One command does it all: all-dimension cultural research → content distillation → org document + portable card.
Methodology draws on archaeological Desk-Based Assessment (DBA): before arrival, exhaust all documentary evidence.
Mode
Forced NATIVE mode. This workflow is a multi-skill pipeline (Research → ContentAnalysis → ljg-card), not the Algorithm seven-step process.
Arguments
| Argument | Description | Example |
|---|---|---|
| City name | Required, target city | Xi'an, Luoyang, Datong |
-f |
Focus topic (optional) | -f 唐代 -f 石窟 -f 青銅器 |
-q |
Quick mode, skips content distillation, only does research + document |
Execution
1. Parse arguments
Extract the city name and optional arguments from the user's message. If there is a focus topic, all subsequent searches center on that topic.
2. All-dimension research (Research extensive — single call, 12 Agents in parallel)
Call the Skill tool to execute Research, using extensive mode.
Core design: not split into "knowledge base map" and "platform discovery" two steps — they are different search angles of the same research operation. 12 Agents set out simultaneously, half doing academic/encyclopedic research, half doing platform content search.
Research outline (prompt passed to Research):
對「{城市}」進行深度文化旅行研究。這不是旅遊攻略,是出發前的考古學式案頭研究(Desk-Based Assessment)。
研究覆蓋以下維度,每個維度需要中英文雙語搜尋:
**維度 A — 歷史分層**
該城市經歷了哪些重要歷史時期?每個時期在這座城市留下了什麼物質遺存?朝代更迭如何影響城市格局?
**維度 B — 博物館重點**
該城市有哪些重要博物館?各博物館的鎮館之寶和核心館藏?哪些展品有重大考古意義?必須給出具體文物名稱和展廳位置。
**維度 C — 古建遺存**
現存哪些重要古建築和遺址?營造年代、建築形制、結構特點。哪些是全國重點文物保護單位?看建築時應關注什麼細節(斗拱、彩畫、碑刻等)?
**維度 D — 考古發現**
該城市及周邊有哪些重大考古發現?出土文物現藏於哪些博物館?發掘過程中有什麼重要故事?
**維度 E — 人文脈絡**
與該城市相關的重要歷史人物、文學作品、文化傳統。幫助理解這座城市的文化性格。
**維度 F — 深度內容發現**
搜尋 B站(bilibili.com)、知乎(zhihu.com)、微信公眾號(mp.weixin.qq.com)、抖音(douyin.com)、小紅書(xiaohongshu.com)上關於該城市博物館和古建的深度講解內容。
篩選標準:
- 要:有知識增量的內容(講背景、講工藝、講考古過程、講建築細節)
- 不要:純打卡拍照、純推薦無分析、廣告軟文
- B站影片優先 10 分鐘以上的講解類
- 公眾號文章優先有參考文獻或明確作者身份的
返回內容標題、URL、一句話摘要。
{如有聚焦主題:特別關注與「{聚焦主題}」相關的內容,其他維度作為背景補充。}
Wait for Research to complete and obtain all-dimension research results.
3. Content distillation (ContentAnalysis — optional)
If the user specified -q quick mode, skip this step.
From the results returned in step 2, extract all valid URLs (article links, video links).
Launch Agent subagents in parallel for each URL:
Each subagent calls the Skill tool to execute ContentAnalysis, passing in the URL, using fast depth level, to extract core knowledge points.
Fallback rules:
- If ContentAnalysis fails for a URL (inaccessible, no subtitles, etc.), skip that URL, do not block
- If all URLs fail, the process does not stop — the step 2 research results are already enough to generate the document
- ContentAnalysis is an enhancement layer, not a necessary layer
Collect all successfully distilled content summaries.
4. Synthesize org-mode document
Synthesize step 2 (research results) and step 3 (content distillation, if any) into a structured org-mode document.
Document structure:
#+title: {城市}旅行研究
#+date: {當前日期}
#+filetags: :travel:museum:architecture:
* 城市概覽
{城市}的文明座標——為什麼值得去,去了看什麼。一段話勾勒這座城市在中國文明史中的位置。
* 歷史分層
** {時期1}({年代範圍})
核心事件、遺留痕跡、對應可看的實物。
** {時期2}
...
* 博物館指南
** {博物館1名稱}
地址、開放時間、預約方式(如需要)。
*** 鎮館之寶
- {文物名}:{為什麼重要} | 看什麼細節:{具體觀察點}
*** 重點展廳
- {展廳名}:{核心看點}
*** 容易錯過的
- {被忽視但值得看的內容}
** {博物館2名稱}
...
* 古建遺存
** {古建1名稱}({朝代},{保護級別})
形制概述。
*** 看什麼
- {具體觀察點1}:{為什麼值得注意}
- {具體觀察點2}
** {古建2名稱}
...
* 考古發現
** {遺址/發現1}
發現經過、意義、出土文物現藏地。如果有有趣的發掘故事,講出來。
* 參觀路線
** 路線一:{主題名}({預計時間})
適合誰:{描述}
1. {地點} → 重點看 {什麼}({停留建議時間})
2. ...
** 路線二:{主題名}
...
* 深度內容推薦
從各平臺發現的值得在出發前看的內容。
** 影片
- [[{URL}][{標題}]] — {一句話摘要}
** 文章
- [[{URL}][{標題}]] — {一句話摘要}
** 書籍(如有推薦)
- {書名} — {為什麼值得讀}
File naming: use denote naming schema, save to ~/Documents/notes/ directory:
{YYYYMMDDTHHMMSS}==z--{城市}旅行研究.org
Writing requirements:
- Each recommendation must have "why see it" and "what details to look at", no vagueness allowed
- The tone is notes written for oneself, not tour guide commentary
- Write exact information when available, do not fabricate when not
5. Forge portable cards (ljg-card)
Extract the core content from the org document in step 4, forge two cards,execute in parallel:
Card A — City civilization overview (infographic):
Call the Skill tool to execute ljg-card -i, input content is: essence summary of the city's historical layers + core museum list + must-see ancient architecture list. Understand this city's civilizational skeleton in one image.
Card B — Visit Route Quick Reference (long image):
Call the Skill tool to execute ljg-card -l, with input content: visit route suggestions + the core highlights of each location. View anytime on your phone.
6. Summary Report
════ 旅行研究完成 ═══════════════════════
🏛️ 城市: {城市名}
📝 知識文件: {org 檔案路徑}
🖼️ 文明概覽卡: {PNG 檔案路徑}
🖼️ 路線速查卡: {PNG 檔案路徑}
📊 研究覆蓋: {N}個博物館 | {M}座古建 | {K}處考古遺址
📎 深度內容: {X}個影片 | {Y}篇文章
Key Constraints
- Step 2 is the core — 12 Agents in parallel cover academic research and platform content, completing it in one go
- Step 3 (content distillation) is an enhancement layer; failure does not block the process
- The two cards in Step 5 run in parallel
- The org document is the main output, and the cards are derivative outputs — document quality takes priority
- Do not produce generic travel guides; every recommendation must have "why to see it" and "what details to look at"
- Research searches use bilingual Chinese and English keywords to broaden coverage
- When there is no exact information, leave it blank rather than fabricate
Known Pitfalls
(First creation, no records yet. Accumulate through use.)