AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Ecommerce Video Clip Workflow

skill-yehyakin-hermes-skills-ecommerce-video-clip-workflow · by yehyakin

竞品视频 → AI转写 → 识别带货节点 → 精剪切片 → 烧字幕的完整流水线。用于从长视频(直播/竞品回放)中批量提取挂车带货高光片段。

No reviews yet
0 installs
32 views
0.0% view→install

Install

$ agentstack add skill-yehyakin-hermes-skills-ecommerce-video-clip-workflow

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-yehyakin-hermes-skills-ecommerce-video-clip-workflow)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
4mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Ecommerce Video Clip Workflow? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

电商带货视频切片工作流

从长视频(直播/竞品视频)中自动识别高价值带货片段,精剪为挂车切片。

适用场景

  • 竞品视频话术分析 + 切片提取
  • 直播回放高光时刻挖掘
  • 视频挂车素材批量制作

前置要求

  • whisper.cpp 已编译 Metal 支持(Mac Apple Silicon)
  • ffmpeg-full(支持 libass subtitles 滤镜)
  • 模型:ggml-small.bin(中文识别准确 + 速度快)

Pipeline

Step 1: 提取音频

ffmpeg -i INPUT.mp4 -vn -c:a pcm_s16le audio.wav

Step 2: Whisper 转写(Metal GPU,8x 实时)

cd ~/whisper.cpp
./build/bin/whisper-cli \
  -m models/ggml-small.bin \
  -f audio.wav \
  -l zh \
  --output-txt \
  --output-file transcript

速度:small 模型 Metal 加速 ~8x 实时(base 模型仅 0.8x 且中文差)

Step 3: 定位带货节点

# 找"一号链接"、"必买"、"鲜货"、"限量"等关键词
grep -n "一号链接\|必买\|鲜货\|限量\|划算" transcript.txt

# 计算行号 → 视频时间(每行 ≈ 视频秒数 / 总行数)
python3 -c "
video_sec = 13112  # 视频总秒数
lines = 8480       # transcript 总行数
per_line = video_sec / lines  # ≈ 1.54秒/行
"

Step 4: 精剪片段(ffmpeg)

ffmpeg -y -i INPUT.mp4 \
  -ss 1630 -t 25 \
  -c:v libx264 -crf 20 -preset fast \
  -c:a aac -b:a 128k \
  clip.mp4

Step 5: 生成 SRT 字幕文件

# 根据行号范围 + 每行秒数生成 SRT
def lines_to_srt(transcript_path, start_line, end_line, per_line_sec=1.54):
    # 生成带时间戳的 SRT 文件

Step 6: 烧字幕(ffmpeg-full + libass)

/opt/homebrew/opt/ffmpeg-full/bin/ffmpeg -y \
  -ss START -t DURATION \
  -i INPUT.mp4 \
  -vf "subtitles=SRT_PATH:force_style='FontSize=28,Bold=1,OutlineColour=&H00000000,Outline=2'" \
  -c:v libx264 -crf 20 -preset fast \
  -c:a aac -b:a 128k \
  OUTPUT.mp4

关键:必须用 ffmpeg-full(Homebrew),标准 ffmpeg 缺 libass 滤镜。

关键经验

| 问题 | 原因 | 解决方案 | |------|------|---------| | Whisper CPU 太慢(3.6小时要跑7小时)| base 模型 + CPU | 改 whisper.cpp small 模型 + Metal | | base 模型中文识别差 | 模型太小 | 换 ggml-small.bin | | ffmpeg subtitles 滤镜报错 | 标准 ffmpeg 不含 libass | brew install ffmpeg-full | | SRT 时间戳和片段对不上 | transcript 是纯文本无时间戳 | 用 总秒数/总行数 估算每行时长 | | 中文路径导致 ffmpeg 报错 | 路径含中文字符 | 复制到 ~/tmp/ 简单路径 |

输出规格

  • 分辨率:1088x1920(竖屏抖音/视频号)
  • 时长:20-30秒(挂车最佳)
  • 编码:H264 + AAC
  • 字幕:内嵌 SRT burn-in

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.