coollabsio/jean

A dev environment for AI agents.

Jean – An Integrated Desktop & Server Workspace for AI‑Assisted Development

What it is – Jean is a native‑desktop (Tauri) application, with an optional headless jean‑server mode, that lets you run many large‑language‑model (LLM) agents (Claude, Codex, Cursor, Grok, etc.) side‑by‑side while you work on code. Each agent operates inside its own Git worktree, so every feature, bug‑fix, or AI‑generated task lives in an isolated checkout that can be created, archived, or restored automatically.

Why it matters – Modern AI‑coding tools are often single‑purpose (e.g., a chat window that calls one model). Jean treats AI agents as first‑class collaborators: they can read/write the same repository, open terminals, run diffs, and even create pull‑requests, all from a single UI. The design keeps the heavy lifting on machines you control (your laptop or a self‑hosted server), avoiding vendor lock‑in.

Key capabilities

Area What Jean provides
Project & worktree handling Multi‑project view, automatic Git worktree creation/archiving, labels, custom avatars.
Session management Parallel sessions per worktree, “Plan/Build/Yolo” execution modes with approval steps, goal tracking (/goal), AI‑summarized recaps, auto‑naming, retention policies.
AI back‑ends Supports a dozen LLM providers (Claude, Codex, Cursor, OpenCode, Pi, Command Code, Grok, Kimi Code, Antigravity β). You pick model, temperature/effort, and can run multi‑agent collaborations.
Magic commands & prompts One‑click actions for issue triage, PR generation, code review, commit‑message drafting, merge‑conflict resolution, release‑note creation, all with editable system prompts.
Automation (Mr. Robot β) Scans GitHub issues, creates a worktree per issue, drafts a plan, and (once you approve) executes it.
Integrations GitHub (issues, PRs, security alerts, Dependabot, auto‑archive), Linear, Sentry, CodeRabbit.
Developer tools Built‑in terminal (xterm.js), file browser with preview/editor, diff viewer with line comments, command palette, ability to open files in external editors (VS Code, Zed, Xcode, etc.).
Plugins RTK (reduces token usage 60‑90 %), Caveman (short Claude replies).
Local + remote Desktop app can connect to any number of jean‑server instances over SSH or direct URL; projects from all servers appear together.
Web Access Every instance runs an HTTP/WebSocket UI (default port 3456) that can be reached from phones, tablets, or other LAN machines. Token‑based auth is enabled by default; Tailscale is recommended for secure remote access.
Customization Themes, fonts, key‑bindings, mobile swipe gestures.

How to get it – Pre‑built binaries are distributed via GitHub Releases and a Homebrew Cask (brew install --cask jean). For developers, the source can be built with Node 19+, Rust, and the usual C++ toolchain on Windows.

Typical workflow

  1. Add a repository (local or remote). Jean creates a Git worktree for the task.
  2. Start a new session and pick an AI backend.
  3. Use the magic command palette to ask the agent to, say, generate a PR from an open GitHub issue.
  4. Review the diff inside Jean, add line comments, run the built‑in terminal to test, then merge.
  5. Archive the worktree; Jean retains a summarized recap for future reference.

Who it’s for – Software engineers who already use LLMs for code assistance and want a unified, self‑hosted environment that ties together version control, issue tracking, and AI agents without hopping between separate tools.

All details above are taken directly from the repository’s README.

Related

  • 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
    dark-hxx/CLI-Manager

    A cross-platform AI CLI workbench that integrates terminal management, session history, token analytics, and SSH remote access for AI coding agents like Claude Code and Codex.

  • Project
    KunAgent/Kun

    Kun is a local‑first AI desktop workspace that merges code editing, document management, and chat‑style agents into one Electron app. All data stays on your computer; you connect any OpenAI‑compatible, Anthropic‑compatible, or local model (e.g., Ollama, DeepSeek). Features include project‑wide code assistance, a research‑paper workspace, extensible plugins, and a terminal UI. It’s aimed at developers and researchers who want AI‑augmented productivity without cloud‑side data storage, and it’s released under a non‑commercial PolyForm license.

  • Project
    hermes-hq/hermes-ide

    Hermes IDE is a cross‑platform desktop app that lets you run multiple code‑generation agents (Claude, Codex, Copilot, etc.) side‑by‑side, each in its own real terminal and Git worktree. It provides inbox notifications, per‑turn diffs, a Review Desk for merging, and optional structured Claude‑view UI, all built with Tauri, React and Rust.

  • Project
    coulsontl/ai-toolbox

    AI Toolbox is a cross‑platform desktop manager that centralises configuration, prompts, plugins, session history and backup for many AI coding assistants (Claude, Codex, Gemini, Grok, Kimi, etc.). It offers a GUI, local proxy gateway, WSL/SSH sync, skill marketplace, and system‑tray shortcuts, built with React/Tauri and SQLite.