What are the limits for video processing?

Using the 2M token context window, the skill can process up to 6 hours of low-resolution video or approximately 2 hours of standard resolution video.

Does it support image generation?

Yes, it includes full text-to-image generation, image editing, and multi-image composition capabilities using Gemini's specialized image models.

Which models does this skill support?

It supports the Gemini 2.5 and 2.0 series, including Pro, Flash, and Flash-Lite models for various performance and cost requirements.

How do I provide an API key for the skill?

You can set a GEMINI_API_KEY in your system environment variables, or within a .env file located in your project root or .claude directory.

Can I process YouTube videos with this skill?

Yes, the skill supports public YouTube URLs for video analysis, temporal Q&A, and scene detection.

AI Multimodal Processing

Name: AI Multimodal Processing
Author: xiaxingxiaowei1983

byxiaxingxiaowei1983

0•

数据科学与机器学习

Integrates Google Gemini's advanced multimodal capabilities to analyze audio, video, images, and documents directly within Claude Code.

The AI Multimodal Processing skill empowers Claude with the ability to perceive and generate complex media using the Google Gemini API. It provides a unified interface for transcribing hours of audio, analyzing long-form video, performing OCR on multi-page PDFs, and generating high-quality images from text prompts. By supporting Gemini 2.0 and 2.5 models, this skill enables sophisticated data extraction, scene understanding, and creative content generation, making it an essential tool for developers working with diverse data types or building multimedia-rich applications.

主要功能

01Comprehensive audio and video analysis with timestamped transcriptions

02High-fidelity document extraction from PDFs into structured JSON or Markdown

03Support for massive context windows up to 2M tokens for long-form media

040 GitHub stars

05Text-to-image generation and editing with controllable aspect ratios and styles

06Advanced vision capabilities including object detection and pixel-level segmentation

使用场景

01Building AI-driven content pipelines that generate images and transcribe audio assets

02Automating the extraction of structured data from complex business forms and PDF tables

03Summarizing long video meetings or YouTube tutorials with specific scene-level insights

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add xiaxingxiaowei1983/openclaw ai-multimodal

For use in Claude.ai and ChatGPT

主要功能

01Comprehensive audio and video analysis with timestamped transcriptions

02High-fidelity document extraction from PDFs into structured JSON or Markdown

03Support for massive context windows up to 2M tokens for long-form media

040 GitHub stars

05Text-to-image generation and editing with controllable aspect ratios and styles

06Advanced vision capabilities including object detection and pixel-level segmentation

使用场景

01Building AI-driven content pipelines that generate images and transcribe audio assets

02Automating the extraction of structured data from complex business forms and PDF tables

03Summarizing long video meetings or YouTube tutorials with specific scene-level insights