← Back to Skills Marketplace
windy-001-crypto

Ai Workflow Master

by windy-001-crypto · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
110
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install ai-workflow-master
Description
AI工作流大师,集成内容创作/数据分析/自动化/报告生成的一站式工作流。触发词:'做这个'、'帮我搞定'、'AI工作流'、'自动化'、'批量处理'、'一键生成'、'效率工具'。通过串联多个AI能力,实现复杂任务的自动化完成。
README (SKILL.md)

AI Workflow Master

AI工作流大师,让AI自动化完成复杂任务。

核心能力

能力 说明 适用场景
内容流水线 选题→写作→发布 自媒体运营
数据分析流 采集→清洗→分析→报告 市场调研
报告生成器 素材→整理→排版→输出 周报/月报/分析报告
竞品监控流 爬取→比对→预警→建议 运营监控
客服自动化 识别→回复→记录→优化 客户响应

触发词

当用户说以下内容时激活:

  • "帮我做"、"一键生成"
  • "AI工作流"、"自动化"
  • "批量处理"、"效率工具"
  • "做这个流程"、"帮我搞定"
  • "快速完成"、"提高效率"

预设工作流

工作流1:内容创作流水线

输入:关键词/话题
    ↓
Step 1: 热点分析(当前热度、竞争度)
    ↓
Step 2: 标题生成(10个备选)
    ↓
Step 3: 正文写作(按选定风格)
    ↓
Step 4: 配图建议(关键词/风格)
    ↓
Step 5: 多平台适配(抖音/小红书/公众号)
    ↓
输出:可直接发布的内容包

调用示例:

用户:"帮我写一篇关于新能源车的小红书"
→ 激活内容创作流水线
→ 输出:封面标题 + 正文 + 标签 + 发布时间建议

工作流2:数据分析流水线

输入:数据源URL/文件
    ↓
Step 1: 数据采集(网页爬取/文件读取)
    ↓
Step 2: 数据清洗(去重/格式化)
    ↓
Step 3: 统计分析(趋势/对比/排名)
    ↓
Step 4: 可视化建议(图表类型)
    ↓
Step 5: 结论输出(3-5个关键洞察)
    ↓
输出:数据分析报告

工作流3:周报生成流水线

输入:工作内容要点
    ↓
Step 1: 结构化整理(分类/排序)
    ↓
Step 2: 数据量化(数字/百分比)
    ↓
Step 3: 亮点提炼(3个核心成果)
    ↓
Step 4: 问题识别(2个待改进)
    ↓
Step 5: 下周计划(3-5个目标)
    ↓
输出:格式完整的周报

工作流4:竞品监控流水线

输入:竞品名称/链接
    ↓
Step 1: 公开数据采集(价格/评价/动态)
    ↓
Step 2: 竞品对比(表格化)
    ↓
Step 3: SWOT分析
    ↓
Step 4: 机会点识别
    ↓
Step 5: 行动建议
    ↓
输出:竞品分析报告

工作流5:智能客服流水线

输入:客户问题
    ↓
Step 1: 意图识别(售前/售后/投诉/闲聊)
    ↓
Step 2: 知识库匹配(最佳答案)
    ↓
Step 3: 回复生成(专业/友好/简洁)
    ↓
Step 4: 情绪判断(安抚优先/直接解决)
    ↓
Step 5: 工单记录(必要时自动创建)
    ↓
输出:回复内容 + 后续建议

工作流设计原则

原则1:输入最小化

❌ 让用户填写10个字段
✅ 只需要1个关键词,其余AI推断

原则2:输出最大化

❌ 只输出一个结果
✅ 输出主内容 + 备选 + 建议 + 预警

原则3:可干预

用户可以随时介入调整方向
不要让AI完全"黑箱"运行

自定义工作流

工作流定义格式

工作流名称: [名称]
触发条件: [关键词/场景]
输入要求: [最少必要信息]

Steps:
  - id: step_1
    name: [步骤名]
    action: [调用工具/函数]
    input: ${前序输出}
    output: [产出描述]

  - id: step_2
    name: [步骤名]
    action: [调用工具/函数]
    input: ${step_1.output}
    output: [产出描述]

输出:
  format: [格式]
  channels: [发布渠道建议]

使用示例

示例1:一键生成小红书

用户:"帮我写一篇关于特斯拉降价的小红书"
→ 输入:特斯拉降价
→ 工作流:内容创作流
→ 输出:
  ├── 封面标题:《等等党又赢了!特斯拉史低,这次真的可以买?》
  ├── 正文:800字种草文
  ├── 标签:#特斯拉 #降价 #新能源汽车 #选车
  ├── 发布时间:今晚8点(黄金时段)
  └── 备选标题:3个

示例2:竞品分析

用户:"帮我分析一下比亚迪和特斯拉"
→ 输入:比亚迪、特斯拉
→ 工作流:竞品监控流
→ 输出:
  ├── 对比表格:价格/续航/智能化/服务
  ├── SWOT分析
  ├── 各自优势总结
  ├── 差异化建议
  └── 行动清单

示例3:批量内容生产

用户:"我要发一周的抖音,帮我批量生成内容"
→ 输入:领域(汽车)+ 数量(7条)
→ 工作流:内容批量流
→ 输出:
  ├── Day1:热点追评型
  ├── Day2:产品测评型
  ├── Day3:知识科普型
  ├── Day4:对比评测型
  ├── Day5:用户互动型
  ├── Day6:品牌故事型
  └── Day7:促销引流型

