ultra-research
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
Chinese深度研究 Skill
Deep Research Skill
三段式：理解需求 → 迭代探索 → 整合报告。
通过 Playwright MCP 控制多个 AI 服务进行并行研究。各服务交互细节见 reference 文件。
Three-stage Process: Understand Requirements → Iterative Exploration → Integrate Report.
Control multiple AI services for parallel research via Playwright MCP. Details of interactions with each service can be found in reference files.
前置条件
Prerequisites
- Playwright MCP 需以持久化浏览器模式运行（），这样各 AI 服务的登录态才会保留
--user-data-dir - 用户需要预先在 Playwright 浏览器中登录要使用的 AI 服务（Gemini、ChatGPT、Claude、Grok）——不需要全部登录，登录了哪些就用哪些
- 本 skill 的选择器基于中文 UI 环境；英文 UI 下按钮名和 placeholder 会不同，需用 实时探测
browser_snapshot
- Playwright MCP must run in persistent browser mode () so that the login states of each AI service are retained
--user-data-dir - Users need to log in to the desired AI services (Gemini, ChatGPT, Claude, Grok) in the Playwright browser in advance — you don't need to log in to all, only use the ones you've logged into
- The selectors of this skill are based on Chinese UI environment; button names and placeholders will differ in English UI, so real-time detection via is required
browser_snapshot
可用的 AI 服务与模式
Available AI Services and Modes
| 服务 | 模式 | 耗时 | 特点 |
|---|---|---|---|
| Gemini Pro | 快速 | 10-60s | 快速问答、事实查证（有搜索能力） |
| Gemini Deep Research | 深度 | 5-10min | 多源综合、广度覆盖 |
| ChatGPT Thinking | 快速 | 10-120s | 深度推理、结构化分析 |
| ChatGPT Deep Research | 深度 | 5-30min | 多源综合、广度覆盖 |
| Claude Opus Extended | 快速 | 10-120s | 深度推理、方案设计 |
| Grok 快速/专家 | 快速 | 10-60s | X/Twitter 内容搜索、时效性讨论和趋势 |
| Grok DeepSearch | 深度 | 数分钟 | 深度搜索、社交媒体观点汇集 |
这些是工具箱，不是流水线。 根据研究需要灵活选用和组合。
交互参考：
- references/gemini.md — Gemini 操作、选择器、提取
- references/chatgpt.md — ChatGPT 操作（最脆弱的服务，注意防御）
- references/claude-web.md — Claude 操作、选择器、提取
- references/grok.md — Grok 操作（X/Twitter 内容搜索的独特入口）
- references/extraction.md — 通用提取技术（分块、引用链接、保存）
- references/prompt-guide.md — 研究 prompt 构造建议
| Service | Mode | Time Cost | Features |
|---|---|---|---|
| Gemini Pro | Fast | 10-60s | Quick Q&A, fact-checking (with search capability) |
| Gemini Deep Research | Deep | 5-10min | Multi-source synthesis, broad coverage |
| ChatGPT Thinking | Fast | 10-120s | In-depth reasoning, structured analysis |
| ChatGPT Deep Research | Deep | 5-30min | Multi-source synthesis, broad coverage |
| Claude Opus Extended | Fast | 10-120s | In-depth reasoning, solution design |
| Grok Fast/Expert | Fast | 10-60s | X/Twitter Content Search, real-time discussions and trends |
| Grok DeepSearch | Deep | Several minutes | In-depth search, social media opinion aggregation |
These are toolboxes, not assembly lines. Flexibly select and combine them based on research needs.
Interaction references:
- references/gemini.md — Gemini operations, selectors, extraction
- references/chatgpt.md — ChatGPT operations (the most fragile service, pay attention to defense)
- references/claude-web.md — Claude operations, selectors, extraction
- references/grok.md — Grok operations (unique entry for X/Twitter content search)
- references/extraction.md — General extraction techniques (chunking, reference links, saving)
- references/prompt-guide.md — Suggestions for constructing research prompts
Part 1: 理解需求
Part 1: Understand Requirements
目标：搞清楚用户要什么，然后开始。
Goal: Clarify what the user needs, then get started.
先判断：用户的需求清楚吗？
First Judgment: Is the User's Requirement Clear?
收到用户 query 后，先判断你是否能理解研究的对象、范围、目的：
- 清楚 → 直接向用户确认你的理解（如需微调范围），然后开始
- 大致清楚但有细节模糊 → 先 WebSearch 搜 2-3 次补充认知，再和用户确认
- 不清楚 → 先 WebSearch 建立领域认知，再带着具体问题向用户澄清
澄清不是走流程——如果用户的意图一目了然，直接干活比问一堆问题有用。
After receiving the user's query, first judge whether you can understand the object, scope, and purpose of the research:
- Clear → Directly confirm your understanding with the user (if scope adjustment is needed), then start
- Generally clear but with ambiguous details → First perform 2-3 WebSearch to supplement your cognition, then confirm with the user
- Unclear → First perform WebSearch to establish domain cognition, then clarify with specific questions to the user
Clarification is not a formality — if the user's intention is obvious, taking action directly is more useful than asking a bunch of questions.
