# 使用 Homebrew 安装 omnara

查看 omnara 的安装路径、可执行文件、元数据以及面向 AI 代理工作流的安全说明。

## 安装

```sh
sudo av install brew:omnara
```

其他安装命令:

### macOS

- Homebrew (100%):

```sh
brew install omnara
```

  证据: local Homebrew formula metadata

## 软件包事实

- **软件包键:** brew:omnara
- **软件包管理器:** Homebrew
- **版本:** 1.7.0
- **来源摘要:** Talk to Your AI Agents from Anywhere
- **主页:** <https://omnara.com/>
- **最后更新:** 2026-06-22T14:05:40-07:00
- **已生成:** 2026-08-03T19:37:03+00:00

## 可执行文件

- omnara (别名)

## 安装行为

- Bottle: 不可用

## 版本和新鲜度

- 页面生成时间: 2026-08-03
- 管理器版本: 1.7.0
## 项目历史与用法

Omnara is a command center for AI coding-agent sessions. Its CLI starts or attaches to agent work in a local project, while the service exposes progress, prompts, diffs, and approvals through web and mobile surfaces.

### 项目历史

The public GitHub repository was created on July 9, 2025 and describes Omnara as a YC S25 project for talking to AI agents from anywhere. The repository README positions it around Claude Code, Codex CLI, n8n, and similar agents, with a dashboard for real-time visibility and user responses.

The repository is archived, while the product site and documentation describe a broader desktop, CLI, web, mobile, Apple Watch, and sandbox workflow. That split suggests the open repository captured an early public implementation while the service documentation became the stronger source for user-facing behavior.

### 采用历史

Y Combinator's launch page presents Omnara as an agent command center for terminal, web, and mobile, focused on switching between devices and receiving push notifications when agents need input. The GitHub repository accumulated thousands of stars and hundreds of forks, which is a visible adoption signal for a young AI-agent tool.

Mobile app listings and documentation show the package fitting a 2025-2026 developer workflow: agents perform long-running coding tasks, and the human needs a remote approval and steering layer rather than another model provider.

### 使用方式

The CLI reference documents `omnara` as the command that starts an agent session in the current directory, with branch-specific startup, Claude Code and Codex checks, login/logout/status commands, a background daemon, process-kill commands, and workspace sync.

The core concepts documentation models usage around workspaces, worktrees, sessions, providers, and local versus sandbox execution. In the default local model, coding work runs on the user's machine while Omnara keeps the session visible and steerable from phone, web, or another machine.

### 为什么软件包爱好者会关心

Omnara is significant as a packaging example because the CLI is a control plane for other CLIs. It bundles or coordinates agent providers, a daemon, authentication, workspace state, and remote UI rather than being the agent itself.

It also marks the arrival of Homebrew packages whose primary job is human-in-the-loop orchestration for AI coding tools: approvals, diffs, resumed sessions, and remote intervention become packageable developer-tool behavior.

### 时间线

- July 9, 2025: the omnara-ai/omnara GitHub repository is created.
- 2025: Y Combinator's launch page presents Omnara as a command center for AI agents across terminal, web, and mobile.
- November 9, 2025: GitHub release v1.7.0 publishes Python wheel artifacts for macOS and Linux.
- 2026: official documentation describes Codex and Claude Code provider support, workspace sync, remote control, and sandbox execution.

### Related projects

- Omnara is adjacent to Claude Code, Codex CLI, n8n, remote-development dashboards, and phone-based approval tools. Its package identity is less like an editor plugin and more like a session broker between local agent processes and remote human attention.

### 来源

- <https://api.github.com/repos/omnara-ai/omnara>
- <https://docs.omnara.com/cli>
- <https://docs.omnara.com/core-concepts>
- <https://docs.omnara.com/quickstart>
- <https://github.com/omnara-ai/omnara>
- <https://github.com/omnara-ai/omnara/releases/tag/v1.7.0>
- <https://play.google.com/store/apps/details?hl=en_US&id=com.omnara.app>
- <https://www.omnara.com/>
- <https://www.ycombinator.com/launches/OCT-omnara-the-first-command-center-for-ai-agents-terminal-web-and-mobile>


## 安全说明

narrow executable package without higher-risk signals.

- **Geiger 风险:** 绿色 / 低
- narrow executable package without higher-risk signals


## Combined YAML source

View the package source record on GitHub. [combined/omnara.yml](https://github.com/mxcl/pkgdb/blob/main/combined/omnara.yml)


## 来源

- pkg.so package database
- Geiger risk classifier
- curated package history
- pkgdb category and tag curation
- cross-ecosystem install command graph