lark-fashion-cockpit-blogger-monitor

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

对标博主监控 + 视频拆解一站式

One-Stop Competitor Blogger Monitoring + Video Parsing

🎯 痛点： 老板娘每天花 2 小时刷竞品博主，看完没数据沉淀也不知道哪条值得学；想二创还要切工具拉视频 + 拆脚本，工作流断裂。

核心能力：

监控 N 个博主 → 算法评分 → 飞书卡片汇报 TOP 3

同 skill 内一键拆脚本 —— 选中爆款 → yt-dlp 拉视频 → faster-whisper 转录 → DOUBAO 多模态拆解 5 维度（镜头 / 台词 / 视觉 / 声音 / 节奏）→ 输出可执行二创 md

🎯 Pain Point: The store manager spends 2 hours daily browsing competitor bloggers, with no accumulated data and no clarity on which videos are worth learning from; secondary creation requires switching tools to download videos and parse scripts, resulting in a fragmented workflow.

Core Capabilities:

Monitor N bloggers → Algorithm scoring → Feishu card report of TOP 3

One-click script parsing within the same skill — Select viral videos → yt-dlp downloads videos → faster-whisper transcribes → DOUBAO multimodal parsing of 5 dimensions (shots / lines / visuals / sound / rhythm) → Output executable secondary creation md

二、视频拆解能力（合并自原 video-script-parser）

II. Video Parsing Capability (Merged from original video-script-parser)

scripts/parse-video-script.py

主脚本：

bash

undefined

scripts/parse-video-script.py

main script:

bash

undefined

单条拆解

Single video parsing

python scripts/parse-video-script.py --url "https://www.douyin.com/video/xxx"

不写飞书 wiki 只本地存

Save locally only without writing to Feishu Wiki

python scripts/parse-video-script.py --url "..." --no-wiki


输出 5 维度结构化拆解：
- 一、镜头脚本（按秒粒度）
- 二、文案台词（含转场点）
- 三、视觉设计（构图 / 配色 / 道具）
- 四、声音设计（旁白节奏 / BGM 选型）
- 五、节奏分析（高潮点位置）

依赖外部能力：
- `yt-dlp` 拉抖音 / 快手 / 视频号视频
- `faster-whisper` 本地 ASR（备选：DOUBAO ASR API，HF 网络受阻时用）
- `DOUBAO doubao-1-5-vision-pro-32k-250115` 多模态视觉拆帧
- 防盗链：requests Referer + Cookie header 解决 CDN 403

python scripts/parse-video-script.py --url "..." --no-wiki


Output structured parsing in 5 dimensions:
- 1. Shot Script (second-level granularity)
- 2. Copy & Lines (including transition points)
- 3. Visual Design (composition / color scheme / props)
- 4. Sound Design (narration rhythm / BGM selection)
- 5. Rhythm Analysis (climax position)

External Dependencies:
- `yt-dlp` to download videos from Douyin / Kuaishou / WeChat Channels
- `faster-whisper` local ASR (Alternative: DOUBAO ASR API, used when HF network is blocked)
- `DOUBAO doubao-1-5-vision-pro-32k-250115` multimodal visual frame parsing
- Anti-hotlinking: Solve CDN 403 with requests Referer + Cookie header

三、典型工作流（监控 → 拆解一气呵成）

III. Typical Workflow (Seamless Monitoring → Parsing)

每天 9:00 自动跑 monitor-bloggers.py
  ↓
飞书卡片"今日 Top 3 爆款"
  ↓
老板娘点其中一条的【拆脚本】按钮
  ↓
parse-video-script.py 自动跑
  ↓
30 秒后飞书 wiki 收到完整拆解 md
  ↓
朱健豪基于 md 二创新视频

Automatically run monitor-bloggers.py at 9:00 daily
  ↓
Feishu card "Today's Top 3 Viral Videos"
  ↓
Store manager clicks the [Parse Script] button for one of the videos
  ↓
parse-video-script.py runs automatically
  ↓
Complete parsed md is received in Feishu Wiki after 30 seconds
  ↓
Zhu Jianhao creates new videos based on the md

