planning-task

Task Planning Skill

Breaks down complex user requests into structured, trackable task plans with dependencies.

When to Use

Load this skill when:

• User has a multi-step request (e.g., "制作产品PPT需要市场数据")
• Need to organize work into phases/steps
• User mentions: "plan", "tasks", "steps", "breakdown", "organize"
• Complex deliverable requiring coordination

Capabilities

1. Task Decomposition: Break complex goals into atomic tasks
2. Dependency Management: Identify which tasks must complete before others
3. Progress Tracking: Generate plan.json and todo.md for monitoring
4. Format Generation: Create both machine-readable (JSON) and human-readable (Markdown) formats

Workflow

Phase 1: Analyze User Intent

Understand the goal and identify key deliverables:

# Use code_execution to analyze
user_intent = "制作AI产品介绍PPT，包含市场数据"

# Identify components
components = [
    "搜索市场数据",
    "分析竞品信息", 
    "设计PPT结构",
    "生成SlideSpeak配置",
    "验证配置",
    "渲染PPT"
]

Phase 2: Generate Structured Plan

Use the helper script to create plan.json:

# Load and execute the plan generator
with open('skills/library/planning-task/scripts/generate_plan.py', 'r') as f:
    exec(f.read())

plan = generate_task_plan(
    user_intent="制作AI产品介绍PPT，包含市场数据",
    tasks=[
        {"id": "task_001", "description": "搜索AI客服市场数据", "dependencies": []},
        {"id": "task_002", "description": "生成PPT配置", "dependencies": ["task_001"]},
        {"id": "task_003", "description": "渲染PPT", "dependencies": ["task_002"]}
    ]
)

# Save plan.json
import json
with open('workspace/plan.json', 'w') as f:
    json.dump(plan, f, ensure_ascii=False, indent=2)

Phase 3: Generate Human-Readable Todo

Create todo.md for user visibility:

# Load and execute the todo generator
with open('skills/library/planning-task/scripts/generate_todo.py', 'r') as f:
    exec(f.read())

todo_markdown = generate_todo_markdown(plan)

# Save todo.md
with open('workspace/todo.md', 'w') as f:
    f.write(todo_markdown)