← 返回 Skills 市场
zhuang-he

Fusion Workflow Hub

作者 zhuang-HE · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
86
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install fusion-workflow-hub
功能描述
融合 Graphify 知识图谱 + Everything Claude Code (ECC) 工作流 + OpenClaw 的超级效率中心。 Use when: 用户想要快速理解代码库结构、执行规划/审查/TDD 等标准化工作流、结合知识图谱进行深度代码分析、一站式解决复杂编程任务,或整合多个工具能力提升效率。...
使用说明 (SKILL.md)

\r \r

Fusion Workflow Hub - 融合工作流中心\r

\r

核心定位\r

\r 本技能整合三大 AI 开发工具的能力,形成完整的智能编码工作流:\r \r | 工具 | 核心能力 | Token 节省 |\r |------|----------|------------|\r | Graphify | 代码知识图谱构建与查询 | 71.5x |\r | ECC | 68 命令、36 Agent、151 Skill | - |\r | OpenClaw | 平台集成规则与自动化 | - |\r \r

工作流程\r

\r

工作流一:代码库深度分析\r

\r

1. 使用 Graphify 构建知识图谱\r
   python -m graphify \x3C目录路径>\r
   \r
2. 读取图谱报告\r
   graphify-out/GRAPH_REPORT.md\r
   \r
3. 查询具体问题\r
   python -m graphify query "\x3C问题>"\r
   \r
4. 追踪概念路径\r
   python -m graphify path "\x3C概念A>" "\x3C概念B>"\r
   \r
5. 交互探索(可选)\r
   打开 graphify-out/graph.html\r
```\r
\r
### 工作流二:功能开发(TDD)\r
\r
```\r
1. 使用 tdd-workflow 规划\r
   use_skill tdd-workflow\r
   \r
2. 增量实现代码\r
   \r
3. 代码审查\r
   use_skill code-review\r
   \r
4. 增量更新图谱\r
   python -m graphify . --update\r
```\r
\r
### 工作流三:持续学习优化\r
\r
```\r
1. 经验积累\r
   use_skill continuous-learning-v2\r
   \r
2. 上下文压缩\r
   use_skill strategic-compact\r
   \r
3. 代码复盘\r
   use_skill code-review\r
```\r
\r
## 工具命令速查\r
\r
### Graphify 命令\r
\r
```bash\r
# 构建知识图谱\r
python -m graphify \x3C目录>\r
\r
# 增量更新\r
python -m graphify \x3C目录> --update\r
\r
# 查询\r
python -m graphify query "\x3C问题>"\r
\r
# 路径追踪\r
python -m graphify path "\x3CA>" "\x3CB>"\r
```\r
\r
### ECC 推荐命令\r
\r
```bash\r
# 代码审查\r
/use code-review\r
\r
# TDD 工作流\r
/use tdd-workflow\r
\r
# 持续学习\r
/use continuous-learning-v2\r
\r
# 战略压缩\r
/use strategic-compact\r
```\r
\r
## 输出产物\r
\r
```\r
graphify-out/\r
├── graph.html        # 交互式可视化图谱\r
├── GRAPH_REPORT.md   # 高层摘要(God Nodes、建议问题)\r
├── graph.json        # 持久化图谱数据\r
└── cache/           # SHA256 缓存\r
```\r
\r
## 前置要求\r
\r
- Python 3.8+\r
- Graphify: `pip install graphifyy`\r
- 文档支持: `pip install python-docx openpyxl pillow`\r
\r
## 集成文件\r
\r
本技能包包含:\r
- `SKILL.md` - 本说明文件\r
- `references/graphify-commands.md` - Graphify 命令参考\r
- `references/ecc-best-practices.md` - ECC 最佳实践\r
\r
## 使用示例\r
\r
**用户**: "帮我分析这个代码库的结构"\r
\r
**AI 执行**:\r
1. 调用 Graphify 构建图谱\r
2. 读取 GRAPH_REPORT.md 理解架构\r
3. 使用 query 查询关键模块\r
4. 汇总分析结果\r
\r
---\r
\r
*本技能包整合了 Graphify (15.8k stars)、Everything Claude Code (50k+ stars) 和 OpenClaw 平台能力*\r
安全使用建议
This skill appears to be a wrapper that coordinates Graphify (a code-graph tool) and ECC workflows; that is coherent, but exercise caution before installing or running anything it instructs. Verify the provenance of the referenced packages and projects: find the official Graphify and ECC repositories or PyPI pages (the SKILL references 'pip install graphifyy' but provides no link). Confirm whether 'graphifyy' is the intended package or a typo/fork. Avoid running 'pip install' on unknown packages without reviewing their source code and publisher. Also check that the other ECC/OpenClaw skills it invokes (tdd-workflow, code-review, etc.) actually exist in your agent environment—otherwise the skill may fail or attempt unexpected fallbacks. If you cannot verify upstream sources, treat the skill as untrusted.
功能分析
Type: OpenClaw Skill Name: fusion-workflow-hub Version: 1.0.0 The skill bundle is classified as suspicious due to a potential typosquatting risk in the installation instructions. Both SKILL.md and references/graphify-commands.md direct the user to install 'graphifyy' via pip, while the provided source link points to a repository named 'graphify' (https://github.com/safishamsi/graphify). This discrepancy is a common indicator of a supply chain attack. Additionally, references/ecc-best-practices.md contains hardcoded local Windows file paths (C:\Users\40549\...), which is highly unusual for a general-purpose skill bundle and suggests it may have been captured from a specific, non-sanitized environment.
