Can Claude process videos using this skill?

Yes, this skill allows Claude to analyze MP4 files and YouTube links by leveraging Gemini's native video understanding capabilities.

Can I compare multiple different file types at once?

Absolutely. You can pass multiple files of different types, such as a PDF and a JPG, to the ai-gem tool for comparative analysis.

What are the requirements to use the gem skill?

You must have a valid Gemini API key set in your GEMINI_API_KEY environment variable and the 'hamel' Python package installed via pip.

How does the search functionality work?

The skill utilizes the --search flag in the ai-gem tool to perform live web searches, providing Claude with up-to-date information beyond its training data.

Is it suitable for large documents?

Yes, it is specifically designed for large documents and long-context tasks that require Gemini's expansive context window.

Gemini Multimodal Tools

Name: Gemini Multimodal Tools
Author: hamelsmu

byhamelsmu

•

データサイエンスとML

Integrates Google Gemini's multimodal capabilities to analyze PDFs, images, videos, and YouTube links directly within Claude.

概要

The Gemini Multimodal skill connects Claude to the powerful Google Gemini API via the ai-gem CLI. This allows Claude to transcend text-based limitations by processing complex visual and audio-visual data. It is the ideal solution for developers needing to summarize long YouTube tutorials, extract data from dense PDF documentation, perform visual audits on UI screenshots, or conduct real-time web searches to supplement coding tasks with current information.

主な機能

Multimodal analysis of images (PNG, JPEG, GIF, WebP) and video files
Integrated web search functionality for real-time information retrieval
Direct processing and summarization of YouTube URLs
Complex document analysis for PDFs and large text files
38 GitHub stars
Cross-file comparison across multiple formats simultaneously

ユースケース

Comparing frontend screenshots against design assets for visual regression testing
Extracting data from complex PDF specifications and technical diagrams
Summarizing technical video tutorials or conference talks into actionable notes

概要

主な機能

Multimodal analysis of images (PNG, JPEG, GIF, WebP) and video files
Integrated web search functionality for real-time information retrieval
Direct processing and summarization of YouTube URLs
Complex document analysis for PDFs and large text files
38 GitHub stars
Cross-file comparison across multiple formats simultaneously

ユースケース

Comparing frontend screenshots against design assets for visual regression testing
Extracting data from complex PDF specifications and technical diagrams
Summarizing technical video tutorials or conference talks into actionable notes