whatsmyagentdoing: Visual Office Dashboard for Local AI Coding Agents

RepositoryCharonyuu/whatsmyagentdoingTool

whatsmyagentdoing visualises active local AI agents such as Claude Code, Codex, or Gemini CLI inside a virtual browser-based office. Developers can monitor the state of parallel sessions in a 2D canvas, an interactive 3D scene, or a card list. The tool runs directly on Node.js without requiring npm installation.

Key points

  1. Tracks concurrent local sessions of Claude Code, Codex, Gemini CLI, and OMP (oh-my-pi).
  2. Animates agent states such as thinking, tool execution, or waiting for human feedback in real time.
  3. Provides three viewing modes: zero-dependency 2D canvas, Three.js 3D environment, and a straightforward list.
  4. Requires only Node.js 18+ with no npm build step, usable directly via CLI or integrated as an agent skill.
  5. Supports downloadable 3D theme packs verified via pinned SHA-256 checksums using CC0 Kenney assets.

Why it matters

Running multiple autonomous coding agents in separate terminal tabs makes tracking status tedious. This tool offers immediate ambient awareness across concurrent tasks in a single visual interface.

Analysis

Problem and Use Case

As developers increasingly operate multiple terminal-based coding agents (such as Claude Code or Codex) simultaneously across various branches or repositories, keeping track of which agent requires intervention becomes challenging. whatsmyagentdoing addresses this friction by providing a visual, centralized dashboard in the browser.

Architecture and Features

The tool monitors local agent sessions and maps their activity to character behaviours in a virtual office. Agents actively processing commands sit at desks, tool runs trigger typing animations, and agents waiting for user input stand up and wave. Idle agents wander to break areas, and environmental lighting reflects local system time.

Technically, the project requires Node.js 18+ and runs directly without an npm install step. It features three rendering modes:

  • Medium (2D): A zero-dependency hand-drawn HTML5 canvas representation.
  • High (3D): Rendered using Three.js (~180 KB gzipped) and meshopt-compressed Kenney CC0 assets (~1.3 MB) with real-time lighting and shadows.
  • List view: Standard cards displaying session details.

Optional 3D environments (such as Space Station or Café) are distributed as isolated data packages validated against pinned SHA-256 checksums before local installation.

Target Audience and Maturity

The tool targets developers running multi-agent workflows locally. With 270 stars and an MIT license, it represents an emerging, functional utility that combines workflow monitoring with an engaging visual metaphor.

Mentioned links

Found in

Summary generated by AI (Gemini 3.8 Flash, 28 October 2026). It may contain errors – the original source is authoritative.

Related entries

Similar in content, found by the AI search.

  • Repository:Leonxlnx/tastecode

    TasteCode: Local Desktop Workspace for AI Coding Agents

    TasteCode is an open-source desktop application for Windows and macOS that combines multiple AI coding agents such as Claude Code, Codex, and Grok into a unified interface. The tool follows a local-first security architecture and includes a dedicated Design Mode for editing web interfaces.

    367starsTypeScript

    AI & LLMs· Tool

  • Repository:hardbeat920/monocode

    MonoCode: Desktop GUI for Coding Agents

    MonoCode is a cross-platform desktop UI that unifies various CLI-based coding agents into a single graphical application. It relies on existing local installations and logins from tools like Claude Code, Codex, and Cursor.

    618starsTypeScript

    AI & LLMs· Tool

  • Repository:getagentseal/codeburn

    CodeBurn: Local Token Usage and Cost Tracking for AI Coding Tools

    CodeBurn is an open-source, locally run utility that monitors token consumption and spending across AI coding assistants. It aggregates metrics across more than 37 tools and agents, breaking down expenditure by model, project, and task.

    10.8kstarsTypeScript

    AI & LLMs· Tool

  • Repository:anthropics/claude-code

    Claude Code: Terminal-Based Agentic Coding Tool

    Claude Code by Anthropic is an agentic coding tool that operates in the terminal. It understands codebases, executes routine development tasks, explains complex code, and handles Git workflows using natural language commands.

    144.1kstarsPython

    AI & LLMs· Tool

  • Repository:pingdotgg/t3code

    T3 Code: Unified Control Surface for Local AI Coding Agents

    T3 Code is an open-source agent harness control surface that lets developers interact with and control local AI coding agents via desktop, web, and mobile applications.

    21.6kstarsTypeScript

    AI & LLMs· Tool

Back to the knowledge base

Let's talk about your project

Get in touch

Locations

  • Mattersburg
    Johann Nepomuk Bergerstraße 7/2/14
    7210 Mattersburg, Austria
  • Vienna
    Ungargasse 64-66/3/404
    1030 Wien, Austria

Parts of this content were created with the assistance of AI.