.
BotBus
BotBus lets developers monitor and control AI coding agents running on their computers from an iPhone, Apple Watch, or Android device. It supports task approvals, follow-up prompts, interruptions, new tasks, remote screen access on Mac, and ACP-compatible custom agents.
AI agent control from a phone and watch
BotBus is a companion app for monitoring and controlling AI coding agents that run on a computer. Its desktop component lives in the menu bar, discovers local agents, and syncs task activity to a paired phone and Apple Watch so users can check progress without returning to the computer.
From the mobile apps, users can start new work, continue conversations, answer approval requests, interrupt tasks, and review returned files or media. The product supports six built-in agents and can also connect to other agents that implement Agent Client Protocol (ACP). Availability depends on the connector, task state, supported platform, and whether the computer is online.
Features
Start work remotely
Choose a computer, project, and agent, then submit a one-sentence request from the phone. You can select a model or run the work in its own Git worktree when supported.
Approvals and follow-up
Receive prompts when an agent wants to run a command, edit files, or ask a question. Approve or answer from a phone or Apple Watch, interrupt a task, and enable auto-approval for a project.
Remote screen control
View a live computer screen from the phone and tap, swipe, or type. Screen input starts locked, and the screen stream and input are end-to-end encrypted; this capability currently supports Mac.
Task outputs on mobile
Receive screenshots, files, and videos with task results. Images and PDFs referenced in a conversation can be opened from the phone.
Mobile development workflow
Preview a development server started by an agent on the phone, and use additional controls such as voice input, photo sending, model or reasoning-effort switching, and review of uncommitted changes where the connector supports them.
Use cases
Monitor long-running development work
Keep an eye on coding tasks started at the desk while away from the computer, and return only when an agent needs input or a task produces an important result.
Start a coding task remotely
Launch a task from a phone by selecting the target computer, project, and agent, then provide a short instruction when an idea or bug fix comes up away from the desk.
Handle approvals while mobile
Respond to command, file-edit, or question prompts from a phone or watch so an agent can continue without requiring access to the computer keyboard.
Review remote work and outputs
Use live screen control or open returned screenshots, files, videos, and PDFs when a remote review requires more than a task status update.
Use a custom or company agent
Connect a private or company-built ACP agent to BotBus for mobile task control, using an ACP registry entry or a local manifest when the agent is not in the registry.
Pros and Cons
Pros
- Controls several AI coding agents from a phone and Apple Watch, including approvals, follow-up prompts, interruptions, and new tasks.
- Supports multiple computers and paired devices, with tasks separated by device and project.
- Can discover ACP-compatible third-party or company agents, in addition to six built-in integrations.
- Uses on-device AES-256-GCM encryption for chats, commands, screenshots, and other task content; the relay server receives ciphertext rather than the decryption key.
- Returns practical task artifacts such as screenshots, files, videos, images, and PDFs to the mobile device.
Cons
- Many actions are connector- and task-state dependent; some sessions are view-only.
- Remote screen control currently supports Mac, while Windows is in development and Linux availability is presented as planned or ready to install depending on the installation path.
- Developer previews of web pages are proxied through the relay and are not end-to-end encrypted.
FAQ
How do I set up BotBus?
Install BotBus on the computer running the agent, then choose “Pair a Phone” from the desktop menu and scan the QR code with the phone. The desktop app runs in the background and discovers supported local agents.
Which AI agents can BotBus connect to?
The built-in integrations are Codex, Claude Code, Hermes, Pi, OpenClaw, and DeepSeek Harness. Other agents that implement Agent Client Protocol (ACP), including examples such as OpenCode, can also connect on Mac, Linux, and Windows when installed on the same computer as BotBus.
Does the connected computer need to stay online?
Viewing tasks that have already synchronized does not require the computer to be online. Approving requests, replying, starting or continuing tasks, and interrupting work require the target computer to be online and supported by the connector.
Can BotBus’s servers read my conversations?
Chats, tasks, commands, images, notification text, and remote-control screens are encrypted on the device with AES-256-GCM before transmission. The pairing key is transferred through the computer’s QR code and remains in the paired devices’ keychains; the relay server handles ciphertext and limited routing metadata. Developer web-page previews are proxied through the server and are not end-to-end encrypted.
Can I pair more than one computer, and can every task be controlled?
BotBus supports multiple paired computers and phones, with tasks listed separately by device and project. Control availability can vary by connector and task state, so some tasks may be view-only.
Quick Facts
- Category
- Developer tool; AI agent companion
- Built-in agents
- Codex, Claude Code, Hermes, Pi, OpenClaw, and DeepSeek Harness
- Custom integrations
- Agents implementing Agent Client Protocol (ACP), installed on the same computer as BotBus
- Desktop support
- macOS 26 or later; Linux installation documented/planned; Windows 10/11 in development
- Mobile support
- iOS 26/watchOS 26 or later; Android 8.0 or later, currently in internal testing
- Pairing method
- Scan a QR code displayed by BotBus on the computer
BotBus Alternativen
ByteAsk
ByteAsk is a terminal-first AI coding agent for C and C++ that edits repositories and verifies changes with the real compiler, debugger, sanitizers, and tests before showing a diff. It offers a free tier plus paid plans, with editor connectors and zero-retention handling described in the source.
MeetStream
MeetStream is a meeting bot API for Zoom, Google Meet, Microsoft Teams, and Webex. It helps developers record, stream, and analyze meetings programmatically, with usage-based pricing and a $5 free credit to start.
CreateOS Sandbox
CreateOS Sandbox is an isolated compute environment for running code and agent workloads inside Firecracker micro-VMs. It is designed for workflows that need machine-level isolation, private networking between sandboxes, and programmatic control through SDK, CLI, or MCP.
GitBot
GitBot is a local browser-based workspace for creating reusable coding bots powered by Claude Code, Codex, or OpenCode. It helps developers run those agents across repository folders with persistent threads and configurable permissions.
hob
hob is an independent workspace for coding agents that keeps agent sessions, terminals, history, and follow-up work organized around the tools and providers you already use. It is aimed at developers who want local control over routing, history, and workspace structure rather than a bundled model stack.
Grok 4.7
Grok 4.7 is an xAI model for software development and knowledge work, including extended coding tasks and professional analysis. It is available through Grok Build, Cursor, the Grok API, and other developer platforms, with token pricing starting at $2 per million input tokens and $6 per million output tokens.