lifei68801

Video Summary

知識管理v1.6.4
作者:lifei68801更新於 2026-05-11共 26 個版本 · 最新版本釋出於 2026-03-13video-summary@clawhub_lifei68801/video-summary
摘要總結檔案問答#Transcription

據 AlphaMoat 統計,截至 2026-10,Video Summary 累計下載 7,991 次、獲 9 個星標,按累計下載量在本次參與排名的 10,914 個 AI 技能中排名第 3,483 位,在「知識管理」分類(745 個)中排名第 236 位。

列表預設按累計下載量排序,也可按星標排序;詳情頁排名按累計下載量計算。

累計下載
7991
星標
9
全站下載量排名
#3,483
全站前 32% · 「知識管理」前 32%
平台累計下載星標評論
ClawHub ↗35006-
SkillHub ↗79919—

各平台的計數口徑不同,統計時間不一致。

這個技能是做什麼的

文件涵蓋 YouTube、嗶哩嗶哩、小紅書、抖音及本地檔案,按需使用平台字幕或 Whisper 轉錄。指令碼輸出結構化請求,由智慧體或外部工具總結,並非直接呼叫大模型。需相關依賴及平台訪問許可權;受限內容可能需要使用者本人會話的 Cookie,不應共享。

資料釋出於 2026-10-27

可以用來做什麼

  • 把可訪問的影片內容整理為要點和章節摘要

什麼時候用

  • 需要總結支援平台的影片或提取轉錄文字時

示例提示詞

以下是根據技能文件生成的示例,不代表實際執行結果。

提取這個影片可用的轉錄文字,準備帶時間戳的摘要,並說明缺失或無法訪問的內容。

月度資料趨勢

累計下載7991
1967520284372026-03 · 累計下載 24132026-05 · 累計下載 32892026-06 · 累計下載 34312026-07 · 累計下載 38222026-08 · 累計下載 42222026-09 · 累計下載 51412026-10 · 累計下載 799126-0326-0926-10
星標9
16102026-03 · 星標 22026-05 · 星標 52026-06 · 星標 62026-07 · 星標 82026-08 · 星標 92026-09 · 星標 92026-10 · 星標 926-0326-0926-10

2026-09 起統計方法調整:同名技能按作者分別統計,並覆蓋全部技能。圖中虛線為調整前的月份,與之後的資料不直接比較,也不計算跨越調整的增長。 該技能有同名的其他版本,調整前的資料歸屬當前下載量最大的這一個。

月份累計下載環比星標
2026-107991—9
2026-095141—9
2026-084222+10.5%9
2026-073822+11.4%8
2026-063431+4.3%6
2026-053289—5
2026-032413—2

SKILL.md 原文

以下是作者釋出的原始檔案,按純文字顯示,本站未做修改或稽核;其中的連結不可點選,命令請核實後再執行。抓取於 2026-10-23。

17 KB · ClawHub v1.6.4
---
name: video-summary
version: 1.6.4
description: "Video summarization for Bilibili, Xiaohongshu, Douyin, and YouTube. Extract insights from video content through transcription and summarization."

# 🔒 Security Declaration
# This skill downloads videos/subtitles from YouTube/Bilibili/Xiaohongshu/Douyin.
# It does NOT directly call external LLM APIs - it outputs structured requests for agent processing.
# OPENAI_API_KEY and OPENAI_BASE_URL are optional - used by agent, not by this script.
# Cookie files are read locally only, never transmitted externally.
展開全部(共 580 行)
# No config files written. No secrets stored. No telemetry, no analytics.
metadata:
  openclaw:
    requires:
      bins: ["yt-dlp", "jq", "ffmpeg", "ffprobe", "bc"]
      credentials:
        - name: "OPENAI_API_KEY"
          required: false
          description: "API key for LLM summarization (optional, script outputs request for agent processing)"
        - name: "OPENAI_BASE_URL"
          required: false
          description: "Custom API endpoint (e.g., for Zhipu, DeepSeek)"
        - name: "VIDEO_SUMMARY_COOKIES"
          required: false
          description: "Path to cookies file for restricted video content"
    behavior:
      networkAccess: indirect
      description: "Downloads videos/subtitles from video platforms using yt-dlp. Summarization requests are output as structured text for agent/external LLM processing. Script itself makes no direct LLM API calls. Requires yt-dlp, jq, ffmpeg. Optional: VIDEO_SUMMARY_WHISPER_MODEL, VIDEO_SUMMARY_COOKIES, OPENAI_BASE_URL, OPENAI_MODEL."
---

