vertex-video-reader

Analyze video content with Google Cloud Vertex AI's Gemini-3.1-Flash-lite-preview model.

4|Updated Jan 28, 2026
One-click install
npx skills add https://github.com/hwj123hwj/custom-skills --skill vertex-video-reader
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: vertex-video-reader
Source: https://github.com/hwj123hwj/custom-skills/tree/main/skills/vertex-video-reader
Command: npx skills add https://github.com/hwj123hwj/custom-skills --skill vertex-video-reader

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires google-cloud-storage, google-cloud-aiplatform, python-dotenv, and includes scripts (resource) components.

What problem does it solve?

This Skill provides an efficient way to analyze and understand video content by leveraging Google Cloud Vertex AI's lightweight video model.

Core Features & Use Cases

  • Video Analysis: Ask questions about the contents of a video, including transcription and description.
  • Action Extraction: Identify and describe actions or characters within a video.
  • Network Compatibility: Handles network proxies to ensure smooth API interaction.
  • Data Size Flexibility: Passes small to medium videos directly via Base64 encoding to the API.
  • Model Choice: Use different models based on the complexity and length of the content to be analyzed.

Quick Start

Use the vertex-video-reader skill to analyze a video file by executing the following command:

python /path/to/video.mp4 -p "What are the key actions in the video?"

Frequently Asked Questions about vertex-video-reader

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

FAQPage Schema
How do I analyze video content using Vertex AI to summarize and transcribe clips?

To analyze video content using Vertex AI, you can use a script that passes your video files to the Gemini-3.1-Flash-lite-preview model to generate summaries, transcriptions, and action descriptions directly through the API.

Can I use Vertex AI video analysis with small to medium video files via Base64 encoding?

Yes, Vertex AI video analysis supports small to medium video files by passing them directly to the API via Base64 encoding, allowing you to process video content without needing prior cloud storage upload for smaller data sizes.

Do I need a Google Cloud Vertex AI API key to extract actions and transcribe videos?

Yes, you need a valid Google Cloud Vertex AI API key to authenticate requests for extracting actions and transcribing videos, ensuring your script can securely interact with the video analysis model.

What is the best way to handle network proxies when running video analysis with Google Cloud Vertex AI?

The best way to handle network proxies for Google Cloud Vertex AI video analysis is to use a skill that explicitly manages network proxy compatibility, ensuring smooth and uninterrupted API interaction during video processing.

Are there limitations on video size when using the Vertex AI lightweight video model for transcription?

Yes, there are limitations on video size when using the Vertex AI lightweight video model; you must provide video files of appropriate size for the Base64 encoding method to function correctly during transcription and analysis.

When should I choose different models for video analysis on Google Cloud Vertex AI?

You should choose different models for video analysis on Google Cloud Vertex AI based on the complexity and length of your video content, allowing you to optimize processing for either lightweight summarization or detailed action extraction.