AI Skill Hub 强烈推荐:magnus Agent工作流 是一款优质的Agent工作流。AI 综合评分 8.0 分,在同类工具中表现稳健。如果你正在寻找可靠的Agent工作流解决方案,这是一个值得深入了解的选择。
magnus Agent工作流 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
magnus Agent工作流 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
# 克隆仓库 git clone https://github.com/hrishikeshs/magnus cd magnus # 查看安装说明 cat README.md # 按 README 完成环境依赖安装后即可使用
# 查看帮助 magnus --help # 基本运行 magnus [options] <input> # 详细使用说明请查阅文档 # https://github.com/hrishikeshs/magnus
# magnus 配置说明 # 查看配置选项 magnus --config-example > config.yml # 常见配置项 # output_dir: ./output # log_level: info # workers: 4 # 环境变量(覆盖配置文件) export MAGNUS_CONFIG="/path/to/config.yml"
A magit-inspired interface for managing multiple Claude Code and Codex instances within Emacs.
Run multiple AI agents in parallel, let them communicate to avoid conflicts, and handle their permission requests one at a time.
Website · Getting started · Command reference · Architecture
- Emacs 28.1 or newer - vterm 0.0.2 or newer - transient 0.4.0 or newer - magit-section 3.3.0 or newer - Claude Code, Codex CLI, or both
Install and authenticate each provider CLI before launching it through Magnus. Only the CLI you use is required; install both for the default cross-provider review workflow.
Magnus is available from MELPA:
(use-package magnus
:ensure t
:bind (("C-c m" . magnus)
("C-c M" . magnus-create-instance)))
Or run M-x package-install RET magnus RET.
1. Run M-x magnus. 2. Press c for Claude Code, or press ? and then X for Codex. 3. Choose a Git project and work in the native TUI that opens. 4. Return to *magnus* and create another agent when the work benefits from a second pair of hands. 5. Use RET to visit an agent, m to steer it, and a to handle the next agent that needs attention.
Press ? in *magnus* for the complete dispatcher. The minibuffer shows contextual actions as point moves. In an agent terminal, C-g sends Escape to the provider TUI because Emacs normally reserves ESC as a Meta prefix.
Three agents working side by side in their own vterm buffers, with Magnus keeping the team visible:

Named agents, active work, attention, and coordination in one buffer:

The main interactive entry points are:
| Command | Purpose |
|---|---|
M-x magnus | Open the status buffer |
M-x magnus-create-instance | Create a Claude Code agent |
M-x magnus-create-codex | Create a Codex agent |
M-x magnus-create-headless | Run a one-off headless Claude task |
M-x magnus-context | Open the current project's shared context |
M-x magnus-doctor | Check dependencies, storage, and runtime health |
M-x magnus-upgrade | Reinstall Magnus with a guarded agent shutdown |
Inside *magnus*, press ? for the contextual command menu. See the full command reference for status, trace, context, and review commands.
高质量的AI工作流管理工具
AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。
✅ MIT 协议 — 最宽松的开源协议之一,可自由商用、修改、分发,仅需保留版权声明。
总体来看,magnus Agent工作流 是一款质量优秀的Agent工作流,在同类工具中具备一定竞争力。AI Skill Hub 将持续追踪其更新动态,建议收藏备用,结合自身场景选择合适时机引入使用。
| 原始名称 | magnus |
| 原始描述 | 开源AI工作流:Manage multiple Claude Code AI agents in Emacs — coordination, attention queue, 。⭐11 · Emacs Lisp |
| Topics | ai-agentsemacsclaude-code |
| GitHub | https://github.com/hrishikeshs/magnus |
| License | MIT |
| 语言 | Emacs Lisp |
收录时间:2026-05-30 · 更新时间:2026-05-30 · License:MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。
选择 Agent 类型,复制安装指令后粘贴到对应客户端