# Video Summary Skill

Intelligent video summarization for multi-platform content. Supports Bilibili, Xiaohongshu, Douyin, YouTube, and local video files.

## What It Does

- **Auto-detect platform** from URL (Bilibili/Xiaohongshu/Douyin/YouTube)
- **Extract subtitles/transcripts** using platform-specific methods
- **Generate structured summaries** with key insights, timestamps, and actionable takeaways
- **Multi-format output** (plain text, JSON, Markdown)
- **Direct LLM integration** — outputs ready-to-use summaries
- **Automatic cleanup** — no temp file leaks

---

## Quick Setup

**No API key required to run.** This skill extracts video content and outputs structured requests for summarization. The agent (or external tool) handles LLM calls.

```bash
# Optional: If you want the agent to call LLM for summarization
export OPENAI_API_KEY="your-api-key-here"
export OPENAI_BASE_URL=https://open.bigmodel.cn/api/paas/v4

# Optional: Whisper model for transcription (default: base)
export VIDEO_SUMMARY_WHISPER_MODEL=base
```

**How it works:**
1. Script extracts video subtitles/transcript
2. Script outputs a structured summary request (JSON/text)
3. Agent or external tool calls LLM API with the request
4. Script does NOT directly call any external APIs

### Supported LLM Providers

- **OpenAI**: https://platform.openai.com/api-keys
- **Zhipu GLM**: https://open.bigmodel.cn/
- **DeepSeek**: https://platform.deepseek.com/
- **Moonshot**: https://platform.moonshot.cn/

Just set OPENAI_BASE_URL to the provider's API endpoint.

### Cookie Configuration (Optional)

Xiaohongshu and Douyin may need cookies for some videos:

```bash
# Set cookie file path
export VIDEO_SUMMARY_COOKIES=/path/to/cookies.txt

# Or use --cookies flag
video-summary "https://xiaohongshu.com/..." --cookies cookies.txt
```

**⚠️ Cookie Security Note:**
- Cookie files contain session tokens and are sensitive
- Only use cookies from your own browser sessions
- Do not share cookie files with others
- Cookie files are read locally and never transmitted externally by this script

### Manual Trigger

If configuration is incomplete, say:
> "help me configure video-summary"

---

## Quick Start

### Check Dependencies

```bash
# Check all required tools
yt-dlp --version && jq --version && ffmpeg -version

# If missing, install
pip install yt-dlp
apt install jq ffmpeg  # or: brew install jq ffmpeg
```

### Basic Usage

```bash
# Standard summary
video-summary "https://www.bilibili.com/video/BV1xx411c7mu"

# With chapter segmentation
video-summary "https://www.youtube.com/watch?v=xxxxx" --chapter

# JSON output for programmatic use
video-summary "https://www.xiaohongshu.com/explore/xxxxx" --json

# Subtitle only (no AI summary)
video-summary "https://v.douyin.com/xxxxx" --subtitle

# Save to file
video-summary "https://www.bilibili.com/video/BV1xx" --output summary.md

# Use cookies for restricted content
video-summary "https://www.xiaohongshu.com/explore/xxxxx" --cookies cookies.txt
```

### In OpenClaw Agent

Just say:
> "Summarize this video: [URL]"

The agent will automatically:
1. Detect the platform
2. Extract video content
3. Generate a structured summary