构造查询
Construct Queries
为选定的 AI 模式各准备一条查询。不同 AI 给不同角度比给同一个 prompt 更有效。见 prompt-guide.md。
Prepare a query for each selected AI mode. Providing different perspectives to different AIs is more effective than using the same prompt for all. See prompt-guide.md.
创建研究目录
Create Research Directory
在合适位置创建研究目录（如 ），创建 todo.md 记录状态。
research-主题/Create a research directory at an appropriate location (e.g., ), and create todo.md to record the status.
research-topic/Part 2: 迭代探索
Part 2: Iterative Exploration
目标：多路并发搜索，逐步深入，直到信息充分。
Goal: Conduct multi-channel concurrent searches, gradually deepen, until sufficient information is obtained.
深度由进展决定
Depth is Determined by Progress
不要预设研究深度，让研究过程自己说话：
- 起步：先用快速模式铺开（Gemini Pro、ChatGPT Thinking、Claude Opus、Grok 专家），快速建立基线认知
- 发现复杂度：如果快速模式的结果显示话题确实多维、有矛盾、需要更多来源 → 追加 Deep Research / DeepSearch
- 收敛：如果快速模式已经给出了一致的、充分的答案 → 不需要开 DR，直接整合
简单问题不要大炮打蚊子。复杂问题不要浅尝辄止。让信息本身告诉你该挖多深。
Do not preset research depth; let the research process speak for itself:
- Start : First use fast modes to lay the groundwork (Gemini Pro, ChatGPT Thinking, Claude Opus, Grok Expert) to quickly establish baseline cognition
- Discover Complexity : If results from fast modes show the topic is indeed multi-dimensional, contradictory, or requires more sources → Add Deep Research / DeepSearch
- Converge : If fast modes have provided consistent and sufficient answers → No need to enable DR, directly integrate
Do not use a sledgehammer to crack a nut for simple questions. Do not stop at the surface for complex questions. Let the information itself tell you how deep to dig.
派发
Dispatch
每个 AI 服务的具体操作步骤见各自 reference 文件。通用流程：
- 新开 Tab → 导航到服务 → 检查登录
- 选择模式 → 输入查询
- 记录到 todo.md（Tab 编号 + 查询内容）
- 发完一个立刻切下一个，不等结果
登录处理：这些服务使用的是用户的账号。如果发现某个服务未登录，立即暂停并告知用户，让用户选择：
- 去 Playwright 浏览器中登录该服务，然后继续
- 跳过该服务，用其他已登录的服务继续
不要自动跳过——用户可能想登录。
Specific operation steps for each AI service can be found in their respective reference files. General process:
- Open a new Tab → Navigate to the service → Check login status
- Select mode → Enter query
- Record to todo.md (Tab number + query content)
- Switch to the next one immediately after sending, do not wait for results
Login Handling : These services use the user's account. If a service is found to be unlogged, immediately pause and inform the user, allowing the user to choose:
- Log in to the service in the Playwright browser, then continue
- Skip this service and continue with other logged-in services
Do not skip automatically — the user may want to log in.
轮询与提取
Polling and Extraction
先到先提取，不等所有流完成。
- 快速模式：~30s 后开始检查
- DR/DeepSearch 模式：~90s 后开始检查
- 用 screenshot 判断是否完成（比 snapshot 可靠）——各服务完成标志见 reference
提取技术见 extraction.md。每份结果保存为独立 md（、...）。
01-gemini-pro.md02-chatgpt-thinking.mdExtract as soon as results are available, do not wait for all streams to complete.
- Fast modes: Start checking after ~30s
- DR/DeepSearch modes: Start checking after ~90s
- Use screenshots to judge completion (more reliable than snapshot) — completion markers for each service can be found in reference files
Extraction techniques can be found in extraction.md. Save each result as an independent md file (e.g., , ...).
01-gemini-pro.md02-chatgpt-thinking.md定向深入
Targeted Deepening
每提取一份结果，立即评估：
- 发现了哪些子问题？
- 哪些维度数据不够或有矛盾？
- 有没有意外发现值得追查？
如果有 → 追加查询。选择最合适的工具：
- 快速模式 — 定向查证具体子问题（秒级）
- WebSearch/WebFetch — 交叉验证事实（比开新 AI Tab 更快）
- DR/DeepSearch — 仅用于发现的子问题确实需要广度覆盖时
- Grok — 特别适合查看某话题在 X/Twitter 上的讨论和观点
Evaluate immediately after extracting each result:
- What sub-issues have been discovered?
