SuperAPI AI Directory

content-factory-dispatcher

Writing & research

View repository

Updated 10-15 07:54

About this skill

Content Factory Dispatcher — Content Factory Dispatch Center

A content production task scheduling system based on the Ralph CEO loop.

How to Use

【CEO指令】跑一遍内容工厂流水线
【CEO指令】今天的内容主题是 AI 编程
【CEO指令】检查内容团队工作状态

Dispatch Flow

1. Hotspot Collection Stage

派给小data: 采集今天热点 → 输出热点池

2. Topic Selection Stage

派给小research: 调研热点 → 输出选题建议
派给小content: 评估选题 → 确认内容方向

3. Content Generation Stage

派给小content: 生成初稿 → 输出多版本草稿

4. Review Stage

派给小pm: 审核草稿 → 质量评分 + 修改建议

5. Publishing Stage

派给小market: 制定发布策略
派给小ops: 执行发布

Ralph Loop Mode

while not 任务完成:
    # 1. 派活
    派给Agent(任务)
    
    # 2. 等待产出
    等待文件产出
    
    # 3. 质量检查
    if 质量不达标:
        派给小pm反馈 → Agent修改
    else:
        进入下一阶段

Agent Dispatch Mapping

Stage Agent Tool
Collection Xiao Data content-source-aggregator
Research Xiao Research content-research-writer
Creation Xiao Content content-creator
Review Xiao PM project-management
Publishing Xiao Market seo
Operations Xiao Ops linux-service-triage

Output Directory

~/clawd/projects/content-factory-agents/output/
├── 2026-03-05/
│   ├── hotpool.json      # 热点池
│   ├── topics.json       # 选题
│   ├── drafts/           # 草稿
│   ├── reviewed/         # 审核后
│   └── published/        # 已发布

Extension

To add a new stage, simply:

  1. Add the stage definition to the phases array
  2. The corresponding Agent skill is ready
  3. The Ralph loop dispatches automatically