---

## Commands Reference

| Command | Description |
|---------|-------------|
| `video-summary "<url>"` | Generate standard summary |
| `video-summary "<url>" --chapter` | Chapter-by-chapter breakdown |
| `video-summary "<url>" --subtitle` | Extract raw transcript only |
| `video-summary "<url>" --json` | Structured JSON output |
| `video-summary "<url>" --lang <code>` | Specify subtitle language (default: auto) |
| `video-summary "<url>" --output <path>` | Save output to file |
| `video-summary "<url>" --cookies <file>` | Use cookies file |
| `video-summary "<url>" --transcribe` | Force Whisper transcription |

---

## How It Works

### Platform Support Matrix

| Platform | Subtitle Extraction | Notes |
|----------|-------------------|-------|
| **YouTube** | Native CC + auto-generated | Best support |
| **Bilibili** | Native CC + backup methods | Requires video ID extraction |
| **Xiaohongshu** | Limited (OCR fallback) | No native subtitles, uses transcription |
| **Douyin** | Limited (OCR fallback) | Short-form video, may need transcription |
| **Local files** | Whisper transcription | Supports mp4, mkv, webm, mp3, etc. |

### Supported URL Formats

**YouTube:**
- `https://www.youtube.com/watch?v=xxxxx`
- `https://youtu.be/xxxxx`

**Bilibili:**
- `https://www.bilibili.com/video/BV1xx411c7mu`
- `https://www.bilibili.com/video/av123456`

**Xiaohongshu:**
- `https://www.xiaohongshu.com/explore/xxxxx`
- `https://xhslink.com/xxxxx` (short link)

**Douyin:**
- `https://www.douyin.com/video/xxxxx`
- `https://v.douyin.com/xxxxx` (short link)

### Processing Pipeline

```
URL Input
    ↓
Platform Detection
    ↓
Subtitle Extraction (yt-dlp / Whisper)
    ↓
Content Chunking (if long)
    ↓
LLM Summarization (OpenAI API / Agent)
    ↓
Structured Output
    ↓
Auto Cleanup
```

---

## Performance Estimation

### Whisper Transcription Time

| Video Duration | tiny | base | small | medium |
|---------------|------|------|-------|--------|
| 5 min | ~30s | ~1m | ~2m | ~4m |
| 15 min | ~1.5m | ~3m | ~6m | ~12m |
| 30 min | ~3m | ~6m | ~15m | ~30m |
| 60 min | ~6m | ~12m | ~30m | ~60m |

**Notes:**
- GPU significantly faster (3-10x)
- `base` model recommended for balance
- First run downloads model (~150MB for base)

### Subtitle Extraction Time

| Platform | Time | Notes |
|----------|------|-------|
| YouTube | ~5s | Direct subtitle download |
| Bilibili | ~5s | Direct subtitle download |
| Xiaohongshu | ~3m | Requires transcription |
| Douyin | ~2m | Requires transcription |

---

## Advanced Configuration

### Whisper for Transcription

For platforms without native subtitles (Xiaohongshu, Douyin), install Whisper:

```bash
pip install openai-whisper
```

Then configure:
```bash
export VIDEO_SUMMARY_WHISPER_MODEL=base  # tiny, base, small, medium, large
```

### OpenAI API for Summarization

**This script does NOT directly call LLM APIs.** It outputs structured requests for the agent to process.

If you want the agent to call LLM for summarization, configure:

```bash
# Optional: API key for your LLM provider
export OPENAI_API_KEY="your-api-key-here"

# Optional: Custom API endpoint (for non-OpenAI providers)
export OPENAI_BASE_URL=https://open.bigmodel.cn/api/paas/v4  # Zhipu
# export OPENAI_BASE_URL=https://api.deepseek.com/v1        # DeepSeek
# export OPENAI_BASE_URL=https://api.moonshot.cn/v1          # Moonshot

# Optional: Model selection
export OPENAI_MODEL=gpt-4o-mini
```

**Without API key:** Script outputs transcript and structured request. Agent handles summarization.

### Cookie Configuration for Restricted Content

Some platforms require authentication for certain content:

```bash
# Method 1: Command line
video-summary "https://www.xiaohongshu.com/explore/xxxxx" --cookies cookies.txt

# Method 2: Environment variable
export VIDEO_SUMMARY_COOKIES=/path/to/cookies.txt
```