- Which dimensions lack data or have contradictions?
- Are there any unexpected findings worth investigating?
If yes → Add supplementary queries. Choose the most suitable tool:
- Fast modes — Targeted verification of specific sub-issues (seconds-level)
- WebSearch/WebFetch — Cross-validate facts (faster than opening a new AI Tab)
- DR/DeepSearch — Only use when the discovered sub-issues truly require broad coverage
- Grok — Especially suitable for viewing discussions and opinions on a topic on X/Twitter
Tab 管理
Tab Management
- 控制在 ≤8 个活跃 Tab，多了浏览器会不稳定
- 提取完的 Tab 可以保留（可能需要追问），但不要无限堆积
- 浏览器断连时：→
browser_close→pkill -f "mcp-chrome"→ 重新 navigatesleep 2
- Keep active Tabs ≤8; too many will make the browser unstable
- Tabs that have been extracted can be retained (may need follow-up questions), but do not accumulate indefinitely
- If the browser disconnects: →
browser_close→pkill -f "mcp-chrome"→ Re-navigatesleep 2
何时停止
When to Stop
- 核心问题有了交叉验证的答案
- 新搜索不再产出实质性新信息
- 用户要求的范围已充分覆盖
- Core questions have cross-validated answers
- New searches no longer produce substantial new information
- The scope required by the user has been fully covered
Part 3: 整合报告
Part 3: Integrate Report
目标：按主题维度重组内容，交叉验证，输出一份深度报告。
Goal: Restructure content by topic dimensions, cross-validate, and output an in-depth report.
原则
Principles
- 按主题组织，不按来源罗列（不要 "Gemini 说... ChatGPT 说..."）
- 标注来源（、
[Gemini DR]、[Grok]等），让读者知道可信度[ChatGPT Thinking] - 共识（多路一致）和分歧（各执一词）都要标注
- 保留引用 URL — 原始网络来源链接是高价值信息，必须传递到最终报告
- 分歧不强行裁决 — 保留并标注，让用户判断
- 信息权重：DR 通常更全面，快速模式更聚焦，WebSearch 最即时，Grok 的 X/Twitter 链接是独家信息
- Organize by topic, not by source (do not use "Gemini said... ChatGPT said...")
- Label sources (e.g., ,
[Gemini DR],[Grok]) so readers can judge credibility[ChatGPT Thinking] - Both consensus (consistent across multiple channels) and disagreements (conflicting views) must be labeled
- Retain citation URLs — Original web source links are high-value information and must be included in the final report
- Do not force resolution of disagreements — Retain and label them, letting the user judge
- Information weight: DR is usually more comprehensive, fast modes are more focused, WebSearch is the most real-time, and Grok's X/Twitter links are exclusive information
引用体系
Citation System
报告有两层引用，都要做好：
1. 网络来源引用（报告 → 原始网页）
正文中关键事实、数据、观点旁标注原始网络来源。格式灵活——脚注、行内链接、段落末尾引用块都行，根据报告风格选择最可读的方式。核心原则：读者看到一个关键论断，能直接找到原始网页验证。
各 AI 的引用链接已在原始结果文件的 章节。整合时从那里挑选最相关的链接嵌入报告正文。不需要全部搬过来——只保留支撑关键论点的。
## 参考链接2. AI 来源标注（报告 → 哪个 AI 说的）
用行内标签 、、 等标注信息来自哪个 AI 模式。这帮助读者判断可信度（DR 多源交叉 vs 快速模式单次推理）。多路一致的可以标 。
[Gemini DR][Grok][ChatGPT Thinking][多路共识]3. 原始结果索引（报告 → 原始文件）
报告末尾附 ，列出所有原始结果文件和对应的查询内容：
## 原始结果索引undefinedThe report has two layers of citations, both of which must be properly handled:
1. Web Source Citations (Report → Original Webpage)
Label the original web source next to key facts, data, and viewpoints in the text. The format is flexible — footnotes, inline links, citation blocks at the end of paragraphs are all acceptable; choose the most readable way based on the report style. Core principle: When readers see a key assertion, they can directly find the original webpage for verification.
Citation links for each AI are in the section of the original result files. When integrating, select the most relevant links from there to embed in the report text. No need to move all over — only retain those that support key arguments.
## Reference Links2. AI Source Labeling (Report → Which AI said it)
Use inline tags like , , to label which AI mode the information comes from. This helps readers judge credibility (multi-source cross-validation in DR vs single inference in fast modes). Consensus across multiple channels can be labeled .
