# An Orca alternative with Gemini CLI in structured views Orca calls itself [the ADE for working with a fleet of parallel agents](https://github.com/stablyai/orca): an open-source desktop app that runs coding agents side by side, each in its own git worktree, with an editor, terminals and a browser built in. Moontors is an agentic development environment (ADE) that runs Claude Code, Codex and Gemini CLI side by side, each through its own protocol, in Moontors' own views. ## At a glance | | Orca | Moontors | | ---------------------- | ---------------------------------------------------------------------------------------------------------- | -------------------- | | Agents | [Any CLI agent](https://onorca.dev/docs/agents/supported), run in a terminal, with about 40 set up already | Claude Code, Codex and Gemini CLI, through each one's own protocol | | Views | A terminal per agent; an experimental [Chat UI](https://onorca.dev/docs/agents/native-chat) for some | Moontors' own views for every agent, the only way it shows them | | Platforms | [macOS, Windows and Linux](https://github.com/stablyai/orca), with mobile companion apps | macOS 13+, Windows 10+ (x64), Linux (x86-64); an Android app on [Google Play](https://play.google.com/store/apps/details?id=com.moontors.mobile), the iPhone's coming soon; no build is signed yet, and macOS and Windows ask before the first open; on Windows 11, Smart App Control, when it's on, blocks unsigned apps, Moontors among them; early preview (0.x) ([Known limits](/docs/known-limits/)) | | Isolation | A [git worktree per agent](https://github.com/stablyai/orca) | A git worktree per session, when you want one ([Worktrees](/docs/worktrees/)) | | When the app is closed | A [background daemon](https://onorca.dev/docs/model/session-restore) keeps the agents running | Sessions keep working in a background service ([Sessions and the board](/docs/sessions-and-board/)) | | Remote work | [SSH, and a headless server](https://onorca.dev/docs/ways-to-run) | Local only | | Licence | [MIT, open source](https://github.com/stablyai/orca) | Free, closed source | | Data | [Anonymous usage telemetry](https://onorca.dev/docs/telemetry), which you can turn off; no account | No Moontors account and no Moontors servers. Moontors itself sends no telemetry; each agent's own usage data follows that agent's settings | ## What each has that the other doesn't Orca: - open source, under the [MIT licence](https://github.com/stablyai/orca); - [any command-line agent](https://onorca.dev/docs/agents/supported), not only three; - a built-in [browser](https://github.com/stablyai/orca), and comments on diff lines that go back to the agent; - agents on other machines, over [SSH or a headless server](https://onorca.dev/docs/ways-to-run); - [GitHub](https://onorca.dev/docs/review/github), [Linear](https://onorca.dev/docs/review/linear) and [Jira](https://onorca.dev/docs/review/jira) in the app; - [scheduled automations](https://onorca.dev/docs/cli/automations) and a [scripting CLI](https://onorca.dev/docs/cli/overview). Moontors: - each agent driven through its own protocol, with its permission requests, models, efforts and commands in Moontors' own views, the only way it shows them, Gemini CLI's too, where Orca's [Chat UI](https://onorca.dev/docs/agents/native-chat) is an experimental setting that doesn't cover Gemini CLI; - the agents' own settings, MCP servers and skills listed, and switched on and off for Moontors' sessions without editing your files ([Configuration](/docs/configuration/)); - an Android app on [Google Play](https://play.google.com/store/apps/details?id=com.moontors.mobile) that follows and answers your computer's own sessions, end to end, with no Moontors server between them; the iPhone's coming soon. ## Where Orca is the better fit - You want an open-source app. - You use agents beyond Claude Code, Codex and Gemini CLI. - You want a browser in the same window as the agents. - Your agents run on other machines, over SSH or on a server. ## Questions ### What's the difference between Moontors and Orca? Both run several agents at once on macOS, Windows and Linux, and keep them running with the app closed: Orca in a [background daemon](https://onorca.dev/docs/model/session-restore), Moontors in a background service. Orca runs [any command-line agent](https://onorca.dev/docs/agents/supported) in a terminal, each in [its own git worktree](https://github.com/stablyai/orca). Moontors drives Claude Code, Codex and Gemini CLI through each one's own protocol, so their permission requests, models and commands appear in Moontors' own views, and gives a session its own git worktree when you ask for one. ### Does Orca show the agents in its own views? Partly. Orca's [Chat UI](https://onorca.dev/docs/agents/native-chat), an experimental setting, shows Claude Code, Codex, Grok, OMP and OpenCode as structured chats, with a newer structured mode for Claude Code, Codex and Grok, and question cards, model and effort pills and a list of slash commands; it doesn't cover Gemini CLI. In Moontors, its own views are the only way it shows an agent, Gemini CLI's included. ### Is Moontors open source? No. Moontors is free, but closed source. Orca is open source, under the [MIT licence](https://github.com/stablyai/orca). ### Can Moontors run agents on another machine? No. Moontors runs agents on your own computer only. Orca can run them over [SSH or on a headless server](https://onorca.dev/docs/ways-to-run). ## Related - [What is an agentic development environment (ADE)?](/docs/what-is-an-ade/) - [How to use Claude Code, Codex and Gemini CLI together](/docs/how-to/claude-code-codex-and-gemini-cli-together/) - [How to use git worktrees for AI coding agents](/docs/how-to/git-worktrees-for-ai-agents/)