一、5 项算法 + 4 路筛选（复刻 douyin-monitor AI 博主监控）

I. 5 Algorithms + 4 Screening Rules (Replicate douyin-monitor AI Blogger Monitoring)

算法（爆款指数 0-10）

Algorithms (Viral Score 0-10)

指标	公式	阈值
赞粉比	likes / fans	≥0.5 → 5 分 / ≥0.2 → 3 分 / ≥0.05 → 1 分
分享率	shares / likes	≥0.1 → 2 分 / ≥0.03 → 1 分
收藏率	collects / likes	≥0.1 → 2 分 / ≥0.05 → 1 分
评赞比	comments / likes	≥0.05 → 1 分

Metric	Formula	Threshold
Like-to-Fan Ratio	likes / fans	≥0.5 → 5 points / ≥0.2 → 3 points / ≥0.05 → 1 point
Share Rate	shares / likes	≥0.1 → 2 points / ≥0.03 → 1 point
Collect Rate	collects / likes	≥0.1 → 2 points / ≥0.05 → 1 point
Comment-to-Like Ratio	comments / likes	≥0.05 → 1 point

4 路筛选（任一满足 → 已选中）

4 Screening Rules (Meet any → Selected)

AI 评分 ≥ 7（豆包/DeepSeek 主观打分）
收藏率 ≥ 10%（高保留价值）
评赞比 ≥ 8%（讨论度高）
赞粉比 ≥ 40%（爆款潜力）

不满足 → 待评估（不浪费老板娘时间看）

AI Score ≥7 (Subjective scoring by DOUBAO/DeepSeek)
Collect Rate ≥10% (High retention value)
Comment-to-Like Ratio ≥8% (High discussion)
Like-to-Fan Ratio ≥40% (Viral potential)

Not meeting any → Pending evaluation (Don't waste the store manager's time)

二、前置条件

II. Prerequisites

bash

pip install httpx openai
lark-cli auth login --scope "bitable:app:readwrite im:message:send"

bash

pip install httpx openai
lark-cli auth login --scope "bitable:app:readwrite im:message:send"

配凭证

Configure credentials

export LARK_FASHION_COCKPIT_BASE_TOKEN="..." export TABLE_BLOGGER_VIDEOS="tblxxxxxxxx" # 27 表 export DEEPSEEK_API_KEY="..." export DOUYIN_COOKIE="..." # 抖音爬数据用

---

export LARK_FASHION_COCKPIT_BASE_TOKEN="..." export TABLE_BLOGGER_VIDEOS="tblxxxxxxxx" # Table 27 export DEEPSEEK_API_KEY="..." export DOUYIN_COOKIE="..." # Used for crawling Douyin data

---

三、初始化（建表 + 灌 mock 数据）

III. Initialization (Create table + populate mock data)

bash

python scripts/seed-mock-bloggers.py

bash

python scripts/seed-mock-bloggers.py

→ 在飞书 27 表灌入 10 条 mock 女装博主数据（含真实风格的口播文案）

→ Populate 10 mock women's fashion blogger data (including real-style narration scripts) in Feishu Table 27


19 个字段：博主名 / 粉丝 / 视频标题 / 发布日期 / 视频链接 / 点赞 / 评论 / 收藏 / 分享 / 赞粉比 / 互动率 / 分享率 / 收藏率 / 评赞比 / 爆款指数 / AI 评分 / 推荐角度 / 口播文案 / 状态

---


19 fields: Blogger Name / Fans / Video Title / Publish Date / Video Link / Likes / Comments / Collects / Shares / Like-to-Fan Ratio / Engagement Rate / Share Rate / Collect Rate / Comment-to-Like Ratio / Viral Score / AI Score / Recommendation Angle / Narration Script / Status