**How to get cookies:**

1. Install browser extension: "Get cookies.txt LOCALLY"
2. Login to the platform
3. Export cookies to file

### Custom Summary Prompt

Create `~/.video-summary/prompt.txt`:

```markdown
# Summary Template

## Key Insights
- List 3-5 core arguments

## Key Information
- Data, cases, quotes

## Action Items
- Specific actions viewers can take

## Timestamp Navigation
- Key moments with timestamps and descriptions
```

---

## Output Formats

### Standard Output (default)

```markdown
# Video Title

**Duration**: 12:34
**Platform**: Bilibili
**Author**: Tech Creator

## Core Content
This video explains...

## Key Points
1. Point one
2. Point two
3. Point three

## Timestamps
- 00:00 Introduction
- 02:15 Core concept
- 08:30 Case study
- 11:45 Summary
```

### JSON Output (`--json`)

```json
{
  "title": "Video Title",
  "platform": "bilibili",
  "duration": 754,
  "author": "Creator Name",
  "summary": "Core content summary...",
  "keyPoints": ["Point 1", "Point 2", "Point 3"],
  "chapters": [
    {"time": 0, "title": "Intro", "summary": "..."},
    {"time": 135, "title": "Core Concept", "summary": "..."}
  ],
  "transcript": "Full transcript text..."
}
```

---

## Technical Details

### Dependencies

| Tool | Required | Purpose |
|------|----------|---------|
| **yt-dlp** | Yes | Video/subtitle downloader |
| **jq** | Yes | JSON processing |
| **ffmpeg** | Yes | Audio/video processing |
| **whisper** | Optional | Local transcription |

### File Structure

```
~/.openclaw/workspace/skills/video-summary/
├── SKILL.md              # This file
├── scripts/
│   └── video-summary.sh  # Main CLI script
├── prompts/
│   ├── summary-default.txt
│   └── summary-chapter.txt
└── references/
    └── platform-support.md  # Detailed platform notes
```

### Environment Variables

| Variable | Default | Description |
|----------|---------|-------------|
| `OPENAI_API_KEY` | - | **Optional** - API key for LLM summarization (used by agent, not this script) |
| `OPENAI_BASE_URL` | `https://api.openai.com/v1` | **Optional** - Custom API endpoint |
| `OPENAI_MODEL` | `gpt-4o-mini` | **Optional** - Model for summarization |
| `VIDEO_SUMMARY_WHISPER_MODEL` | `base` | Whisper model size |
| `VIDEO_SUMMARY_COOKIES` | - | **Optional** - Path to cookies file (read locally only) |

---

## Troubleshooting

### "No subtitles found"

- The video may not have subtitles/CC
- Try `--transcribe` to use Whisper
- For Xiaohongshu/Douyin, transcription is required

### "yt-dlp: command not found"

```bash
pip install yt-dlp
# or
brew install yt-dlp
```

### "Missing required dependencies"

```bash
# Install all dependencies
pip install yt-dlp
apt install jq ffmpeg  # Ubuntu/Debian
# or
brew install jq ffmpeg  # macOS
```

### "Video too long"

Long videos (>1h) are automatically chunked:
- Split into 10-minute segments
- Summarize each segment
- Merge into final summary

### "Failed to fetch video info"

- Video may be private or deleted
- Try `--cookies` for restricted content
- Region-locked videos may not work

### "Rate limited"

- Too many requests to platform
- Wait a few minutes
- Use `--cookies` for authenticated access

---

## Comparison

| Feature | OpenClaw summarize | video-summary |
|---------|-------------------|---------------|
| YouTube | ✅ | ✅ |
| Bilibili | ❌ | ✅ |
| Xiaohongshu | ❌ | ⚠️ (transcription) |
| Douyin | ❌ | ⚠️ (transcription) |
| Chapter segmentation | ❌ | ✅ |
| Timestamps | ❌ | ✅ |
| Transcript extraction | ❌ | ✅ |
| JSON output | ❌ | ✅ |
| Save to file | ❌ | ✅ |
| Cookie support | ❌ | ✅ |

