Can this skill perform actual video editing?

Yes, it can execute FFmpeg commands to perform common editing tasks such as clipping specific segments, merging multiple videos, or splitting files.

Does transcribing video require an external API key?

Speech transcription is performed locally using OpenAI's Whisper model, which does not require an API key or an internet connection for the transcription itself.

How does Video Toolkit handle 'cannot read binary files' errors?

It uses FFmpeg to extract visual frames and audio tracks into formats Claude can interpret, bypassing the need to read raw binary video data directly.

How does the music identification feature work?

It uses a sequential workflow: Gemini Audio API first detects the presence of music and provides timestamps, then Shazam (shazamio) identifies the specific track and artist metadata.

Which video formats are supported by the toolkit?

The skill supports a wide range of formats including .mp4, .mov, .avi, .mkv, and .webm, provided FFmpeg is installed on the host system.

Video Toolkit

Name: Video Toolkit
Author: emdashcodes

byemdashcodes

•

Data Science & ML

Analyzes, transcribes, and edits video files using AI-powered frame extraction, local speech-to-text, and multi-modal audio understanding.

About

The Video Toolkit is a specialized Claude Code skill designed for comprehensive video processing and multi-modal analysis directly within your development environment. It solves the common 'cannot read binary files' error by leveraging FFmpeg for frame extraction and OpenAI Whisper for local, private speech transcription. By integrating Google Gemini for deep audio understanding and Shazam for music identification, this skill allows Claude to 'see' and 'hear' video content, enabling complex tasks like automated summarization, scene detection, foreign language translation, and programmatic video editing like clipping or merging.

Key Features

Automated music identification and timestamping via Shazam and Gemini API integration.
Native video editing capabilities including clipping, merging, and splitting via FFmpeg automation.
4 GitHub stars
Intelligent frame extraction using either time-based intervals or change-based scene detection.
Foreign language translation and non-speech audio analysis (e.g., detecting applause or ambient noise).
Multi-modal analysis combining visual frame extraction with local Whisper speech-to-text transcription.

Use Cases

Analyzing and summarizing screen recordings, technical presentations, or meetings to extract key visual and verbal insights.
Automating the preparation of video datasets by trimming clips and extracting specific audio segments programmatically.
Identifying copyrighted music and translating dialogue in video assets for localization and compliance checks.

About

Key Features

Automated music identification and timestamping via Shazam and Gemini API integration.
Native video editing capabilities including clipping, merging, and splitting via FFmpeg automation.
4 GitHub stars
Intelligent frame extraction using either time-based intervals or change-based scene detection.
Foreign language translation and non-speech audio analysis (e.g., detecting applause or ambient noise).
Multi-modal analysis combining visual frame extraction with local Whisper speech-to-text transcription.

Use Cases

Analyzing and summarizing screen recordings, technical presentations, or meetings to extract key visual and verbal insights.
Automating the preparation of video datasets by trimming clips and extracting specific audio segments programmatically.
Identifying copyrighted music and translating dialogue in video assets for localization and compliance checks.