gem

Analyze PDFs, images, videos, and YouTube links via the Gemini API.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/rajshah4/my-agent-skills --skill gem
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gem
Source: https://github.com/rajshah4/my-agent-skills/tree/main/skills/gem
Command: npx skills add https://github.com/rajshah4/my-agent-skills --skill gem

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill enables multimodal content analysis by interfacing with Google's Gemini API to process and extract insights from documents and media.

Core Features & Use Cases

  • Multimodal analysis: process PDFs, images, videos and YouTube links to extract summaries, highlights, or answers.
  • Visual content understanding: identify objects, scenes, and text within images or frames.
  • Use Case: Quickly summarize long research papers by feeding PDFs and images to generate a concise abstract.

Quick Start

Use ai-gem to summarize a PDF like document.pdf or analyze an image like photo.jpg. Also supports YouTube URLs.

Frequently Asked Questions about gem

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I extract key entities and summaries from a PDF using multimodal AI?

To extract summaries from a PDF using multimodal AI, you interface with the Gemini API to process documents and generate concise abstracts. This approach enables extraction of key entities and highlights from research papers.

Can I analyze images and identify objects or scenes with Gemini?

Yes, you can analyze images to identify objects, scenes, and text using Gemini's multimodal capabilities. By processing visual content through the API, the model understands frames and extracts actionable insights from photos.

Do I need an API key to process YouTube links and videos?

Yes, processing YouTube links and videos requires the GEMINI_API_KEY to be set in your environment. You also need the hamel package installed to enable the multimodal processing capabilities for media extraction.

What's the best way to summarize long research papers that contain both text and images?

The best way to summarize research papers with text and images is using a multimodal AI approach. By feeding PDFs and images to the Gemini API, you can generate a concise abstract that captures both textual and visual insights.

Are there limitations when extracting insights from video frames?

The metadata does not specify explicit limitations for video frame extraction. However, processing videos requires the hamel package and a valid GEMINI_API_KEY to successfully interface with the Gemini API for scene analysis.