mikefutiamikefutiaCommunityยท1 Agent Skills Included

claude-vision

Video analysis with scene breakdowns, transcripts, and timestamps

Analyzes any video file and returns a structured markdown report with summaries, scene-by-scene timestamps, transcripts, and key moments. Turns competitor ads, screen recordings, tutorials, and meeting videos into written breakdowns without manual rewatching. Built-in accuracy rules prevent invented narrators, fake transcripts, or guessed speaker names.
npx skills add mikefutia/claude-vision --all -g -y

All Skills in This Repository (1)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install claude-vision?โ–ผ

Run `npx skills add mikefutia/claude-vision --all -g -y` in your terminal to install the video analyzer skill globally.

How to analyze a video with Claude Code?โ–ผ

Point the skill at any video file, for example: /video-analyzer ~/Downloads/demo.mp4. It uploads the video to the Gemini API and returns a structured markdown report with timestamps and transcripts.

What video formats does claude-vision support?โ–ผ

It supports mp4, mov, webm, avi, mpeg, mpg, flv, wmv, 3gpp, and 3gp files of any length.

Do I need an API key to use claude-vision?โ–ผ

Yes. You need a free Gemini API key from Google AI Studio, set as the GEMINI_API_KEY environment variable, plus the google-genai Python package.

Can I use claude-vision for competitor ad teardowns?โ–ผ

Yes. Drop in a competitor's UGC ad and get a beat-by-beat breakdown with on-screen text, visual details, and key moments, without inventing narrators or dialogue.

Related Repositories in Marketing & Sales

View All in Marketing & Salesโ†’