---

四、运行监控

IV. Run Monitoring

bash

undefined

bash

undefined

完整流程：抓数据 → 算分 → 写表 → 发卡片

Full workflow: Crawl data → Calculate scores → Write to table → Send card

python scripts/monitor-bloggers.py

自定义参数

Custom parameters

python scripts/monitor-bloggers.py --top 5 --reply-chat-id oc_xxx

---

python scripts/monitor-bloggers.py --top 5 --reply-chat-id oc_xxx

---

五、典型汇报卡片

V. Typical Report Card

👀 对标博主每日监控

📊 监控博主：10 位 ｜ 今日抓取视频：10 条
🎯 已选中（值得二创）：3 条 ｜ 待评估：7 条

🔥 爆款指数 TOP 3（按算法排序）

[1] 程程姐_穿搭教学 · 粉丝 85.0w · 🎤 有口播
📝 150 显高显瘦秘籍 ｜ 衬衫塞进高腰裤的 3 种正确方式
👍 42.5w  💬 8.5k  ⭐ 5.6w  🔄 1.2w
赞粉比 50.00%  收藏率 13.18%  爆款指数 9/10
💡 推荐角度：教学型痛点解决
口播文案（前 120 字）：嗨大家好我是程程姐。今天教你们三招衬衫塞进高腰裤的正确方式...
[▶ 看视频]

[2] Sandy小苹果 · ...
[3] 大码姑娘子涵 · ...

[📋 查看完整 27 表]

👀 Competitor Blogger Daily Monitoring

📊 Monitored Bloggers: 10 | Videos Crawled Today: 10
🎯 Selected (Worthy of Secondary Creation): 3 | Pending Evaluation: 7

🔥 Top 3 Viral Scores (Sorted by Algorithm)

[1] Cheng Cheng Jie_Outfit Tutorial · 850k Fans · 🎤 With Narration
📝 150cm Height Enhancement & Slimming Tips | 3 Correct Ways to Tuck Shirts into High-Waisted Pants
👍 425k  💬 8.5k  ⭐ 56k  🔄 12k
Like-to-Fan Ratio 50.00%  Collect Rate 13.18%  Viral Score 9/10
💡 Recommendation Angle: Pain Point Solving via Tutorial
Narration Script (First 120 words): Hi everyone, I'm Cheng Cheng Jie. Today I'll teach you three correct ways to tuck shirts into high-waisted pants...
[▶ Watch Video]

[2] Sandy Little Apple · ...
[3] Plus-Size Girl Zihan · ...

[📋 View Complete Table 27]

六、与其他 skill 协作

VI. Collaboration with Other Skills

视频拆解能力已合并入本 skill（见上方"二、视频拆解能力"），同 skill 内一键调
```
parse-video-script.py
```
```
content-pipeline
```
— 选中的视频作为二创素材源
```
competitor-monitor
```
— 趋势分析联动

Video parsing capability has been merged into this skill (see "II. Video Parsing Capability" above), one-click call to
```
parse-video-script.py
```
within the same skill
```
content-pipeline
```
— Selected videos serve as secondary creation material sources
```
competitor-monitor
```
— Trend analysis linkage

七、cron 定时跑（推荐）

VII. Cron Scheduled Run (Recommended)

bash

undefined

bash

undefined

早 8:30 自动跑（搭配 morning-report 9:00）

Auto-run at 8:30 AM (paired with morning-report at 9:00)

crontab -e 30 8 * * * cd /path/to/lark-fashion-cockpit && python skills/blogger-monitor/scripts/monitor-bloggers.py

---

crontab -e 30 8 * * * cd /path/to/lark-fashion-cockpit && python skills/blogger-monitor/scripts/monitor-bloggers.py

---

八、参考

VIII. References

监控主脚本：
```
scripts/monitor-bloggers.py
```
mock 数据脚本：
```
scripts/seed-mock-bloggers.py
```
卡片样式 sample：
```
examples/sample-card.md
```

Monitoring main script:
```
scripts/monitor-bloggers.py
```
Mock data script:
```
scripts/seed-mock-bloggers.py
```
Card style sample:
```
examples/sample-card.md
```