Skip to content
Research
Command

/refine-question

将原始研究问题细化为结构化的深度研究提示词

From plugin
claude-code-stock-deep-research-agent
3625 skills5 commands
Install
$ npx -y skills add liangdabiao/Claude-Code-Stock-Deep-Research-Agent --agent claude-code

How it fires

How this command gets triggered: by you, by Claude, or both.

  • Fires itselfClaude auto-loads it when your prompt matches the work.
  • You can call itInvoke it directly when you want it.
  • Slash command/refine-question

Context preview

What this command does when you run it.

将原始研究问题细化为结构化的深度研究提示词

Command definition

refine-question.md
description: 将原始研究问题细化为结构化的深度研究提示词
argument-hint: [原始研究问题]

Question Refinement

Refine the user's research question into a structured research prompt.

Raw Question

$ARGUMENTS

Your Task

Use the **question-refiner** skill to:

1. **Ask Clarifying Questions**: Ask 5-6 relevant questions about:

  • Specific focus areas within the topic
  • Output format requirements (report, presentation, etc.)
  • Geographic and time scope
  • Target audience
  • Special requirements

2. **Wait for User Response**: Do NOT generate the structured prompt until the user answers.

3. **Generate Structured Prompt**: Create a structured research prompt with:

  • **TASK**: Clear, concise statement of what needs to be researched
  • **CONTEXT/BACKGROUND**: Why this research matters
  • **SPECIFIC QUESTIONS**: 3-7 concrete sub-questions
  • **KEYWORDS**: Search terms and synonyms
  • **CONSTRAINTS**: Timeframe, geography, source types, length
  • **OUTPUT FORMAT**: Detailed structure with lengths and components
  • **FINAL INSTRUCTIONS**: Citation requirements

Important Notes

  • Be patient: Better to ask one more question than deliver a vague prompt
  • Be specific: Every field should be filled with concrete, actionable details
  • Focus on the user's needs, not what you think they should want

Begin the question refinement process.

Ships withclaude-code-stock-deep-research-agent

Investment Research Edition - 专业股票投资尽调系统 ⚖️ 免责声明 本研究报告不构成投资建议或推荐。所有投资存在风险,包括本金损失。 重要提示: 本报告仅供教育和信息用途 部分数据需要通过官方渠道验证 过往业绩不代表未来表现 投资决策前请自行进行尽职调查 建议咨询合格的财务顾问 🎓 研究框架 本研究基于 Claude Code Deep Research 系统: 方法论: 8阶段股票投资尽调框架 智能体: 28个并行研究智能体 工具: WebSearch、WebFetch、综合分析

Get the whole plugin