Best open-source alternatives to Conductor

Orca is a desktop application for coordinating AI coding agents across multiple worktrees and workflows. It is aimed at developers who want to run tools like Claude Code, Codex, OpenCode, and other terminal-based agents in parallel while keeping their work organized in one interface. The project combines an editor-like workspace, terminals, browser-assisted design mode, issue and PR browsing, SSH-backed remote worktrees, and a mobile companion for monitoring agents on the go. It also includes a CLI for scripting common actions, making it useful both as an interactive control center and as an automation tool for agent-driven development.

Herdr is a terminal-based runtime for coding agents, designed for people who run tools like Claude Code, Codex, Cursor, Opencode, and similar agent workflows. It acts as a background server that owns terminal sessions so work can continue even if the user closes the lid, loses network connectivity, or restarts the machine. The project is aimed at developers who want persistent, inspectable agent sessions with minimal friction. It emphasizes a tmux-like experience with both keyboard and mouse interaction, pane status visibility, and a socket API that agents can use directly. The README also highlights plugin support and remote reattachment over SSH, suggesting it is meant for local and remote interactive workflows rather than a hosted SaaS product.

Paseo is a self-hosted platform for coordinating AI coding agents across several interfaces, including desktop, mobile, web, and a command-line client. It is aimed at developers who want to run agents on their own machines and keep their existing tools, configs, and credentials in place while switching between surfaces as needed. The project centers on a local daemon that manages agent execution and exposes connectivity for clients to attach to it. The README describes support for multiple agent providers, voice interaction, device pairing, remote access over TCP or VPN, and a skill system that helps agents hand work off to one another. It also includes a Docker deployment path and a CLI workflow for headless or remote use.

T3 Code is a control surface for agent-based coding tools that runs on a user’s machine and can be operated from mobile, web, or desktop clients. It is designed for people who already use agents such as Claude Code, Codex, Cursor, Grok Build, and OpenCode, and it acts as a remote-ready interface for controlling those local tools. The project emphasizes being open and forkable, with a focus on performance and remote access. The README points readers to in-repo documentation for installation, permissions, keyboard shortcuts, project customization, remote access, synchronization, and source control integrations, and it also describes how to run the backend locally via `npx` or install the desktop app through common distribution channels.

OpenRig is a CLI tool for defining, launching, and managing a team of AI coding agents in a persistent shared rig.

Codeman is a self-hosted control panel for AI coding agents. It is aimed at developers who want to run tools like Claude Code, OpenCode, Codex, or Gemini CLI in persistent sessions, observe their progress in real time, and keep them running even when the browser disconnects or the user steps away. The project runs a web dashboard backed by tmux sessions, allowing agents to continue in the background with features such as idle-based reprompting, automatic resumption after usage limits reset, scheduled jobs, and replayable terminal history. It also supports mobile-friendly interaction, QR-based login, and a multi-user mode with separate logins and workspaces for small teams.

Ouijit is a desktop-style tool and CLI for coordinating agent-driven development work. It is aimed at users who work with agent harnesses such as Claude Code, Codex, Pi, and OpenCode, and need a way to manage tasks, terminal sessions, hooks, and parallel workstreams from a single interface. The project emphasizes practical workflow support: it can track task state, expose a session-aware `ouijit` command in every terminal it opens, manage worktrees automatically, and sandbox untrusted code in VMs. It also surfaces live agent status, notifications, markdown files, and web previews as panels. The README describes it as free and open source, with no account, sign-in, or telemetry requirements.

Vibe Kanban is a self-hostable kanban-style workspace designed to help software teams plan and review work done by coding agents. It is aimed at developers who use tools like Claude Code, Gemini CLI, Codex, Amp, Cursor, and similar agents to implement changes while humans coordinate tasks and review results. The project combines issue planning, agent execution, and code review in one interface. Users create kanban issues, spin up workspaces that provide an agent with a branch, terminal, and dev server, then inspect diffs, leave inline comments, preview the app in a built-in browser, and create pull requests for merge. The README also describes support for remote deployment, reverse proxy setups, and SSH-based editor integration.

Claude Squad is a terminal-based task manager for AI coding assistants. It is aimed at developers who want to run several agent sessions at once, keep each session isolated, and monitor or steer them from a single TUI.

Emdash is a desktop application designed for developers who want to run multiple AI coding agents at the same time. It keeps each task in its own Git worktree and branch, allowing users to explore fixes or features in parallel, compare diffs, and merge the results from one interface. The project targets local development workflows as well as remote codebases accessed over SSH/SFTP. It integrates with existing agent CLIs such as Claude Code, Codex, OpenCode, Amp, and others, and can track agent status, notifications, and resumable sessions through lifecycle hooks where supported. It also emphasizes local-first privacy by storing app state in a local SQLite database and not sending code or chat data to Emdash servers.

Nimbalyst is a desktop and mobile workspace built for people using coding agents such as Codex, Claude Code, OpenCode, and Copilot. It gives users a shared visual environment for editing agent-produced work, reviewing diffs, managing sessions, and organizing tasks, all while keeping content as plain files in a git repository. The project is aimed at developers who want to coordinate multiple AI-assisted sessions in parallel and keep their work linked across documents, worktrees, commits, and notes. It includes a desktop Electron app, an iOS companion app, built-in Git and MCP support, extension points for custom editors, and documentation for both end users and contributors.

Jean is an opinionated workspace for AI-assisted software development. It is aimed at developers who want to run several coding agents at once, keep each task isolated in its own git worktree, and manage sessions, reviews, terminals, and issue-driven workflows from one place. The project offers both a native desktop application and a headless server mode. The desktop app can combine local and remote servers in one sidebar, while the server exposes the same UI over HTTP/WebSocket with token authentication. It integrates with major coding agents and developer services, supports customized prompts and execution modes, and lets users control everything on infrastructure they own rather than through a vendor-locked cloud service.