Back to Details

huashu-research

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

调研 Skill

Research Skill

结构化的网络调研流程，核心目标：调研成果实时持久化，防止会话截断丢失工作。
Structured web research workflow with the core goal: real-time persistence of research results to prevent work loss due to session truncation.

何时使用

When to Use

  • 为写文章做前期调研
  • 了解新产品、新技术、新发布
  • 搜集竞品信息或行业动态
  • 任何需要多次 WebSearch 的信息搜索任务
  • Pre-research for article writing
  • Learn about new products, technologies, and releases
  • Collect competitor information or industry trends
  • Any information search task requiring multiple WebSearches

执行流程

Execution Process

Step 1: 立即创建调研文件

Step 1: Immediately Create a Research File

  • 在开始搜索之前，先创建文件
  • 路径：
    _knowledge_base/research-<主题>-<YYYYMMDD>.md
  • 初始内容包含：调研目标、关键问题、预期输出
markdown
undefined
  • Create the file before starting the search
  • Path: 
    _knowledge_base/research-<Topic>-<YYYYMMDD>.md
  • Initial content includes: research objective, key questions, expected output
markdown
undefined

[主题] 调研笔记

[Topic] Research Notes

调研日期：YYYY-MM-DD 调研目标：[一句话说明]
Research Date: YYYY-MM-DD Research Objective: [One-sentence description]

关键问题

Key Questions

  1. [问题1]
  2. [问题2]
  3. [问题3]
  1. [Question 1]
  2. [Question 2]
  3. [Question 3]

发现

Findings

（调研中逐步填充）
(Fill in incrementally during research)

来源列表

Source List

（每次搜索后追加）
undefined
(Append after each search)
undefined

Step 2: 搜索并增量保存

Step 2: Search and Incrementally Save

  • 每次 WebSearch 后，立即将发现追加到文件
  • 每条发现附上来源 URL 和日期
  • 遵循信息源优先级（见 SHARED-RULES.md）
  • After each WebSearch, immediately append the findings to the file
  • Attach the source URL and date to each finding
  • Follow information source priority (see SHARED-RULES.md)

Step 3: 阶段摘要

Step 3: Phase Summary

  • 每完成3次搜索，在文件中保存一次「阶段摘要」
  • 格式：
    ### 阶段摘要 (第N轮)
    + 当前关键发现
  • After completing 3 searches, save a "Phase Summary" in the file
  • Format: 
    ### Phase Summary (Round N)
    + key findings so far

Step 4: 最终简报

Step 4: Final Brief

调研结束时，整理文件为结构化简报：
markdown
undefined
At the end of the research, organize the file into a structured brief:
markdown
undefined

调研结论

Research Conclusions

关键事实

Key Facts

  1. [事实1]（来源：URL）
  2. [事实2]（来源：URL）
  1. [Fact 1] (Source: URL)
  2. [Fact 2] (Source: URL)

来源列表

Source List

来源URL发布日期可信度
.........高/中/低
SourceURLPublication DateCredibility
.........High/Medium/Low

待确认问题

Pending Verification Issues

  • [还需要进一步验证的点]
  • [Points requiring further verification]

写作建议

Writing Recommendations

  • [基于调研结果，对后续写作的建议]
undefined
  • [Recommendations for subsequent writing based on research results]
undefined

关键原则

Key Principles

  • 先建文件再搜索：确保第一次搜索结果就被保存
  • 增量保存不等到最后：每次搜索后立即追加
  • 调研和写作分离：本 Skill 只做调研，不开始写草稿
  • 标注可信度：区分一手信息（官方）和二手信息（媒体/社区）
  • 忽略过时信息源：知乎/百度（2025年前）、营销软文
  • Create file first, then search: Ensure the first search results are saved
  • Incremental saving, don't wait until the end: Append immediately after each search
  • Separate research and writing: This Skill only handles research, does not start drafting
  • Mark credibility: Distinguish primary sources (official) and secondary sources (media/community)
  • Ignore outdated sources: Zhihu/Baidu (before 2025), marketing articles

与其他 Skill 的关系

Relationship with Other Skills

  • 调研完成后，用户可触发 /选题生成 来确定写作方向
  • 调研文件将作为后续写作的输入素材
  • 如果调研中发现的信息适合长期留存，保存到对应的 _knowledge_base 分类目录
  • After research is completed, users can trigger /Topic Generation to determine writing direction
  • Research files will be used as input materials for subsequent writing
  • If information found during research is suitable for long-term retention, save it to the corresponding _knowledge_base category directory

输出位置

Output Location

  • 调研笔记：
    _knowledge_base/research-<主题>-<YYYYMMDD>.md
  • 长期知识：
    _knowledge_base/<分类>/<主题>-<YYYYMM>.md
最后更新: 2026-02-06
花叔出品 | AI Native Coder · 独立开发者 公众号「花叔」| 30万+粉丝 | AI工具与效率提升 代表作：小猫补光灯（AppStore付费榜Top1）·《一本书玩转DeepSeek》
  • Research Notes: 
    _knowledge_base/research-<Topic>-<YYYYMMDD>.md
  • Long-term Knowledge: 
    _knowledge_base/<Category>/<Topic>-<YYYYMM>.md
Last Updated: 2026-02-06
Produced by Uncle Hua | AI Native Coder · Independent Developer Official Account "Uncle Hua" | 300,000+ Followers | AI Tools & Efficiency Improvement Masterpieces: Kitten Fill Light (Top 1 in AppStore Paid Charts) · Mastering DeepSeek in One Book