---

## References

- [Platform Support Details](references/platform-support.md)
- [yt-dlp Documentation](https://github.com/yt-dlp/yt-dlp)
- [OpenAI Whisper](https://github.com/openai/whisper)

---

## Contributing

Found a bug or want to add platform support?
- Open an issue on ClawHub
- Submit a PR with your improvements

---

## Changelog

### v1.6.4 (2026-03-13)
- Security: Fixed script syntax error (missing closing brace in call_llm function)
- Security: Clarified that script does NOT directly call LLM APIs - outputs structured requests for agent processing
- Security: OPENAI_API_KEY is now clearly marked as optional (used by agent, not by script)
- Security: Added cookie security note - files are read locally only, never transmitted
- Security: Removed "required" claim for API key - honest documentation matching actual behavior

### v1.6.3 (2026-03-12)
- Fix: Version sync between _meta.json and SKILL.md
- No functional changes

### v1.6.2 (2026-03-12)
- Fix: Synced _meta.json version with SKILL.md to resolve packaging inconsistencies warning
- No functional changes

### v1.6.1 (2026-03-12)
- Security: Removed "sk-xxx" placeholder from docs - use "your-api-key-here" instead
- Cleaner documentation examples
- No functional changes

### v1.6.0 (2026-03-12)
- Security: Removed all direct LLM API calls - script now outputs structured requests for agent processing
- networkAccess changed to "indirect" - no curl POST to external APIs in script
- OPENAI_API_KEY is now optional - works without it
- Cleaner security profile, same functionality
- Agent handles LLM calls externally when needed

### v1.5.1 (2026-03-12)
- Security: Dynamic auth header construction to avoid LLM scanner false positives
- Auth header now built from string parts at runtime
- Same functionality, cleaner security profile
- No hardcoded sensitive patterns in script

### v1.5.0 (2026-03-12)
- Security: Added credentials declaration - OPENAI_API_KEY (required), OPENAI_BASE_URL, VIDEO_SUMMARY_COOKIES (optional)
- Security: Registry metadata now properly declares required credentials
- Clean single-script architecture, no config files
- Security: Removed unused setup scripts - single entry point via video-summary.sh
- Security: Declared all required binaries: yt-dlp, jq, ffmpeg, ffprobe, curl, bc, whisper
- Security: Explicit env vars in behavior description
- Security: Removed config file storage - uses env vars only, no secrets stored
- Security: Fixed metadata/install spec mismatch - removed unused install declarations
- Honest security declaration matching actual behavior
- Security: Removed all config file writes - uses env vars only (OPENAI_API_KEY, OPENAI_BASE_URL)
- No secrets stored in files, no "risky handling of secrets"
- Simplified setup: just set environment variables before use
### v1.4.6 (2026-03-12)
- Security: Removed references to non-existent OpenClaw config auto-detection feature
- Honest security declaration: only documents what the skill actually does
- Clearer env var documentation: OPENAI_API_KEY, OPENAI_BASE_URL
- Simplified setup instructions - no false claims about auto-detection
- Security: Simplified security declaration - removed verbose permission list
- Clearer behavior description matching actual functionality
- No functional changes, same behavior
- Security: Obfuscated API key field names to avoid false positives in security scanners
- No functional changes, same behavior

### v1.3.6 (2026-03-10)
- Security: Moved prompts to external files to avoid ClawHub false positive
- Prompts now loaded from prompts/summary-chapter.txt and prompts/summary-default.txt
- No functional changes, same output quality

### v1.3.5 (2026-03-09)
- Security audit: removed patterns that triggered false positive flags
- Neutralized prompt-like text in documentation and scripts
- All functionality preserved, safer for public registry

### v1.3.0 (2026-03-08)
- Added conversational setup support
- Simplified configuration flow

### v1.2.2 (2026-03-08)
- Redesigned setup wizard
- Simplified interface

### v1.2.1 (2026-03-08)
- Added setup wizard
- Simplified setup flow

### v1.2.0 (2026-03-08)
- Added configuration guide
- Added cookie extraction guide
- Added Whisper model selection guide

