LodyAI/Lody

Share coding agents with your team on phone and desktop

Lody – Shared Workspace for Coding Agents

What it is – Lody is an open‑source platform that lets a team use AI‑powered coding agents (Claude Code, Codex, Kimi, OpenCode, etc.) from any device. It synchronises conversations, code changes, and machine resources across desktop, mobile, web and a CLI, turning each AI‑agent session into a collaborative, version‑controlled workspace.

Key capabilities

  • Shared conversations – Open the same chat with teammates, see the full transcript, runtime status, generated files and diffs, and add new instructions without copying screenshots.
  • Bring your own agents & machines – Connect any ACP‑compatible coding agent and any machine (laptop, server, cloud VM) that already has the model subscription and credentials. Machines stay private until explicitly shared.
  • Multi‑surface dispatch – From the desktop app, iOS/Android app, web UI, or the lody CLI you can start work on a chosen machine, watch progress, approve permission requests and inspect results.
  • Agent coordination – An agent can spawn child conversations, read their status, send follow‑up prompts, or cancel them, enabling one “coordinator” agent to delegate bug investigation, implementation and testing.
  • Git‑aware workspaces – Each session gets its own Git worktree, so parallel agents never clash. Lody shows per‑turn diffs, lets you comment line‑by‑line, and displays PR status and CI alongside the chat.
  • Visual feedback loop – Run a web app inside a session, annotate UI elements, and send those annotations back to the agent for visual debugging.
  • Built‑in tools – Terminal, command palette, shortcuts, file explorer, and mobile Live Activities for notifications and approvals.
  • Usage insights – Token, quota, machine and agent resource consumption are displayed per session.

How to get started

  1. Install the CLI – npx lody daemon start launches a daemon on any workstation, server or cloud VM. It opens a sign‑in link that registers the machine with your Lody workspace.
  2. Create a session – From any terminal (or script) you can launch a coding task, e.g.
    npx lody session create --workspace my‑team \
    --agent-config codex \
    --repo owner/repo "Fix the failing test"
  3. Join from other surfaces – Open the Lody desktop app, mobile app, or web UI, select the same workspace, and you’ll see the live conversation, file tree, diffs and terminal.
  4. Collaborate – teammates can add instructions, comment on diffs, attach files, or fork the session into a new worktree for alternative approaches.

Project layout

  • apps/cli – the command‑line daemon and session manager.
  • apps/electron – desktop client (macOS, Windows, Linux).
  • packages/* – UI components, shared schemas, CRDT‑based sync (Loro), ACP extensions, optional cloud‑API definitions, and experimental end‑to‑end encryption core.
  • site-docs – website, docs and blog.

Why it matters Lody bridges the gap between powerful AI coding assistants and real‑world software development workflows. By treating each agent conversation as a first‑class, version‑controlled artifact that can be shared, forked and inspected, teams can safely harness AI productivity while keeping code history, reviews and security under their control.

Links

Related

  • Project
    Innei/lody-ios

    Lody iOS is a community‑maintained iOS client for the Lody AI‑agent platform. It combines React Native UI with Swift native modules to deliver a fully native‑look chat interface, real‑time CRDT sync (Loro/Flock), code‑diff viewing, and live preview of remote dev servers via a Quick‑Tunnel. Distributed through TestFlight and licensed under AGPL‑3.0.

  • Project
    makecindy/cindy

    Cindy is an open-source AI agent that runs locally on your own machine, combining multiple agent engines (harnesses like Claude Code and Codex), models, and tools into one agent that does real work in your projects. This repo contains the desktop and mobile client apps (Electron and Expo/React Native) plus shared packages, organized as a pnpm monorepo. The backend service lives in a separate repository. You can bring models via the official Cindy service, your existing Claude Code/Codex Coding Plan, your own API keys, or local models. The client is free and Apache-2.0 licensed, customizable through memory, skills, automation, MCP, and plugins, with official builds including opt-out TapDB analytics.

  • Project
    codesstar/loci

    Loci is an open‑source, file‑based memory system that lets different AI agents read and write shared context stored locally as Markdown/JSON. It offers personal memory, task scheduling, project indexing, contact logs, a knowledge vault, and a local Node.js dashboard, all without requiring external databases.

  • Dispatch
    Codex Goes Mobile: Bridging the Gap Between the Desk and the World

    OpenAI integrates Codex into the ChatGPT mobile app, allowing developers to steer long-running AI agents and approve code changes from anywhere.

  • Project
    RizRiyz/luvus

    Luvus is a cross‑platform terminal UI that lets you launch, monitor, and orchestrate AI coding agents (Claude, Codex, Grok, etc.) alongside shells, editors, and Git tools. It provides persistent workspaces, pane management, agent‑aware status panels, Git/GitHub integration, remote‑machine support, a read‑only web view, and an extensible module system. Install via a script, Homebrew, or PowerShell, then use `luvus` to start a session, `luvus doctor` to verify setup, and `luvus update` for upgrades.