AIAny
AI Coding2025
Icon for item

TUIOS

Runs many shells inside one terminal and treats coding agents as first-class panes. Combines daemon-backed persistent sessions, vim-style modal control, BSP tiling, a session rail and an Inbox that surfaces approvals, questions and finished turns from LLM-based coding agents — useful when you run and manage agent workflows alongside shells.

Introduction

Developers increasingly run language-model-powered helpers inside terminal workflows, but traditional multiplexers treat those processes as opaque PTYs. TUIOS flips that assumption: it makes agent activity first-class UI state (working, waiting-for-you, done, errored) and gathers everything that needs human attention into a single Inbox, so you can triage and reply without hunting through panes.

What Sets It Apart
  • Agent-aware session model: Panes running coding agents report state to the daemon and appear on the session rail and in a global Inbox. That lets you answer approvals or questions from a compact interface rather than switching to the agent's pane.
  • Persistent, daemon-backed sessions with layout resurrection: Window trees, working directories and layouts survive detaches and daemon restarts (running processes and scrollback do not), letting you reattach from other machines or share sessions across hosts.
  • Terminal-native UX for power users: Vim-style modal window management, BSP tiling with smart preselection, multifocus and a command palette combine terminal ergonomics with a window manager's features; tape scripting records and replays workflows for automation.
  • Agent orchestration features: Integrations for many agent CLIs, Inbox approvals, agent-to-agent messaging, fleets (fan-out prompts to multiple agents), pane grants (fine-grained agent permissions) and optional MCP surface for model-context tooling.
  • Rich terminal capabilities: Kitty and sixel passthrough, vim-style copy mode across long scrollback, multifocus copy, and a live theme/settings UI keep the experience fully terminal-native while supporting graphical output where available.
Who It's For and Tradeoffs

Great fit if you: are a developer who runs LLM-based coding agents inside terminals; want a tmux-like multiplexer that understands and surfaces agent work; need cross-host / ssh-attached sessions and reproducible window layouts; or want to script and replay terminal workflows.

Look elsewhere if you: need GUI-first LLM tooling or web UIs for non-terminal users; require full process resurrection including in-memory program state and scrollback after a reboot (TUIOS restores layout and directories but not running processes or scrollback); or cannot run in a terminal with decent protocol support (truecolor/kitty/sixel recommendations improve features).

Where It Fits

Think of TUIOS as a modern, agent-aware alternative to tmux: it keeps the terminal at the center of development, but treats agents, approvals and inter-agent messaging as first-class concerns. It complements GUI LLM tools by letting engineers keep an integrated, keyboard-driven workflow while coordinating multiple model-based helpers across machines.

Information

  • Websitegithub.com
  • AuthorsGaurav Gosain
  • Published date2025/09/06

More Items

GitHub
AI Client2024

Unifies visual creator tools, reusable Skills, and a Codex-powered Agent to handle video translation, image/video generation, voiceovers, and editing in one local multimodal workspace. Desktop and web clients share a synchronized state machine and keep data local by default.

GitHub
AI Coding2026

Provides a rule set and agent skill that lets AI coding assistants review and improve SwiftUI code for modern APIs, performance, accessibility, and maintainability. Packs checks for deprecated APIs, data flow, navigation, and accessibility tailored to mistakes LLMs commonly make.

GitHub
AI Agent2026

Audits deployed AI agents for governance and operational misalignment, running up to 60 inspections that detect deception, manipulation, fabrication, unpredictability and opacity; outputs JSON/Markdown scorecards, a graded A–F score, and optional Pro reports and dashboards.