能力评估
Purpose & Capability
The name and description (codebase analysis + ECC workflows) align with the runtime instructions (invoke Graphify to build/query a knowledge graph and call ECC workflows). However the package provenance is unclear (no homepage/source) despite claims of popular third-party projects; the skill asserts integration with large projects but provides no authoritative links, which reduces trustworthiness.
Instruction Scope
SKILL.md only instructs running Graphify commands, reading local output files (GRAPH_REPORT.md, graphify-out/graph.html), and invoking other ECC/OpenClaw skills via 'use_skill' or '/use' commands. It does not ask the agent to read unrelated system files, environment variables, or transmit data to unknown external endpoints. The instructions are narrowly scoped to code analysis workflows.
Install Mechanism
There is no formal install spec, but SKILL.md directs the user to 'pip install graphifyy' and optional deps. Asking users/agents to pip-install an unverified package is a moderate risk: PyPI packages run arbitrary install-time code and the repository/source for 'graphifyy' is not provided. There is also a minor inconsistency: commands run as 'python -m graphify' while the package name is 'graphifyy'—this could be legitimate (different package/module name) or a typo/fork, but it increases the risk and should be verified before installing.
Credentials
The skill declares no required environment variables, secrets, or config paths. It only references local output files produced by Graphify. There is no disproportionate credential request in the metadata or SKILL.md.
Persistence & Privilege
The skill does not request persistent/always-on presence and does not modify other skills or system-wide settings. Autonomous invocation is enabled by default (not flagged by itself) and is consistent with how the skill expects to call other ECC/OpenClaw skills.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install fusion-workflow-hub
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /fusion-workflow-hub 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
Fusion Workflow Hub version 1.0.0 released. - Initial release combining Graphify 知识图谱, Everything Claude Code (ECC) workflows, and OpenClaw automation into a unified efficiency hub. - Provides step-by-step workflows for codebase analysis, TDD development, and continuous learning. - Includes quick command references for Graphify and ECC, plus recommended setup instructions. - Bundles documentation files for usage instructions and best practices. - Not intended for simple file edits or non-code-analysis queries.
元数据
Slug fusion-workflow-hub
版本 1.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

Fusion Workflow Hub 是什么?

融合 Graphify 知识图谱 + Everything Claude Code (ECC) 工作流 + OpenClaw 的超级效率中心。 Use when: 用户想要快速理解代码库结构、执行规划/审查/TDD 等标准化工作流、结合知识图谱进行深度代码分析、一站式解决复杂编程任务,或整合多个工具能力提升效率。... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 86 次。

如何安装 Fusion Workflow Hub?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install fusion-workflow-hub」即可一键安装,无需额外配置。

Fusion Workflow Hub 是免费的吗?

是的,Fusion Workflow Hub 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Fusion Workflow Hub 支持哪些平台?

Fusion Workflow Hub 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Fusion Workflow Hub?

由 zhuang-HE(@zhuang-he)开发并维护,当前版本 v1.0.0。

💬 留言讨论