Repolode
Submit
Nexting-ai

Nexting-ai/nexting

Remote control for Claude Code, Codex, Grok, and Cursor on your Mac or PC, driven from a phone or a voice wearable.

View on GitHubWebsite
Stars
1,968
Forks
9
Open issues
2
Last commit
2 days ago

What is it?

Nexting lets you view live sessions of AI coding agents running on your own Mac or Windows PC, send them tasks, and drive them remotely from an iPhone app or a voice wearable called the PIN. Supported agents include Claude Code, Codex, Grok Build, and Cursor Desktop, with OpenClaw and any OpenAI-compatible local endpoint such as Ollama or vLLM also available as modes.

The main interaction is push-to-talk: you press the PIN, speak, and the request is transcribed on the iPhone and passed to the agent, which runs it in the background and pushes the result back to the phone. Claude Code, Codex, Grok, and Cursor sessions can optionally use end-to-end encryption, so the relay carries ciphertext only.

The repository holds the OpenClaw channel plugin, a multi-agent bus, and the firmware, enclosure files and schematic for the Co-Builder hardware edition, which is based on a Seeed XIAO nRF52840 Sense running Zephyr. It also contains an open device protocol and C99 SDK for building compatible hardware.

Key features

  • 01View live Claude Code, Codex, Grok, and Cursor sessions on a Mac or Windows PC from a phone.
  • 02Send tasks by voice with push-to-talk, transcribed on the iPhone.
  • 03Optional end-to-end encryption for supported agent sessions.
  • 04Works with OpenClaw and OpenAI-compatible local endpoints such as Ollama and vLLM.
  • 05Open firmware, enclosure files, and schematics for a DIY voice wearable.

Use cases

  • Checking on a long-running coding agent session while away from your desk.
  • Dispatching a quick task to a background agent by voice without unlocking your phone.
  • Running a local model through an OpenAI-compatible endpoint and controlling it from a wearable.
  • Building a custom voice input device on the open Zephyr firmware and device protocol.

Quick start

  1. 1

    Install the Hermes bridge CLI from npm (for a local OpenAI-compatible AI such as Ollama or LM Studio)

    npm install -g nexting-hermes-bridge
  2. 2

    Sign in to link your Nexting account

    nexting-hermes-bridge login
  3. 3

    Start the bridge pointed at your local endpoint

    nexting-hermes-bridge start --endpoint http://localhost:8642 --model hermes-agent

Things to know

  • Full use is built around the Nexting PIN hardware ($129), and the Ring is still in private beta.
  • Remote control relies on a Nexting account and the iOS app, so the bridge is not standalone.
  • Several README setup paths require piping a remote install script into a shell, which you should review first.

Topics

#ai-agent#claude-code#codex#remote-control#wearable#voice-assistant#openclaw#embedded#agent-remote-control#ai-wearable

FAQ

What is nexting?
Remote control for Claude Code, Codex, Grok, and Cursor on your Mac or PC, driven from a phone or a voice wearable. Nexting lets you view live sessions of AI coding agents running on your own Mac or Windows PC, send them tasks, and drive them remotely from an iPhone app or a voice wearable called the PIN. Supported agents include Claude Code, Codex, Grok Build, and Cursor Desktop, with OpenClaw and any OpenAI-compatible local endpoint such as Ollama or vLLM also available as modes.
Is nexting free to use? What license does it have?
nexting is open source under the MIT license. Check the license terms before using it in commercial products.
Is nexting still maintained?
Actively maintained. The last commit to nexting was on Oct 20, 2026.
What can I use nexting for?
Checking on a long-running coding agent session while away from your desk. Dispatching a quick task to a background agent by voice without unlocking your phone. Running a local model through an OpenAI-compatible endpoint and controlling it from a wearable. Building a custom voice input device on the open Zephyr firmware and device protocol.
How do I install nexting?
Install the Hermes bridge CLI from npm (for a local OpenAI-compatible AI such as Ollama or LM Studio): npm install -g nexting-hermes-bridge Sign in to link your Nexting account: nexting-hermes-bridge login Start the bridge pointed at your local endpoint: nexting-hermes-bridge start --endpoint http://localhost:8642 --model hermes-agent
Where can I get nexting?
The source code and installation instructions are on GitHub: https://github.com/Nexting-ai/nexting

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects