Back to Details

multi-search

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

多主题深度研究技能

Multi-Topic In-Depth Research Skill

概述

Overview

智能多主题深度研究工具，自动分析材料并生成系统化研究文档。支持任意材料输入，通过并行启动多个独立研究 Agent进行深度检索，形成精简的研究知识库。
核心原则：
  • 只做信息检索、归纳与专业表述转化
  • 不新增事实，不虚构信息
  • 聚焦单一问题，精简输出，能够解决问题即可
  • 通用设计，可适用于法律、商业、技术、学术等各领域
An intelligent multi-topic in-depth research tool that automatically analyzes materials and generates systematic research documents. It supports input of any materials, and forms a streamlined research knowledge base by launching multiple independent research Agents in parallel for in-depth retrieval.
Core Principles:
  • Only perform information retrieval, induction and professional expression conversion
  • Do not add new facts or fabricate information
  • Focus on single issues, output streamlined content that suffices to solve the problem
  • Universal design, applicable to fields such as law, business, technology, academia, etc.

触发条件

Trigger Conditions

使用 
/multi-search
命令触发，或当用户请求：
  • 深度研究多个相关课题
  • 进行系统性信息检索
  • 整合多角度分析
  • 需要生成结构化研究报告
Trigger with the 
/multi-search
command, or when the user requests:
  • In-depth research on multiple related topics
  • Systematic information retrieval
  • Integration of multi-angle analysis
  • Generation of structured research reports

输入格式

Input Formats

方式一：基于文件

Method 1: Based on Files

/multi-search @文档路径.md
/multi-search @document-path.md

方式二：直接粘贴

Method 2: Direct Paste

/multi-search
[粘贴材料内容]
/multi-search
[paste material content]

方式三：指定课题

Method 3: Specify Topics

/multi-search
项目：[项目名称]
研究课题：
1. [课题一]
2. [课题二]
3. [课题三]
/multi-search
Project: [Project Name]
Research Topics:
1. [Topic 1]
2. [Topic 2]
3. [Topic 3]

处理流程

Processing Flow

阶段1：分析准备

Phase 1: Analysis & Preparation

  1. 读取输入材料
  2. 提取研究课题清单
    • 有明确课题：直接使用
    • 无明确课题：自动从材料中提取
  3. 课题拆分原则：
    • 方向明确：每个课题对应独特的检索方向
    • 避免重叠：确保课题之间检索关键词不重复
    • 聚焦问题：每个课题解决一个具体问题
  4. 确定项目名称和输出位置
  1. Read input materials
  2. Extract research topic list
    • With explicit topics: use directly
    • Without explicit topics: automatically extract from materials
  3. Topic splitting principles:
    • Clear direction: Each topic corresponds to a unique retrieval direction
    • Avoid overlap: Ensure no duplicate retrieval keywords between topics
    • Focus on problem: Each topic solves one specific problem
  4. Confirm project name and output location

阶段2：输出目录智能检测

Phase 2: Intelligent Output Directory Detection

按优先级检测项目结构：
  1. 优先检测：
    output/
    目录 → 使用 
    output/[项目名]/
  2. 其次检测：当前工作目录 → 使用 
    ./[项目名]/
  3. 兜底方案：用户当前目录 → 使用 
    ./research/
创建目录：
[输出目录]/03 - 🔍 深度研究/
Detect project structure by priority:
  1. Priority detection: 
    output/
    directory → Use 
    output/[Project Name]/
  2. Secondary detection: Current working directory → Use 
    ./[Project Name]/
  3. Fallback solution: User's current directory → Use 
    ./research/
Create directory: 
[Output Directory]/03 - 🔍 In-Depth Research/

阶段3：并行深度研究

Phase 3: Parallel In-Depth Research

为每个研究课题启动独立的 
general-purpose
独立研究 Agent。
上下文传递（主Agent → 独立研究 Agent）：
  • 项目关键信息（背景、目标、核心问题）
  • 完整的课题清单及各课题的检索范围
  • 已分配的关键词方向（避免重复的依据）
  • 具体需求背景
去重检查机制：
每个 独立研究 Agent 在开始检索前，必须遵循以下流程：
  1. 检索前声明：
    • 在当前上下文中声明："我将检索 [关键词A, 关键词B] 用于研究 [课题名称]"
    • 等待主 Agent 确认无重复后再开始
  2. 主 Agent 审核：
    • 检查该 Agent 声明的关键词是否与已分配方向重复
    • 如发现重复，及时通知该 Agent 转向其他方向
  3. 动态调整：
    • 如果某方向已被其他 Agent 覆盖，该 Agent 应转向相关但不同的角度
    • 记录调整后的检索方向
深度检索要求：
  • 4-6轮深度检索
  • 自动选择 WebSearch（搜索发现）或 WebFetch（获取完整内容）
  • 关键词差异化，确保每个 Agent 覆盖独特角度
文档生成：
  • 聚焦解决单一核心问题
  • 简洁明了，能够解决问题即可
  • 包含关键来源链接
  • 直接可用的结论和建议
Launch independent 
general-purpose
research Agents for each research topic.
Context Transfer (Main Agent → Independent Research Agent):
  • Key project information (background, objectives, core issues)
  • Complete topic list and retrieval scope for each topic
  • Assigned keyword directions (basis for avoiding duplicates)
  • Specific requirement background
Duplication Check Mechanism:
Each independent research Agent must follow this process before starting retrieval:
  1. Pre-retrieval Declaration:
    • Declare in the current context: "I will retrieve [Keyword A, Keyword B] for researching [Topic Name]"
    • Wait for confirmation from the Main Agent that there is no duplication before starting
  2. Main Agent Review:
    • Check if the keywords declared by the Agent overlap with already assigned directions
    • If duplication is found, notify the Agent to switch to other directions promptly
  3. Dynamic Adjustment:
    • If a direction has been covered by other Agents, this Agent should switch to a related but different angle
    • Record the adjusted retrieval direction
In-Depth Retrieval Requirements:
  • 4-6 rounds of in-depth retrieval
  • Automatically select WebSearch (search discovery) or WebFetch (full content acquisition)
  • Differentiated keywords to ensure each Agent covers a unique angle
Document Generation:
  • Focus on solving a single core problem
  • Concise and clear, sufficient to solve the problem
  • Include key source links
  • Provide directly usable conclusions and suggestions

阶段4：整合输出

Phase 4: Integration & Output

  1. 生成研究总览文档（000.研究总览.md）
  2. 整合所有 独立研究 Agent 的核心发现
  3. 创建文档间导航链接
  4. 添加综合建议和立即行动清单
  1. Generate research overview document (000.Research Overview.md)
  2. Integrate core findings from all independent research Agents
  3. Create navigation links between documents
  4. Add comprehensive suggestions and immediate action checklist

输出格式

Output Formats

目录结构

Directory Structure

[输出目录]/
└── [项目名]/
    └── 03 - 🔍 深度研究/
        ├── 000.研究总览.md
        ├── YYMMDD [研究课题一].md
        ├── YYMMDD [研究课题二].md
        └── ...
[Output Directory]/
└── [Project Name]/
    └── 03 - 🔍 In-Depth Research/
        ├── 000.Research Overview.md
        ├── YYMMDD [Research Topic 1].md
        ├── YYMMDD [Research Topic 2].md
        └── ...

总览文档格式

Overview Document Format

markdown
undefined
markdown
undefined

[项目名称] 深度研究总览

[Project Name] In-Depth Research Overview

生成时间: YYYY-MM-DD 研究方式: N个独立研究 Agent，各进行4-6轮深度检索 总检索轮次: XX+轮 总文档量: XX KB
Generation Time: YYYY-MM-DD Research Method: N independent research Agents, each conducting 4-6 rounds of in-depth retrieval Total Retrieval Rounds: XX+ rounds Total Document Size: XX KB

研究成果清单

Research Results List

已完成的N份精简研究报告

Completed N Streamlined Research Reports

序号研究课题文件大小核心价值
01课题一XX KB简要描述
No.Research TopicFile SizeCore Value
01Topic 1XX KBBrief description

核心发现

Core Findings

发现1：[最重要发现]

Finding 1: [Most Important Finding]

依据：[简要说明]
结论：[具体结论]
Basis: [Brief explanation]
Conclusion: [Specific conclusion]

综合建议

Comprehensive Suggestions

一、策略建议

1. Strategic Recommendations

推荐方案：[具体方案]
Recommended Solution: [Specific plan]

二、立即行动清单

2. Immediate Action Checklist

  • 行动项1
  • 行动项2
undefined
  • Action Item 1
  • Action Item 2
undefined

详细研究文档格式

Detailed Research Document Format

markdown
undefined
markdown
undefined

[研究课题标题]

[Research Topic Title]

生成时间: YYYY-MM-DD 研究深度: XX+轮深度检索，覆盖XXXX、XXXX、XXXX
Generation Time: YYYY-MM-DD Research Depth: XX+ rounds of in-depth retrieval, covering XXXX, XXXX, XXXX

核心结论

Core Conclusion

[最重要的发现和结论，2-3段，充分详实]
[Most important findings and conclusions, 2-3 paragraphs, sufficiently detailed]

一、[主要内容一]

1. [Main Content 1]

（一）子标题

(1) Subtitle

正文段落。引用来源使用内嵌链接格式：
Body paragraphs. Use inline link format for citations:

二、[主要内容二]

2. [Main Content 2]

[继续结构化内容]
[Continue structured content]

三、应用建议

3. Application Suggestions

（一）建议要点

(1) Key Suggestions

内容：[具体内容]
Content: [Specific content]

（二）注意事项

(2) Notes

⚠️ [注意点]

---
⚠️ [Key Note]

---

链接规范

Link Specifications

核心原则

Core Principle

所有来源链接必须内嵌到正文中相应位置
markdown
✅ 正确：
根据[研究报告](https://链接)显示...

❌ 错误：
根据某报告...
（文末单独列出引用来源）
All source links must be embedded in the corresponding position in the body text
markdown
✅ Correct:
According to [Research Report](https://link)...

❌ Incorrect:
According to a certain report...
(List citations separately at the end of the document)

链接标注约定

Link Label Conventions

  • 🔗 → 通用网页资源
  • 📚 → 学术文献
  • 🏛️ → 机构官网
  • 📄 → 数据来源
  • 🔗 → General web resources
  • 📚 → Academic literature
  • 🏛️ → Institutional official websites
  • 📄 → Data sources

文档命名规范

Document Naming Specifications

编号系统

Numbering System

  • 00.
    - 研究总览
  • 01-09.
    - 核心研究
  • 10-19.
    - 重要研究
  • 20+.
    - 延伸研究
  • 00.
    - Research Overview
  • 01-09.
    - Core Research
  • 10-19.
    - Important Research
  • 20+.
    - Extended Research

主题规范

Topic Specifications

  • ✅ 使用简洁中文标题
  • ✅ 避免特殊字符
  • ✅ 长度15字以内
  • ✅ 清晰反映研究对象
  • ✅ Use concise Chinese titles
  • ✅ Avoid special characters
  • ✅ Within 15 characters
  • ✅ Clearly reflect the research object

质量标准

Quality Standards

独立研究 Agent研究质量

Independent Research Agent Research Quality

  • 聚焦单一问题：每个 独立研究 Agent 仅解决一个核心问题
  • 检索深度：4-6轮检索（够用即可）
  • 精简输出：简洁明了，能够解决问题就行
  • 关键引用：引用关键来源（够用即可）
  • 直接可用：提供直接可用的结论和建议
  • Focus on single problem: Each independent research Agent only solves one core problem
  • Retrieval depth: 4-6 rounds of retrieval (sufficient is enough)
  • Streamlined output: Concise and clear, sufficient to solve the problem
  • Key citations: Cite key sources (sufficient is enough)
  • Directly usable: Provide directly usable conclusions and suggestions

文档质量标准

Document Quality Standards

  • 结构清晰：章节标题层级分明
  • 叙述连贯：段落式叙述，避免过度罗列
  • 链接准确：所有链接内嵌到正文相应位置
  • 格式统一：遵循统一的格式规范
  • 可操作性强：提供具体的步骤、工具、命令
  • Clear structure: Hierarchical chapter titles
  • Coherent narrative: Paragraph-based narration, avoid excessive listing
  • Accurate links: All links embedded in corresponding positions in the body text
  • Unified format: Follow unified format specifications
  • High operability: Provide specific steps, tools, commands

注意事项

Notes

严禁行为

Prohibited Behaviors

  • ❌ 不创建二级子目录（如"引用素材/"）
  • ❌ 不生成独立的执行总结文件
  • ❌ 不使用过度罗列的要点格式
  • ❌ 不在文末单独列出引用来源
  • ❌ 不添加冗余的进度跟踪章节
  • ❌ Do not create secondary subdirectories (e.g., "Reference Materials/")
  • ❌ Do not generate separate executive summary files
  • ❌ Do not use excessive bullet point formats
  • ❌ Do not list citations separately at the end of the document
  • ❌ Do not add redundant progress tracking sections

推荐做法

Recommended Practices

  • ✅ 使用叙述式段落表达
  • ✅ 链接内嵌到相应正文位置
  • ✅ 保持简洁的研究总览
  • ✅ 提供具体的行动建议
  • ✅ 标注清晰的文档编号
  • ✅ Use narrative paragraph expression
  • ✅ Embed links in corresponding positions in the body text
  • ✅ Keep research overview concise
  • ✅ Provide specific action suggestions
  • ✅ Use clear document numbering

依赖

Dependencies

本技能依赖 Claude Code 内置工具，无需额外配置：
  • WebSearch：搜索发现
  • WebFetch：获取完整内容
  • Task：启动独立 独立研究 Agent
This skill relies on built-in tools of Claude Code, no additional configuration required:
  • WebSearch: Search discovery
  • WebFetch: Full content acquisition
  • Task: Launch independent research Agents

变更历史

Change History

版本日期更新内容
v1.0.02026-02-15从 Command 迁移为 Skill，重命名为多主题深度研究（multi-search）
VersionDateUpdate Content
v1.0.02026-02-15Migrated from Command to Skill, renamed to Multi-Topic In-Depth Research (multi-search)