AI Skill Hub 推荐使用:Kukeon AI工作流引擎 是一款优质的Agent工作流。AI 综合评分 7.0 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。
Kukeon AI工作流引擎 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。
Kukeon AI工作流引擎 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。
# 方式一:go install(推荐) go install github.com/eminwux/kukeon@latest # 方式二:从源码编译 git clone https://github.com/eminwux/kukeon cd kukeon go build -o kukeon . # 方式三:下载预编译二进制 # 访问 Releases 页面下载对应平台二进制文件 # https://github.com/eminwux/kukeon/releases
# 查看帮助 kukeon --help # 基本运行 kukeon [options] <input> # 详细使用说明请查阅文档 # https://github.com/eminwux/kukeon
# kukeon 配置说明 # 查看配置选项 kukeon --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 KUKEON_CONFIG="/path/to/config.yml"
Self-hosted runtime for AI coding agents on your own Linux host.
Kukeon runs Claude Code and other agent workloads as isolated containerd cells on a single Linux host you own — from a cloud VM down to a Raspberry Pi. Each agent gets an attachable terminal, logs, scoped secrets, and reproducible blueprints/configs, with the Git forge as the system of record. No SaaS sandbox, no Kubernetes cluster, every action inspectable.
Start with one agent cell; grow into per-project agent teams declared with kuketeam.yaml.
/run/containerd/containerd.sockRelease binaries are published for amd64 and arm64.
curl -fsSL https://kukeon.io/install.sh | bash
The installer detects your platform, verifies the prerequisites above (and prints a distro-aware hint on any miss), downloads the latest release, verifies its sha256 checksum, installs kuke + kukeond to /usr/local/bin, runs sudo kuke init to bring the daemon up, and (on systemd hosts) installs /etc/systemd/system/kukeond.service so kukeond comes back automatically after a host or containerd restart. On systemd-less hosts the unit step is skipped with a notice — bring kukeond up manually after each reboot with sudo kuke daemon start. Pass --check to run the prereq checks only without touching the system:
curl -fsSL https://kukeon.io/install.sh | bash -s -- --check
After install completes you should see the daemon's realm/space/stack hierarchy provisioned and kukeond listening on its unix socket:
Realm: default (namespace: default.kukeon.io)
System realm: kuke-system (namespace: kuke-system.kukeon.io)
Run path: /opt/kukeon
...
kukeond is ready (unix:///run/kukeon/kukeond.sock)
If you would rather drive each step yourself — installing onto an air-gapped host, or pinning a non-default release tag — see the manual install steps.
kuke uninstall removes everything kukeon put on the host — every realm and its workloads, the *.kukeon.io containerd namespaces, /opt/kukeon, /run/kukeon, and the kukeon system user and group. Only the kuke/kukeond binary is left behind:
sudo kuke uninstall
→ See docs/site/cli/kuke-uninstall.md for details, and docs/site/architecture/storage-layout.md for everything kukeon writes on a host.
Get kukeon running on a single Linux host in minutes.
Agent原生设计理念先进���声明式编排降低使用门槛。但项目早期阶段,文档和社区支持不足,适合技术探索而非生产首选。
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。
✅ Apache 2.0 — 宽松开源协议,可商用,需保留版权声明和 NOTICE 文件,含专利授权条款。
总体来看,Kukeon AI工作流引擎 是一款质量良好的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 | kukeon |
| 原始描述 | 开源AI工作流:🌪️ Agent-native orchestration. Declarative, self-hostable runtime for AI agents。⭐6 · Go |
| Topics | Agent编排工作流引擎自托管云原生Go语言 |
| GitHub | https://github.com/eminwux/kukeon |
| License | Apache-2.0 |
| 语言 | Go |
收录时间:2026-06-10 · 更新时间:2026-06-11 · License:Apache-2.0 · 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