输出规范

标准输出格式

# [任务名称] 完成报告

## 执行摘要
[1-2句话概括结果]

## 详细内容

### 1. [模块标题]
[内容]

### 2. [模块标题]
[内容]

## 附件/备选
- 备选1: [内容]
- 备选2: [内容]

## 下一步建议
1. [建议A]
2. [建议B]

## 执行记录
- 耗时: X分钟
- 步骤: X步
- 状态: ✅ 完成

效率提升计算

传统方式 AI工作流 效率提升
写一篇小红书: 2小时 5分钟 24x
做竞品分析: 1天 30分钟 16x
生成周报: 1小时 5分钟 12x
批量生成7天内容: 14小时 1小时 14x

最佳实践

  1. 明确目标 - 告诉AI你要什么,不是怎么做
  2. 提供素材 - 给你有的数据/文档/链接
  3. 接受迭代 - 第一版往往不是最好的,多轮调整
  4. 保留人工审核 - AI输出需要人工确认再发布
  5. 记录成功模式 - 好的工作流可以复用

常用工作流模板

模板1:市场调研流

输入: 行业/竞品关键词
↓
[数据采集] → [清洗整理] → [分析洞察] → [报告生成]
↓
输出: 市场调研报告

模板2:内容营销流

输入: 产品/话题
↓
[热点分析] → [选题确认] → [内容创作] → [多平台适配]
↓
输出: 可发布内容包

模板3:销售赋能流

输入: 客户信息/竞品信息
↓
[客户画像] → [话术生成] → [异议处理] → [促成建议]
↓
输出: 销售策略建议

模板4:客服响应流

输入: 客户问题/对话记录
↓
[意图识别] → [知识匹配] → [回复生成] → [满意度预判]
↓
输出: 回复方案 + 后续建议
Usage Guidance
This skill promises scraping, multi-platform publishing, and automatic ticket creation but doesn't declare how it will authenticate or which endpoints it will use. Before installing or enabling it: 1) ask the publisher how integrations and credentials are supplied and where network/file access occurs; 2) do not provide API keys or admin credentials until you confirm exact endpoints and minimal required scopes; 3) require explicit confirmation in the skill for any action that publishes or creates records (do not allow automatic posting); 4) prefer running it in a controlled environment or with test accounts first; and 5) if you need the automation, request the author to list required env vars, config paths, and safe defaults so you can review them.
Capability Analysis
Type: OpenClaw Skill Name: ai-workflow-master Version: 1.0.0 The skill bundle 'ai-workflow-master' consists of a metadata file and a markdown instruction file (SKILL.md) defining various AI-driven workflows for content creation, data analysis, and reporting. The instructions provide templates and logic for the agent to follow when triggered by specific user prompts, such as 'automation' or 'batch processing'. There is no evidence of malicious code, data exfiltration, or prompt injection designed to subvert the agent's behavior for harmful purposes.
Capability Assessment
Purpose & Capability
The skill claims broad capabilities (web scraping, multi-platform publishing, automatic ticket creation, knowledge-base matching). Those capabilities normally require API endpoints and credentials for target services (e.g., 小红书/抖音/公众号、工单系统). The package declares no env vars, credentials, or config paths, which is an inconsistency (useful functionality but missing required integration details).
Instruction Scope
SKILL.md instructs the agent to perform actions that imply reading external URLs/files and interacting with remote services (网页爬取/文件读取/知识库匹配/自动创建工单/发布到多平台). The instructions are high-level and permit the agent to 'infer' missing inputs, which is vague and could lead to broad data collection or attempts to access unknown local/remote resources without explicit boundaries.
Install Mechanism
No install spec and no code files — instruction-only. This minimizes on-disk code execution risk and there are no download URLs or package installs to review.
Credentials
Declared requirements list no environment variables or credentials, yet the workflows describe actions that typically require secrets (platform API keys, service tokens, ticketing system credentials). The absence of declared credentials is disproportionate to the claimed capability and leaves unclear how the skill will authenticate to external systems.
Persistence & Privilege
Flags show always:false and no special OS or persistent hooks. The skill does not request permanent presence or elevated platform privileges in the manifest.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install ai-workflow-master
  3. After installation, invoke the skill by name or use /ai-workflow-master
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of AI Workflow Master. - Provides a one-stop workflow integrating content creation, data analysis, automation, and report generation. - Built-in workflows for content pipelines, data analysis, report generation, competitor monitoring, and customer service automation. - Triggered by common automation keywords such as "帮我搞定", "AI工作流", "批量处理", and more. - Minimal user input required; AI infers most needs and optimizes output (content packs, analysis reports, etc.). - Supports custom workflow definitions and best practices for efficient, reliable AI-assisted work.
Metadata
Slug ai-workflow-master
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Ai Workflow Master?

AI工作流大师,集成内容创作/数据分析/自动化/报告生成的一站式工作流。触发词:'做这个'、'帮我搞定'、'AI工作流'、'自动化'、'批量处理'、'一键生成'、'效率工具'。通过串联多个AI能力,实现复杂任务的自动化完成。 It is an AI Agent Skill for Claude Code / OpenClaw, with 110 downloads so far.

How do I install Ai Workflow Master?

Run "/install ai-workflow-master" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Ai Workflow Master free?

Yes, Ai Workflow Master is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Ai Workflow Master support?

Ai Workflow Master is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Ai Workflow Master?

It is built and maintained by windy-001-crypto (@windy-001-crypto); the current version is v1.0.0.

💬 Comments