deep-research

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Deep Research Skill

深度研究 Skill

Overview

概述

This skill provides a systematic methodology for conducting thorough web research. Load this skill BEFORE starting any content generation task to ensure you gather sufficient information from multiple angles, depths, and sources.

此 skill 提供进行深入网络研究的系统化方法。在开始任何内容生成任务之前加载此 skill，确保从多个角度、深度和来源收集足够的信息。

When to Use This Skill

何时使用此 Skill

Always load this skill when:

Do not use this skill when a more specialized research skill clearly fits, especially

github-deep-research

openai-docs

, or

codebase-deep-research

始终在以下情况下加载此 skill：

当更专业的研究 skill 明显更适用时，不要使用此 skill，尤其是
```
github-deep-research
```
、
```
openai-docs
```
或
```
codebase-deep-research
```
。

Research Questions

研究问题

User asks "what is X", "explain X", "research X", "investigate X"
User wants to understand a concept, technology, or topic in depth
The question requires current, comprehensive information from multiple sources
A single web search would be insufficient to answer properly

用户询问“X 是什么”“解释 X”“研究 X”“调查 X”
用户希望深入理解某个概念、技术或主题
问题需要从多个来源获取当前、全面且完整的信息
单次网络搜索不足以正确回答问题

Content Generation (Pre-research)

内容生成（研究前）

Creating presentations (PPT/slides)
Creating frontend designs or UI mockups
Writing articles, reports, or documentation
Producing videos or multimedia content
Any content that requires real-world information, examples, or current data

创建演示文稿（PPT/slides）
创建前端设计或 UI 模型
撰写文章、报告或文档
制作视频或多媒体内容
任何需要现实信息、示例或当前数据的内容

Core Principle

核心原则

Never generate content based solely on general knowledge. The quality of your output directly depends on the quality and quantity of research conducted beforehand. A single search query is NEVER enough.

绝不只基于通用知识生成内容。 输出质量直接取决于事先开展的研究质量。单个搜索查询永远不足以完成研究。

Research Methodology

研究方法

Phase 1: Broad Exploration

阶段 1：广泛探索

Start with broad searches to understand the landscape:

Initial Survey: Search for the main topic to understand the overall context
Identify Dimensions: From initial results, identify key subtopics, themes, angles, or aspects that need deeper exploration
Map the Territory: Note different perspectives, stakeholders, or viewpoints that exist

Example:

Topic: "AI in healthcare"
Initial searches:
- "AI healthcare applications 2024"
- "artificial intelligence medical diagnosis"
- "healthcare AI market trends"

Identified dimensions:
- Diagnostic AI (radiology, pathology)
- Treatment recommendation systems
- Administrative automation
- Patient monitoring
- Regulatory landscape
- Ethical considerations

首先进行广泛搜索，以了解整体情况：

初步调查：搜索主要主题，了解整体背景
识别维度：从初步结果中识别需要更深入探索的关键子主题、主题、角度或方面
绘制领域地图：记录存在的不同观点、利益相关者或视角

示例：

Topic: "AI in healthcare"
Initial searches:
- "AI healthcare applications 2024"
- "artificial intelligence medical diagnosis"
- "healthcare AI market trends"

Identified dimensions:
- Diagnostic AI (radiology, pathology)
- Treatment recommendation systems
- Administrative automation
- Patient monitoring
- Regulatory landscape
- Ethical considerations

Phase 2: Deep Dive

阶段 2：深入研究

For each important dimension identified, conduct targeted research:

Specific Queries: Search with precise keywords for each subtopic
Multiple Phrasings: Try different keyword combinations and phrasings
Fetch Full Content: Use
```
web_fetch
```
to read important sources in full, not just snippets
Follow References: When sources mention other important resources, search for those too

Example:

Dimension: "Diagnostic AI in radiology"
Targeted searches:
- "AI radiology FDA approved systems"
- "chest X-ray AI detection accuracy"
- "radiology AI clinical trials results"

