← Back to Skills Marketplace
jeremycooper2077

Video Analyzer

by JeremyCooper2077 · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
38
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install jc-video-analyzer
Description
Parse local video files into transcript and AI analysis. Extract audio, transcribe with faster-whisper, analyze with AI.
README (SKILL.md)

Video Analyzer

本地视频文件解析管线:提取音频 → 语音转文字 → AI 深度分析。

前置条件

  • ffmpeg — 系统 PATH 中可用
  • faster-whisperpip install faster-whisper
  • HF 模型下载 — 国内需设 HF_ENDPOINT=https://hf-mirror.com
  • 中转站 API — 从 openclaw.json 读取 SU2_API_KEY

工作流程

视频文件 → ffmpeg提取音频(16kHz WAV) → faster-whisper转写 → AI修正+分析

Step 1: 检查视频信息

ffprobe -v quiet -print_format json -show_format -show_streams \x3Cvideo.mp4>

Step 2: 提取音频

ffmpeg -i \x3Cvideo.mp4> -vn -acodec pcm_s16le -ar 16000 -ac 1 audio_16k.wav -y

Step 3: 转写(用脚本)

# Windows
$env:HF_ENDPOINT='https://hf-mirror.com'
python scripts/transcribe.py \x3Caudio.wav> \x3Coutput.txt>

# Linux/macOS
HF_ENDPOINT=https://hf-mirror.com python scripts/transcribe.py \x3Caudio.wav> \x3Coutput.txt>

参数:python scripts/transcribe.py \x3Caudio.wav> \x3Coutput.txt> [model=tiny] [language=zh]

Step 4: AI 分析(用脚本)

node scripts/analyze.js \x3Ctranscript.txt> \x3Coutput.md>

分析内容:修正转写稿 + 核心观点 + 结构化对比 + 启示总结。

可选:提取关键帧

ffmpeg -i \x3Cvideo.mp4> -vf "fps=1/60,scale=640:-1" frame_%03d.jpg -y

输出

  • audio_16k.wav — 提取的音频
  • transcript.txt — 带时间戳的原始转写
  • transcript_corrected.md — AI 修正后文字稿
  • analysis.md — AI 分析报告
  • frame_*.jpg — 关键帧(可选)

注意事项 / 踩坑记录

  • tiny 模型识别率一般,专业术语会有同音错误,依赖 AI 修正。可换 basesmall 提升质量。
  • python3 — Windows 上可能指向 Windows Store 别名,用 python 代替。
  • HF_ENDPOINT — 国内必须设 hf-mirror.com,否则模型下载极慢或失败。
  • 中转站 502 — 偶尔出现,重试即可。也可换其他模型。
  • 编码 — Python 输出时设 PYTHONIOENCODING=utf-8
Usage Guidance
Install only if you are comfortable using the configured AI relay for transcript analysis. Review or redact transcripts before running scripts/analyze.js on private, business, regulated, or confidential videos, and run ffmpeg examples in a dedicated output directory to avoid overwriting files.
Capability Tags
requires-sensitive-credentials
Capability Assessment
Purpose & Capability
The artifacts match the stated purpose: ffmpeg audio extraction instructions, local faster-whisper transcription, and AI-based transcript correction and analysis.
Instruction Scope
The remote AI analysis is manually invoked on a user-supplied transcript and output path, but the documentation could state more plainly that the transcript is sent to the configured API endpoint.
Install Mechanism
The package contains a SKILL.md and two helper scripts only; there are no install hooks, background workers, lifecycle scripts, or automatic execution paths.
Credentials
Reading an existing SU2_API_KEY from OpenClaw config and using network access is proportionate for AI analysis, but transcripts can contain sensitive content.
Persistence & Privilege
No persistence, privilege escalation, broad local indexing, credential storage, destructive automation, or hidden behavior was found; files are read and written only through explicit command arguments.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install jc-video-analyzer
  3. After installation, invoke the skill by name or use /jc-video-analyzer
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release: 本地视频解析与AI总结工具
Metadata
Slug jc-video-analyzer
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Video Analyzer?

Parse local video files into transcript and AI analysis. Extract audio, transcribe with faster-whisper, analyze with AI. It is an AI Agent Skill for Claude Code / OpenClaw, with 38 downloads so far.

How do I install Video Analyzer?

Run "/install jc-video-analyzer" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Video Analyzer free?

Yes, Video Analyzer is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Video Analyzer support?

Video Analyzer is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Video Analyzer?

It is built and maintained by JeremyCooper2077 (@jeremycooper2077); the current version is v1.0.0.

💬 Comments