ljg-travel
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
Chineseljg-travel-flow: 旅行研究
ljg-travel-flow: Travel Research
一条命令完成:全维度文化研究 → 内容提炼 → org 文档 + 便携卡片。
方法论借鉴考古学 Desk-Based Assessment(DBA):到达之前,穷尽一切文献证据。
Complete with one command: Full-dimensional cultural research → Content refinement → org document + portable reference cards.
Methodology draws on archaeological Desk-Based Assessment (DBA): Exhaust all documentary evidence before arrival.
模式
Mode
强制 NATIVE 模式。 本 workflow 是多 skill 管道(Research → ContentAnalysis → ljg-card),不走 Algorithm 七步流程。
Mandatory NATIVE Mode. This workflow is a multi-skill pipeline (Research → ContentAnalysis → ljg-card) and does not follow the Algorithm seven-step process.
参数
Parameters
| 参数 | 说明 | 示例 |
|---|---|---|
| 城市名 | 必填,目标城市 | 西安、洛阳、大同 |
| 聚焦主题(可选) | |
| 快速模式,跳过内容提炼,只做研究+文档 |
| Parameter | Description | Example |
|---|---|---|
| City Name | Required, target city | Xi'an, Luoyang, Datong |
| Focus theme (optional) | |
| Quick mode, skip content refinement, only conduct research + document generation |
执行
Execution
1. 解析参数
1. Parse Parameters
从用户消息中提取城市名称和可选参数。如有聚焦主题,后续所有搜索围绕该主题展开。
Extract the city name and optional parameters from the user's message. If there is a focus theme, all subsequent searches will revolve around this theme.
2. 全维度研究(Research extensive — 单次调用,12 个 Agent 并行)
2. Full-Dimensional Research (Research extensive — single call, 12 Agents in parallel)
调用 Skill tool 执行 ,使用 extensive 模式。
Research核心设计:不分"知识底图"和"平台发现"两步——它们是同一个研究操作的不同搜索角度。 12 个 Agent 同时出发,一半做学术/百科研究,一半做平台内容搜索。
研究提纲(传入 Research 的 prompt):
对「{城市}」进行深度文化旅行研究。这不是旅游攻略,是出发前的考古学式案头研究(Desk-Based Assessment)。
研究覆盖以下维度,每个维度需要中英文双语搜索:
**维度 A — 历史分层**
该城市经历了哪些重要历史时期?每个时期在这座城市留下了什么物质遗存?朝代更迭如何影响城市格局?
**维度 B — 博物馆重点**
该城市有哪些重要博物馆?各博物馆的镇馆之宝和核心馆藏?哪些展品有重大考古意义?必须给出具体文物名称和展厅位置。
**维度 C — 古建遗存**
现存哪些重要古建筑和遗址?营造年代、建筑形制、结构特点。哪些是全国重点文物保护单位?看建筑时应关注什么细节(斗拱、彩画、碑刻等)?
**维度 D — 考古发现**
该城市及周边有哪些重大考古发现?出土文物现藏于哪些博物馆?发掘过程中有什么重要故事?
**维度 E — 人文脉络**
与该城市相关的重要历史人物、文学作品、文化传统。帮助理解这座城市的文化性格。
**维度 F — 深度内容发现**
搜索 B站(bilibili.com)、知乎(zhihu.com)、微信公众号(mp.weixin.qq.com)、抖音(douyin.com)、小红书(xiaohongshu.com)上关于该城市博物馆和古建的深度讲解内容。
筛选标准:
- 要:有知识增量的内容(讲背景、讲工艺、讲考古过程、讲建筑细节)
- 不要:纯打卡拍照、纯推荐无分析、广告软文
- B站视频优先 10 分钟以上的讲解类
- 公众号文章优先有参考文献或明确作者身份的
返回内容标题、URL、一句话摘要。
{如有聚焦主题:特别关注与「{聚焦主题}」相关的内容,其他维度作为背景补充。}等待 Research 完成,获得全维度研究结果。
Call the Skill tool to execute using the extensive mode.
ResearchCore Design: No separation between "knowledge base mapping" and "platform discovery" — they are different search angles of the same research operation. 12 Agents start simultaneously, half conducting academic/encyclopedic research and half searching platform content.
Research Outline (prompt passed to Research):
Conduct in-depth cultural travel research on「{City}」. This is not a travel guide, but an archaeological-style desk-based assessment (Desk-Based Assessment) before departure.
The research covers the following dimensions, and each dimension requires bilingual (Chinese and English) searches:
**Dimension A — Historical Layers**
What important historical periods has this city experienced? What material remains did each period leave in this city? How did dynastic changes affect the city's layout?
**Dimension B — Museum Highlights**
What important museums are there in this city? What are the treasure pieces and core collections of each museum? Which exhibits have significant archaeological value? Must provide specific cultural relic names and exhibition hall locations.
**Dimension C — Ancient Architectural Remains**
What important ancient buildings and sites are preserved? Construction era, architectural form, structural features. Which are key national cultural relic protection units? What details should be paid attention to when viewing the buildings (brackets, colored paintings, inscriptions, etc.)?
