Getting it into your agent
There is no command for this one: it runs only inside a plugin, and the catalogue could not identify which plugin ships it. The source is linked below.
Wrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/yehyakin/hermes-skills/whisper-video-clipping-workflow)<a href="https://agentmods.dev/skills/yehyakin/hermes-skills/whisper-video-clipping-workflow"><img src="https://agentmods.dev/badge/skills/yehyakin/hermes-skills/whisper-video-clipping-workflow/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/yehyakin/hermes-skills/whisper-video-clipping-workflow"><img src="https://agentmods.dev/badge/skills/yehyakin/hermes-skills/whisper-video-clipping-workflow.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00051 | $0.06367 |
| Opus 5 | $0.00026 | $0.03184 |
| Sonnet 5 | $0.00010 | $0.01273 |
| Haiku 4.5 | $0.00005 | $0.00637 |
Grade A, and why
whisper-video-clipping-workflow scanned grade A with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -L -o models/ggml-small.bin "https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-small.bin" Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
result = subprocess.run( How it starts
The opening of the file, as written. The whole thing — 584 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Whisper 视频字幕提取 + 带货切片剪辑工作流
适用场景
- 竞品直播间录屏 → 提取话术 → 剪辑带货片段
- 直播回放 → 自动找"一号链接"/"必买"/"鲜货"等高光时刻
- 任意长视频 → 生成可搜索字幕 → 精准定位内容
完整流程
步骤 1:安装 whisper.cpp(Mac Metal 加速)
# 安装 cmake
brew install cmake
# 克隆 whisper.cpp
git clone https://github.com/ggerganov/whisper.cpp.git ~/whisper.cpp
# 编译(不用 CoreML,用 Metal 加速)
cd ~/whisper.cpp
cmake -B build -DCMAKE_BUILD_TYPE=Release -DWHISPER_COREML=OFF -DWHISPER_METAL=ON
cmake --build build -j$(sysctl -n hw.ncpu)
步骤 2:下载模型
| 模型 | 大小 | 速度 | 适用场景 |
|---|---|---|---|
| small | 465MB | ~8x 实时(M1) | 中文推荐,精度高速度快 |
| base | 141MB | ~0.8x 实时 | 仅英文或测试 |
cd ~/whisper.cpp
curl -L -o models/ggml-small.bin "https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-small.bin"
步骤 3:提取音频 + 转录
# 从视频提取音频
ffmpeg -i /path/to/video.mp4 -vn -c:a pcm_s16le /path/to/audio.wav
# 后台运行转录(--output-txt 只输出文本)
cd ~/whisper.cpp
./build/bin/whisper-cli \
-m models/ggml-small.bin \
-f /path/to/audio.wav \
-l zh \
--output-txt \
--output-file /path/to/transcript \
> /path/to/whisper_log.txt 2>&1 &
# 注意:whisper-cli 跑完才写文件,不是流式的
# 查看进程:ps aux | grep whisper-cli
步骤 4:定位高价值片段
# whisper.txt 是纯文本(无时间戳),每行约 1.54 秒
# 计算公式:行号 × 1.54 = 视频秒数
import subprocess
# 找"一号链接"、"鲜货"、"必买"、"好价格"等关键词
result = subprocess.run(
['grep', '-n', '一号链接\\|鲜货\\|必买\\|好价格\\|限量\\|抢', '/path/to/transcript.txt'],
capture_output=True, text=True
)
print(result.stdout)
# 找关键词密度最高的时段
步骤 5:剪辑片段(每段 45 秒)
# 时间戳计算:line_number × 1.54 = seconds
clips = [
("01_开场马甲介绍", 919, 45),
("02_限量15件抢鲜", 1630, 45),
# ...
]
import subprocess
for name, start_sec, duration in clips:
output = f"/path/to/clips/{name}.mp4"
subprocess.run([
'ffmpeg', '-y', '-i', '/path/to/video.mp4',
'-ss', str(start_sec - 5), # 多录5秒前的内容
'-t', str(duration),
'-c:v', 'libx264', '-crf', '23', '-preset', 'ultrafast',
'-c:a', 'aac', '-b:a', '128k',
output
], check=True)
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 584 lines · 51 tokens per session scan A 6942271cb931
whisper-video-clipping-workflow is a skill published in the GitHub repository yehyakin/hermes-skills (9 stars, last pushed 3mo ago), licensed MIT. It adds 51 tokens to every session and 6,367 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 2 findings (makes network calls, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
media-fetch
Download video/audio from 1000+ sites (YouTube, Twitter/X, Spotify, TikTok, etc.), extract audio, download subtitles, transcribe speech to text, and search for videos. Use when asked to download, summarize, or transcribe any video/audio URL, or when asked to search YouTube.
remotion-video
Use when you need to render an actual video file with Remotion — React compositions, the Composition/Sequence/TransitionSeries graph, transitions, burned-in word-by-word captions from a transcript, automatic silence removal, b-roll overlays, headless CI renders, and a final MP4 or MOV. NOT writing the script, hook…
video-streaming-expert
Expert in video streaming technologies, HLS, DASH, adaptive bitrate streaming, CDN delivery, DRM protection, and video encoding/transcoding. Use when the user mentions video, streaming, media, WebRTC, multimedia, or HLS, or when the task involves Streaming Protocols, Adaptive Bitrate Streaming, FFmpeg Video…
media-transcoding
FFmpeg-based media transcoding workflows with preset-driven conversions, batch processing, and safe backups for web/mobile/archive outputs.
chinese-video-transcribe-pdf
A workflow that turns Chinese-language videos, with or without subtitles, into structured text and a Chinese PDF report. It can use videos from sources such as YouTube or local MP4 files.
vlog-auto-edit
A workflow that turns raw travel or everyday video clips into a finished vlog, a short edited video built around a person's experiences.