# Codexia — Agent Workstation (Codex + Claude)

> Codexia is an agent workstation for Codex CLI and Claude Code, with task scheduling, MCP/skill marketplaces, and one-click data previews (PDF/XLSX/CSV).

## Install

Copy the content below into your project:

## Quick Use

```bash
brew tap milisp/codexia
brew install --cask codexia
```

Launch Codexia, add a project folder, then create a scheduled agent job (per README Quick Start).

## Intro

Codexia is an agent workstation for Codex CLI and Claude Code, with task scheduling, MCP/skill marketplaces, and one-click data previews (PDF/XLSX/CSV).

- **Best for:** teams running multiple agent sessions that want scheduling + marketplace + data tools in one UI
- **Works with:** Desktop app; integrates with Codex CLI + Claude Code; includes MCP marketplace and data preview tools (per README)
- **Setup time:** 10–25 minutes

## Practical Notes

- Install via Homebrew cask; prebuilt releases also available for macOS/Linux/Windows (per README).
- Use the task scheduler for recurring maintenance (dependency updates, lint + test, changelog drafts).
- Keep an “agent baseline” prompt + toolset template so scheduled jobs behave consistently.

## Main

A useful pattern for scheduled agents:

1. Create one “maintenance” job per repo: run tests, summarize failures, and open a todo list.
2. Keep job prompts short and operational: inputs, expected output format, and stopping rules.
3. Store artifacts (logs, diffs, reports) in a predictable folder so humans can review quickly.

If you treat scheduled agents like CI with a narrative report, you’ll get repeatable value instead of random “AI output”.

### FAQ

**Q: Is it only for one CLI?**
A: It’s designed as a workstation: pair it with Codex CLI and Claude Code, plus MCP/skills marketplaces (per README).

**Q: What’s the first job to automate?**
A: A nightly “health check”: `lint → typecheck → tests → summary` is the easiest win.

**Q: How do I avoid flaky automation?**
A: Pin tool versions and keep prompts deterministic; require human review for any destructive action.

## Source & Thanks

> Source: https://github.com/milisp/codexia
> License: AGPL-3.0
> GitHub stars: 675 · forks: 65

---

<!-- ZH -->

## 快速使用

```bash
brew tap milisp/codexia
brew install --cask codexia
```

启动 Codexia，添加项目目录，然后按 README 的 Quick Start 创建一个定时 Agent 任务。

## 简介

Codexia 是面向 Codex CLI 与 Claude Code 的桌面 Agent Workstation：任务调度、MCP/技能市场、PDF/XLSX/CSV 预览与会话管理一体化，支持 brew cask 安装。

- **适合谁：** 希望把多会话、任务调度、市场与数据工具统一到一个桌面工作台的团队
- **可搭配：** 桌面应用；可与 Codex CLI + Claude Code 配合；包含 MCP 市场与数据预览工具（见 README）
- **准备时间：** 10–25 分钟

## 实战建议

- README：可通过 Homebrew cask 安装；也提供 macOS/Linux/Windows 预编译 release。
- 用任务调度跑周期性维护（依赖更新、lint+test、changelog 草稿）。
- 为定时任务准备统一的 baseline prompt + toolset 模板，保证行为一致。

## 主要内容

把定时 agent 当“带叙事的 CI”来用：

1. 每个仓库建一个维护任务：跑测试、总结失败、生成待办清单。
2. prompt 写得更像运维手册：输入、输出格式、停止条件都要明确。
3. 产物（日志、diff、报告）放到固定目录，方便人类快速复核。

当你把定时 agent 当成 CI + 报告，它就会变成可复用的生产力，而不是随机生成。

### FAQ

**它只绑定某一个 CLI 吗？**
答：定位是工作台：结合 Codex CLI 与 Claude Code，并通过 MCP/skills 市场扩展（见 README）。

**最先自动化什么任务？**
答：夜间健康检查：`lint → typecheck → tests → summary` 是最稳的起点。

**怎么避免自动化变得不稳定？**
答：固定工具版本、让 prompt 更确定；任何破坏性动作都要人工复核。

## 来源与感谢

> Source: https://github.com/milisp/codexia
> License: AGPL-3.0
> GitHub stars: 675 · forks: 65


---
Source: https://tokrepo.com/en/workflows/codexia-agent-workstation-codex-claude
Author: Script Depot