coding-by-feng/ai-agent-session-center

78 stars · Last commit 2026-06-20

Real-time dashboard that turns AI coding agent sessions (Claude Code, Gemini CLI, Codex) into animated 3D robots — with live terminals, prompt history, tool logs, and queuing. Runs on any device.

README preview

# AI Agent Session Center

**Monitor, drive, queue, and resume every Claude Code, Gemini, and Codex session from one localhost dashboard — each rendered as a live 3D robot you can click into.**

*Built for developers juggling multiple AI coding agents across terminals and machines.*

[![npm version](https://img.shields.io/npm/v/ai-agent-session-center.svg)](https://www.npmjs.com/package/ai-agent-session-center)
[![Node.js](https://img.shields.io/badge/Node.js-18+-green)](https://nodejs.org)
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow)](./LICENSE)

### **[▶ Try it live — zero install, runs in your browser → aasc.work/demo](https://aasc.work/demo)**

<img src="static/screenshot-dashboard.png" alt="3D cyberdrome with active agent sessions across project rooms" width="100%">

**Jump to:** [Install](#install) · [Why power users keep it open](#why-power-users-keep-it-open) · [What's inside](#whats-inside) · [Under the hood](#under-the-hood) · [Commands](#commands) · [Troubleshooting](#troubleshooting) · [FAQ](#faq)

> You're running Claude Code in one terminal, Gemini in another, Codex in a third. Which one is **stuck waiting for approval**? Which one **finished** and needs your next prompt? Which one is **burning tokens on a runaway loop**? Agent Session Center watches all of them at once and surfaces the one that needs you — so you stop tab-juggling and only step in when it matters.

```bash
npx ai-agent-session-center

View full repository on GitHub →