Then fetch and read:
- Key research papers or summaries
- Industry reports
- Real-world case studies

针对识别出的每个重要维度，开展有针对性的研究：

具体查询：使用精准关键词搜索每个子主题
多种表述方式：尝试不同的关键词组合和表达方式
获取完整内容：使用
```
web_fetch
```
完整阅读重要来源，而不只是阅读摘要
追踪引用：当来源提到其他重要资源时，也搜索这些资源

示例：

Dimension: "Diagnostic AI in radiology"
Targeted searches:
- "AI radiology FDA approved systems"
- "chest X-ray AI detection accuracy"
- "radiology AI clinical trials results"

Then fetch and read:
- Key research papers or summaries
- Industry reports
- Real-world case studies

Phase 3: Diversity & Validation

阶段 3：多样性与验证

Ensure comprehensive coverage by seeking diverse information types:

Information Type	Purpose	Example Searches
Facts & Data	Concrete evidence	"statistics", "data", "numbers", "market size"
Examples & Cases	Real-world applications	"case study", "example", "implementation"
Expert Opinions	Authority perspectives	"expert analysis", "interview", "commentary"
Trends & Predictions	Future direction	"trends 2024", "forecast", "future of"
Comparisons	Context and alternatives	"vs", "comparison", "alternatives"
Challenges & Criticisms	Balanced view	"challenges", "limitations", "criticism"

通过寻找不同类型的信息，确保全面的覆盖：

信息类型	目的	示例搜索
事实与数据	具体证据	"statistics", "data", "numbers", "market size"
示例与案例	现实应用	"case study", "example", "implementation"
专家观点	权威视角	"expert analysis", "interview", "commentary"
趋势与预测	未来方向	"trends 2024", "forecast", "future of"
比较	背景与替代方案	"vs", "comparison", "alternatives"
挑战与批评	平衡观点	"challenges", "limitations", "criticism"

Phase 4: Synthesis Check

阶段 4：综合检查

Search Strategy Tips

搜索策略提示

Effective Query Patterns

有效的查询模式

undefined

undefined

Be specific with context

❌ "AI trends" ✅ "enterprise AI adoption trends 2024"

Include authoritative source hints

"[topic] research paper" "[topic] McKinsey report" "[topic] industry analysis"

Search for specific content types

"[topic] case study" "[topic] statistics" "[topic] expert interview"

Use temporal qualifiers — always use the ACTUAL current year from <current_date>

"[topic] 2026" # ← replace with real current year, never hardcode a past year "[topic] latest" "[topic] recent developments"

undefined

"[topic] 2026" # ← replace with real current year, never hardcode a past year "[topic] latest" "[topic] recent developments"

undefined

Temporal Awareness

时间感知

Always check
<current_date>
in your context before forming ANY search query.

<current_date>

gives you the full date: year, month, day, and weekday (e.g.

2026-02-28, Saturday

). Use the right level of precision depending on what the user is asking:

User intent	Temporal precision needed	Example query
"today / this morning / just released"	Month + Day	`"tech news February 28 2026"`
"this week"	Week range	`"technology releases week of Feb 24 2026"`
"recently / latest / new"	Month	`"AI breakthroughs February 2026"`
"this year / trends"	Year	`"software trends 2026"`

Rules:

When the user asks about "today" or "just released", use month + day + year in your search queries to get same-day results
Never drop to year-only when day-level precision is needed —
```
"tech news 2026"
```
will NOT surface today's news
Try multiple phrasings: numeric form (
```
2026-02-28
```
), written form (
```
February 28 2026
```
), and relative terms (
```
today
```
,
```
this week
```
) across different queries

❌ User asks "what's new in tech today" → searching

"new technology 2026"

→ misses today's news ✅ User asks "what's new in tech today" → searching

"new technology February 28 2026"

"tech news today Feb 28"

