jamesrochabrun/AgentHub

Manage all sessions in Claude Code and Codex. Easily create new worktrees, run multiple terminals in parallel, preview edits before accepting them, make inline changes directly from diffs, and more.

What it solves

AgentHub is a desktop application designed to manage and monitor multiple AI coding agents (specifically Claude Code and Codex) simultaneously. It eliminates the need to switch between multiple terminal windows and provides a unified interface for overseeing agent sessions, reviewing their changes, and integrating with development tools like GitHub and iOS Simulators.

How it works

AgentHub acts as a management layer over existing AI CLI tools. It uses file-system watchers to monitor agent sessions in real-time and provides an embedded PTY terminal for interaction. It integrates with the GitHub CLI to surface repository data and uses a dedicated side panel system to render specialized outputs such as MCP (Model Context Protocol) apps, Mermaid diagrams, and data measurements.

Who it’s for

Software developers who use AI coding agents to automate tasks and want a high-level dashboard to coordinate parallel sessions, review code changes via inline diffs, and preview results in web or mobile environments.

Highlights

  • Multi-Agent Orchestration: Launch and monitor Claude Code and Codex sessions side-by-side with a unified hub panel.
  • Integrated Development Tools: Built-in file explorer, editor, and GitHub integration for browsing PRs and issues.
  • Specialized Rendering: Native support for Mermaid diagrams, MCP app UIs, and a "Studio" for rendering agent-generated HTML/CSS mockups.
  • iOS Simulator Integration: Mirror and control iOS simulators in-app, with support for hot-reloading and SwiftUI previews.
  • Measurement Tracking: Records agent-produced data (claims, charts, queries) into a persistent, re-runnable project panel.
  • Web Preview: Automatic detection of localhost servers and Storybook configurations for one-click previews.

Related

  • Project
    asheshgoplani/agent-deck

    Agent Deck is a terminal‑based command center for managing multiple AI coding agents (Claude, OpenCode, Codex, etc.). It offers a unified TUI/CLI, session forking, tool (MCP) and skill attachment, group‑based policies, cost tracking, and a phone‑controlled “conductor” for fleet orchestration.

  • Project
    jazzyalex/agent-sessions

    Agent Sessions is a macOS app that locally indexes and searches conversation histories from many coding‑assistant agents, lets you resume supported CLI sessions, and shows per‑session quota/price usage, all without sending data off‑device.

  • Project
    NanmiCoder/cc-haha

    cc‑haha is a cross‑platform desktop workbench for Claude Code and other LLMs. It provides multi‑session chat, Git branch launching, diff review, built‑in browser preview, permission controls, model selection, image generation, visual sub‑agent and team management, dynamic workflow orchestration, request tracing, and macOS‑only “Computer Use” that lets agents control other apps without hijacking your mouse/keyboard. The app is built with Electron, React, TypeScript, and runs on the Bun runtime, supporting many model providers (Claude, ChatGPT, Grok, local Ollama, etc.) and integrates with IM platforms (Telegram, Slack, WeChat, etc.). Users download a binary for macOS/Windows/Linux, configure API keys, and can also run the underlying CLI for deeper customization.

  • Project
    YoanWai/agent-manager

    A terminal-based manager for AI coding agents that allows developers to run multiple CLI agents side-by-side with unified prompting and integrated diff reviews.

  • Project
    Octane0411/open-vibe-island

    Open Island is a free, native macOS app that lives in the notch or menu bar and provides a live dashboard for AI coding agents (Claude Code, Codex, Cursor, Gemini‑CLI, etc.). It shows session status, handles permission prompts, displays usage stats, and lets you jump back to the exact terminal or IDE pane with a click. Integration is hook‑based: agents emit JSON events, a tiny CLI forwards them via Unix socket to the app, which updates its UI. Supports 13 agents and 15+ terminals/IDEs, runs locally with no telemetry, and is distributed as a signed DMG or Homebrew cask.