Enables AI agents to maintain persistent, observable SSH sessions with real command-completion detection, interactive-program support, and allowlisted access, while users can watch activity live via tmux.
Remote-first tmux co-pilot that lets LLMs operate inside real tmux sessions with SSH-aware discovery, deterministic window/pane control, and pull-based state snapshots for grounded terminal assistance.
Enables you to see, start, read, message, approve, interrupt, and stop Claude Code sessions running in tmux across all your SSH-reachable machines from the Claude app, voice mode, or another session. Hooks report which sessions are waiting on you, a shared git-backed memory is injected into new sessions, and small jobs can be routed to local OpenAI-compatible models.
Turns an Android phone running Termux into a remotely drivable machine, letting an AI agent run shell commands, manage background jobs and tmux sessions, read and write files, install packages, and use Android device features such as battery status, screenshots, notifications, SMS and sensors. It can be reached either over stdio on the same device or over Streamable HTTP with bearer tokens and token scopes from anywhere on the local network, with a confirmation gate on destructive actions, a jailed file API and an audit log.
Enables an AI agent to control an Android phone over a private Tailscale network via accessibility UI automation, screenshots, gestures, notifications, clipboard, intents, SAF file access, and optional Termux/Shizuku shells without root or ADB.