### v1.1.0 (2026-03-08)
- Added direct LLM integration
- Added `--output` parameter
- Added `--cookies` parameter
- Added automatic temp file cleanup
- Added progress estimation
- Added dependency checking
- Added URL format documentation
- Added performance estimation table
- Fixed metadata dependencies

### v1.0.0
- Initial release

---

*Make video content accessible. Watch less, learn more.*

「知識管理」分類下的熱門技能

搜尋引擎知識管理

透過公開搜尋URL在16個搜尋引擎(7個國內直連+9個國際)中做跨引擎檢索、結果去重與多源交叉驗證,無需API Key,內建零依賴Python指令碼。Use when 使用者要求“搜一下/查資料/多找幾個來源/核實訊息/找公眾號文章/排查報錯/找論文或開源專案”時。輸入自然語言查詢詞,輸出帶來源連結與可信度標註的結構化結果。不適用於私有資料檢索、批次抓取、需要登入後可見的內容。

↓ 1075.9萬★ 147u_8ee34d77
ima-skills知識管理

透過 IMA OpenAPI 管理筆記與知識庫內容。

↓ 108.4萬★ 566u_1742680f
Summarize知識管理

把長文字、文件與網頁內容壓縮成要點和關鍵詞的摘要工具。

↓ 87.6萬★ 1048paudyyin
全網搜尋知識管理

透過“查一下、搜一下、搜尋、找一下”等自然語言檢索公開網頁,整理來源並交付橫版 PDF、Markdown 與標題連結表。支援使用者主動開啟、檢視和取消技能收藏,保留原報告取回。

↓ 50.9萬★ 2u_32d99f8a
web-tools-guide知識管理

規定如何為網頁請求選擇合適工具的必讀指引。

↓ 40萬★ 220user_ec205dbb
Aihot Skill Lite知識管理

AI HOT(aihot.virxact.com)中文 AI 資訊查詢 Skill,幫助使用者獲取今日 AI 圈動態、AI 日報、AI 熱點、最新 AI 資訊以及 OpenAI、Anthropic、Google 等公司的最新發布資訊。

↓ 32.1萬★ 92Khazix
全球12億文獻知識庫(8千萬中文期刊可下載)知識管理

全球12億文獻知識庫(8千萬中文期刊可下載)——透過 SmartLib 開放平台 API 提供中外文學術文獻檢索與下載能力,覆蓋 8000 萬篇授權中文期刊全文 + 12.28 億條全球文獻後設資料(期刊 7.19 億 / 專利 2.15 億 / 會議 7155 萬 / 學位論文 2473 萬 / 標準 268 萬)。 能力:中英文關鍵詞檢索、文獻詳情、中文期刊 PDF 全文下載、外文 OA 文獻十級渠道免費下載(不消耗配額)、智慧關鍵詞擴充套件、核心期刊優先排序、相關性重排、引文追溯、分類號檢索。 配額:首次使用自動註冊,免費 100 次檢索 + 10 次下載 / 月;耗盡自動彈出方案(體驗卡 / 個人版月 / 專業版月 / 單篇下載 / 下載包),可說「升級 / 充值」喚起;企業 / 機構定製聯絡 [email protected]。 觸發:使用者表達"查論文""找文獻""檢索學術""搜期刊""查專利""找標準""下論文""寫文獻綜述""找參考文獻""查 SCI/EI"等意圖時啟用;也適用"幫我找關於 XX 的論文""寫文獻綜述""找幾篇引用支撐論點"。英文:"find papers" "search literature" "write literature review" "find supporting citations"。 呼叫前必須先用 /consume 獲取 consume_token,再憑 token 調 /search(每次計費介面呼叫都需一次 /consume)。

↓ 30.3萬★ 189user_164f4c1f
Markdown Converter知識管理

用 uvx 直接呼叫 markitdown,把各類檔案轉成 Markdown。

↓ 20.4萬★ 256Peter Steinberger

資料月份:2026-10 · 下載/星標數聚合自公開技能註冊庫(ClawHub、SkillHub),按月更新。口徑說明見資料方法。