ai-multimodal

Process audio, image, video, and documents via Gemini's multimodal API.

Updated Jan 28, 2026

Search Tags:#ocr #multimodal #transcription #gemini #image-generation #video-analysis

Ready to run

npx skills add https://github.com/brixtonpham/claude-config --skill ai-multimodal-brixtonpham

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: ai-multimodal
Source: https://github.com/brixtonpham/claude-config/tree/main/skills/ai-multimodal
Command: npx skills add https://github.com/brixtonpham/claude-config --skill ai-multimodal-brixtonpham

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires google-genai, python-dotenv, Pillow, and includes scripts (resource) and references (resource) components.

What problem does it solve?

AI teams often struggle to coordinate and execute multimodal tasks across audio, image, video, and documents with a single, unified toolchain.

Core Features & Use Cases

Unified multimodal processing across audio (transcription, summarization), images (captioning, OCR, object detection, segmentation), video (summarization, scene detection, Q&A), and documents (PDF extraction and understanding), plus image generation.
Batch and pipeline oriented: run batch tasks across multiple media types using Google Gemini's multimodal API.
Extensible workflow: leverage optional scripts, references, and assets to customize processing and integrations.

Quick Start

Process a sample media item with the skill to run a multimodal analysis across audio, image, and video inputs.

Frequently Asked Questions about ai-multimodal

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

FAQPage Schema

How do I transcribe audio and extract data from PDFs in a single workflow?▼

Batch multimodal processing allows you to transcribe audio, caption images, and extract PDF data in one workflow. Using Google Gemini's API, you can run batch tasks across multiple media types to generate unified text outputs.

Can I perform OCR and image generation using the Gemini API?▼

Yes, you can perform OCR and image generation using the Gemini API. The multimodal processing capabilities support extracting text from images via OCR and generating new images directly from text prompts.

Do I need a GEMINI_API_KEY to analyze video and caption images?▼

Yes, a GEMINI_API_KEY is required to analyze video and caption images. You must configure this API key in your environment using python-dotenv to authenticate requests to Google Gemini for multimedia processing.

What is the best way to run batch video analysis and audio transcription?▼

The best way to run batch video analysis and audio transcription is through a pipeline-oriented approach. By leveraging Google Gemini's multimodal API, you can process multiple media types concurrently to summarize scenes and transcribe speech.

Does multimodal processing work with Python and Pillow for image captioning?▼

Multimodal processing works with Python and Pillow for image captioning. You can use Pillow to handle image preprocessing before sending the media to the Gemini API for tasks like object detection, segmentation, and captioning.

When should I not use a unified multimodal toolchain for media processing?▼

You should avoid a unified multimodal toolchain when your task requires highly specialized, single-format processing or offline execution. If your workflow cannot rely on external dependencies like the Google Gemini API, use dedicated single-format libraries instead.