**Dimension D — Archaeological Discoveries**
What major archaeological discoveries are there in and around the city? Where are the unearthed cultural relics now collected? What important stories are there during the excavation process?
**Dimension E — Cultural Context**
Important historical figures, literary works, and cultural traditions related to the city. Help understand the cultural character of this city.
**Dimension F — In-Depth Content Discovery**
Search for in-depth explanations about the city's museums and ancient buildings on Bilibili (bilibili.com), Zhihu (zhihu.com), WeChat Official Accounts (mp.weixin.qq.com), Douyin (douyin.com), and Xiaohongshu (xiaohongshu.com).
Selection criteria:
- Preferred: Content with knowledge increment (explaining background, craftsmanship, excavation process, architectural details)
- Not allowed: Pure check-in photos, pure recommendations without analysis, advertising soft articles
- Prioritize Bilibili videos of 10 minutes or longer that are explanatory
- Prioritize official account articles with references or clear author identities
Return content title, URL, and one-sentence summary.
{If there is a focus theme: Pay special attention to content related to「{Focus Theme}」, with other dimensions as background supplements.}Wait for Research to complete and obtain full-dimensional research results.
3. 内容提炼(ContentAnalysis — 可选)
3. Content Refinement (ContentAnalysis — Optional)
如果用户指定了 快速模式,跳过此步。
-q从步骤 2 返回的结果中,提取所有有效 URL(文章链接、视频链接)。
对每个 URL 并行启动 Agent subagent:
每个 subagent 调用 Skill tool 执行 ,传入 URL,使用 fast 深度级别,提取核心知识点。
ContentAnalysis降级规则:
- 如果 ContentAnalysis 对某个 URL 失败(无法访问、无字幕等),跳过该 URL,不阻塞
- 如果所有 URL 都失败,流程不中断——步骤 2 的研究结果已经足够生成文档
- ContentAnalysis 是增强层,不是必要层
收集所有成功提炼的内容摘要。
If the user specifies the quick mode, skip this step.
-qExtract all valid URLs (article links, video links) from the results returned in Step 2.
Launch Agent subagents in parallel for each URL:
Each subagent calls the Skill tool to execute , passing in the URL, using fast depth level, to extract core knowledge points.
ContentAnalysisDegradation Rules:
- If ContentAnalysis fails for a certain URL (unable to access, no subtitles, etc.), skip the URL without blocking the process
- If all URLs fail, the process does not interrupt — the research results from Step 2 are sufficient to generate the document
- ContentAnalysis is an enhancement layer, not a necessary layer
Collect all successfully refined content summaries.
4. 合成 org-mode 文档
4. Synthesize org-mode Document
将步骤 2(研究结果)和步骤 3(内容提炼,如有)合成为一份结构化 org-mode 文档。
文档结构:
org
#+title: {城市}旅行研究
#+date: {当前日期}
#+filetags: :travel:museum:architecture:
* 城市概览
{城市}的文明坐标——为什么值得去,去了看什么。一段话勾勒这座城市在中国文明史中的位置。
* 历史分层
** {时期1}({年代范围})
核心事件、遗留痕迹、对应可看的实物。
** {时期2}
...
* 博物馆指南
** {博物馆1名称}
地址、开放时间、预约方式(如需要)。
*** 镇馆之宝
- {文物名}:{为什么重要} | 看什么细节:{具体观察点}
*** 重点展厅
- {展厅名}:{核心看点}
*** 容易错过的
- {被忽视但值得看的内容}
** {博物馆2名称}
...
* 古建遗存
** {古建1名称}({朝代},{保护级别})
形制概述。
*** 看什么
- {具体观察点1}:{为什么值得注意}
- {具体观察点2}
** {古建2名称}
...
* 考古发现
** {遗址/发现1}
发现经过、意义、出土文物现藏地。如果有有趣的发掘故事,讲出来。
* 参观路线
** 路线一:{主题名}({预计时间})
适合谁:{描述}
1. {地点} → 重点看 {什么}({停留建议时间})
2. ...
** 路线二:{主题名}
...
* 深度内容推荐
从各平台发现的值得在出发前看的内容。
** 视频
- [[{URL}][{标题}]] — {一句话摘要}
** 文章
- [[{URL}][{标题}]] — {一句话摘要}
** 书籍(如有推荐)
- {书名} — {为什么值得读}文件命名:使用 denote naming schema,保存到 目录:
~/Documents/notes/{YYYYMMDDTHHMMSS}==z--{城市}旅行研究.org写作要求:
- 每个推荐必须有「为什么看」和「看什么细节」,不许空泛
- 语气是给自己写的笔记,不是导游词
- 有确切信息写确切的,没有的不编
Combine the results from Step 2 (research results) and Step 3 (content refinement, if any) into a structured org-mode document.
Document Structure:
org
#+title: {City} Travel Research
#+date: {Current Date}
#+filetags: :travel:museum:architecture:
* City Overview
The civilization coordinates of {City} — why it is worth visiting and what to see there. A paragraph outlining the city's position in Chinese civilization history.