→ gets today's results

在构建任何搜索查询之前，始终检查上下文中的
<current_date>
。

<current_date>

包含完整日期：年、月、日和星期（例如

2026-02-28, Saturday

）。根据用户需求使用适当的时间精度：

用户意图	所需时间精度	示例查询
“今天 / 今天早上 / 刚发布”	月 + 日	`"tech news February 28 2026"`
“本周”	周范围	`"technology releases week of Feb 24 2026"`
“最近 / 最新 / 新发布”	月	`"AI breakthroughs February 2026"`
“今年 / 趋势”	年	`"software trends 2026"`

规则：

当用户询问“今天”或“刚发布”时，在搜索查询中使用月 + 日 + 年，以获得同一天的结果
当需要精确到日时，不要仅使用年份；
```
"tech news 2026"
```
无法返回当天的新闻
尝试多种表述方式：数字形式（
```
2026-02-28
```
）、文字形式（
```
February 28 2026
```
）和相对时间（
```
today
```
、
```
this week
```
），并结合使用不同查询

❌ 用户询问“今天科技领域有什么新动态” → 搜索

"new technology 2026"

→ 遗漏当天的新闻 ✅ 用户询问“今天科技领域有什么新动态” → 搜索

"new technology February 28 2026"

"tech news today Feb 28"

→ 获取当天结果

When to Use web_fetch

何时使用 web_fetch

Use

web_fetch

to read full content when:

A search result looks highly relevant and authoritative
You need detailed information beyond the snippet
The source contains data, case studies, or expert analysis
You want to understand the full context of a finding

在以下情况下使用

web_fetch

阅读完整内容：

搜索结果看起来高度相关且来源权威
需要超出摘要范围的详细信息
来源包含数据、案例研究或专家分析
希望理解某项发现的完整背景

Iterative Refinement

迭代优化

Research is iterative. After initial searches:

Review what you've learned
Identify gaps in your understanding
Formulate new, more targeted queries
Repeat until you have comprehensive coverage

研究是一个迭代过程。完成初步搜索后：

回顾已经了解的信息
识别理解上的空白
制定新的、更具针对性的查询
重复上述过程，直到获得全面覆盖

Quality Bar

质量标准

Your research is sufficient when you can confidently answer:

What are the key facts and data points?
What are 2-3 concrete real-world examples?
What do experts say about this topic?
What are the current trends and future directions?
What are the challenges or limitations?
What makes this topic relevant or important now?

当你能够自信地回答以下问题时，说明研究已经充分：

关键事实和数据要点是什么？
有哪些 2-3 个具体的现实案例？
专家对这一主题有什么看法？
当前趋势和未来发展方向是什么？
存在哪些挑战或局限？
为什么这一主题在当前具有相关性和重要性？

Common Mistakes to Avoid

需要避免的常见错误

❌ Stopping after 1-2 searches
❌ Relying on search snippets without reading full sources
❌ Searching only one aspect of a multi-faceted topic
❌ Ignoring contradicting viewpoints or challenges
❌ Using outdated information when current data exists
❌ Starting content generation before research is complete

❌ 仅进行 1-2 次搜索便停止
❌ 只依赖搜索摘要而不阅读完整来源
❌ 只搜索多面主题的其中一个方面
❌ 忽视相反观点或挑战
❌ 在已有最新数据时仍使用过时信息
❌ 在研究完成前开始生成内容

Output

输出

After completing research, you should have:

A comprehensive understanding of the topic from multiple angles
Specific facts, data points, and statistics
Real-world examples and case studies
Expert perspectives and authoritative sources
Current trends and relevant context

Only then proceed to content generation, using the gathered information to create high-quality, well-informed content.

完成研究后，应具备：

从多个角度对主题形成的全面理解
具体事实、数据要点和统计数据
现实示例和案例研究
专家观点和权威来源
当前趋势及相关背景

只有在完成上述工作后，才能继续生成内容，利用收集到的信息创作高质量、有充分依据的内容。