[Gemini DR][Grok][ChatGPT Thinking][Multi-channel Consensus]3. Original Result Index (Report → Original Files)
Attach at the end of the report, listing all original result files and their corresponding query content:
## Original Result Indexundefined原始结果索引
Original Result Index
| 文件 | 服务 | 模式 | 查询摘要 |
|---|---|---|---|
| 01-gemini-pro.md | Gemini | Pro | ... |
| 02-chatgpt-dr.md | ChatGPT | Deep Research | ... |
这让读者能回溯到完整的原始结果（含全部参考链接）。| File | Service | Mode | Query Summary |
|---|---|---|---|
| 01-gemini-pro.md | Gemini | Pro | ... |
| 02-chatgpt-dr.md | ChatGPT | Deep Research | ... |
This allows readers to trace back to the complete original results (including all reference links).流程
Process
- 写初稿 — 基于已有结果先写能写的，同步嵌入引用链接
- 标记缺口 — 哪些数据不够？哪些只有单一来源？
- 缺口重大 → 回到 Part 2 补充
- 逐步更新 — 新信息到手就更新报告
- 最终审校 — 检查引用完整性（关键论断有没有来源支撑？）→ 交付用户
报告保存为 。
00-final-report.md- Write Draft — First write what you can based on existing results, and embed citation links simultaneously
- Mark Gaps — Which data is insufficient? Which has only a single source?
- Significant Gaps → Return to Part 2 to supplement
- Gradually Update — Update the report as new information is obtained
- Final Review — Check citation integrity (Are key assertions supported by sources?) → Deliver to user
Save the report as .
00-final-report.mdtodo.md 持久化
todo.md Persistence
研究可能跨会话。todo.md 是恢复点。
markdown
undefinedResearch may span sessions. todo.md is the recovery point.
markdown
undefined研究：[主题]
Research: [Topic]
状态：[进行中/已完成]
Status: [In Progress/Completed]
搜索流
Search Streams
| # | 服务 | 模式 | Tab | 查询摘要 | 状态 |
|---|---|---|---|---|---|
| 01 | Gemini | Pro | 2 | ... | 已提取 |
| 02 | ChatGPT | DR | 3 | ... | 等待中 |
| 03 | Grok | 专家 | 4 | ... | 进行中 |
| # | Service | Mode | Tab | Query Summary | Status |
|---|---|---|---|---|---|
| 01 | Gemini | Pro | 2 | ... | Extracted |
| 02 | ChatGPT | DR | 3 | ... | Pending |
| 03 | Grok | Expert | 4 | ... | In Progress |
发现的子问题
Discovered Sub-issues
- ...
- ...
整合进度
Integration Progress
- 初稿
- 缺口填补
- 最终审校
**跨会话恢复**：读 todo.md → `browser_tabs` list 检查现有 Tab → 对照恢复。
---- Draft
- Gap Filling
- Final Review
**Cross-session Recovery** : Read todo.md → Check existing Tabs via `browser_tabs` list → Restore accordingly.
---错误处理
Error Handling
| 情况 | 建议 |
|---|---|
| 未登录 | 暂停，告知用户——让用户选择登录还是跳过该服务 |
| 浏览器断连 | |
| Cloudflare 拦截 | 重试 1-2 次 navigate，通常第二次就通过 |
| 单 Tab 超时（>40min） | 标记超时，继续其他流 |
| 提取返回空 | 见 extraction.md 的降级策略 |
| snapshot 返回空 | 等 2-3 秒重试，ChatGPT/Grok 有时会进入瞬时空状态 |
| DR 未启动 | ChatGPT DR 可能先回澄清问题，检测后回复让它直接开始 |
| DR/DeepSearch 配额用尽 | 跳过，改用快速模式或 WebSearch 补充 |
部分失败不影响产出：≥2 路成功仍出报告（注明来源范围）。
| Scenario | Suggestion |
|---|---|
| Not Logged In | Pause and inform the user — Let the user choose to log in or skip the service |
| Browser Disconnection | |
| Cloudflare Interception | Retry navigation 1-2 times; usually passes on the second try |
| Single Tab Timeout (>40min) | Mark as timeout, continue with other streams |
| Extraction Returns Empty | See the downgrade strategy in extraction.md |
| snapshot Returns Empty | Retry after 2-3 seconds; ChatGPT/Grok sometimes enter a transient empty state |
| DR Not Started | ChatGPT DR may first send clarification questions; detect and reply to let it start directly |
| DR/DeepSearch Quota Exhausted | Skip, use fast modes or WebSearch to supplement |
Partial failure does not affect output : A report can still be generated if ≥2 streams succeed (note the source scope).