* Historical Layers
** {Period 1} ({Time Range})
Core events, remaining traces, corresponding visible artifacts.
** {Period 2}
...
* Museum Guide
** {Museum 1 Name}
Address, opening hours, reservation method (if required).
*** Treasure Pieces
- {Cultural Relic Name}: {Why it is important} | Key details to observe: {Specific Observation Points}
*** Key Exhibition Halls
- {Exhibition Hall Name}: {Core Highlights}
*** Easily Missed Spots
- {Neglected but Worthwhile Content}
** {Museum 2 Name}
...
* Ancient Architectural Remains
** {Ancient Building 1 Name} ({Dynasty}, {Protection Level})
Overview of its form.
*** Key Observation Points
- {Observation Point 1}: {Why it is noteworthy}
- {Observation Point 2}
** {Ancient Building 2 Name}
...
* Archaeological Discoveries
** {Site/Discovery 1}
Excavation process, significance, current collection location of unearthed relics. If there are interesting excavation stories, include them.
* Visiting Routes
** Route 1: {Theme Name} ({Estimated Time})
Suitable for: {Description}
1. {Location} → Focus on {What to See} ({Suggested Stay Time})
2. ...
** Route 2: {Theme Name}
...
* In-Depth Content Recommendations
Worthwhile content to review before departure, discovered from various platforms.
** Videos
- [[{URL}][{Title}]] — {One-Sentence Summary}
** Articles
- [[{URL}][{Title}]] — {One-Sentence Summary}
** Books (if recommended)
- {Book Title} — {Why it is worth reading}File Naming: Use the denote naming schema and save to the directory:
~/Documents/notes/{YYYYMMDDTHHMMSS}==z--{City}TravelResearch.orgWriting Requirements:
- Each recommendation must include "why to visit" and "what details to observe"; no vague descriptions allowed
- The tone should be like notes written for oneself, not a tour guide's script
- Include exact information where available; leave blank rather than fabricate if information is uncertain
5. 铸造便携卡片(ljg-card)
5. Generate Portable Cards (ljg-card)
从步骤 4 的 org 文档中提取核心内容,铸造两张卡片,并行执行:
卡片 A — 城市文明概览(信息图):
调用 Skill tool 执行 ,输入内容为:城市历史分层 + 核心博物馆清单 + 必看古建清单的精华摘要。一张图看懂这座城市的文明骨架。
ljg-card -i卡片 B — 参观路线速查(长图):
调用 Skill tool 执行 ,输入内容为:参观路线建议 + 每个地点的核心看点。手机上随时查看。
ljg-card -lExtract core content from the org-mode document in Step 4 and generate two cards, executed in parallel:
Card A — City Civilization Overview (Infographic):
Call the Skill tool to execute , with input content being the condensed summary of the city's historical layers + core museum list + must-see ancient building list. Understand the city's civilization framework at a glance with one image.
ljg-card -iCard B — Visiting Route Quick Reference (Long Image):
Call the Skill tool to execute , with input content being the visiting route suggestions + core highlights of each location. Accessible anytime on mobile phones.
ljg-card -l6. 汇总报告
6. Summary Report
════ 旅行研究完成 ═══════════════════════
🏛️ 城市: {城市名}
📝 知识文档: {org 文件路径}
🖼️ 文明概览卡: {PNG 文件路径}
🖼️ 路线速查卡: {PNG 文件路径}
📊 研究覆盖: {N}个博物馆 | {M}座古建 | {K}处考古遗址
📎 深度内容: {X}个视频 | {Y}篇文章════ Travel Research Completed ═══════════════════════
🏛️ City: {City Name}
📝 Knowledge Document: {org File Path}
🖼️ Civilization Overview Card: {PNG File Path}
🖼️ Route Quick Reference Card: {PNG File Path}
📊 Research Coverage: {N} Museums | {M} Ancient Buildings | {K} Archaeological Sites
📎 In-Depth Content: {X} Videos | {Y} Articles关键约束
Key Constraints
- 步骤 2 是核心——12 个 Agent 并行覆盖学术研究和平台内容,一次完成
- 步骤 3(内容提炼)是增强层,失败不阻塞流程
- 步骤 5 的两张卡片之间并行
- org 文档是主产出,卡片是衍生产出——文档质量优先
- 不产出泛泛的旅游攻略,每个推荐必须有「为什么看」和「看什么细节」
- Research 搜索使用中英文双语关键词,扩大覆盖面
- 没有确切信息时宁可留空,不编造
- Step 2 is the core — 12 Agents in parallel cover academic research and platform content, completed in one go
- Step 3 (content refinement) is an enhancement layer; failure does not block the process
- The two cards in Step 5 are executed in parallel
- The org document is the main output, and the cards are derivative outputs — document quality takes priority
- No generic travel guides are produced; each recommendation must include "why to visit" and "what details to observe"
- Research searches use bilingual (Chinese and English) keywords to expand coverage
- Leave blank rather than fabricate when exact information is unavailable
Known Pitfalls
Known Pitfalls
(首次创建,暂无记录。使用中积累。)
(First creation, no records yet. Accumulate during use.)