ai-video-gen

Generate AI videos from text prompts and reference images via HeyGen and fal.ai.

46.2k|5.7k|Updated Mar 29, 2026
One-click install
npx skills add https://github.com/calesthio/OpenMontage --skill ai-video-gen-calesthio
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ai-video-gen
Source: https://github.com/calesthio/OpenMontage/tree/main/.agents/skills/ai-video-gen
Command: npx skills add https://github.com/calesthio/OpenMontage --skill ai-video-gen-calesthio

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill streamlines the creation of AI-generated video clips from plain-language prompts or reference images, removing the need to manually interact with multiple provider APIs and handle polling, provider selection, and basic configuration.

Core Features & Use Cases

  • Multi-gateway Support: Works with HeyGen and fal.ai gateways to access models like VEO, Kling, MiniMax, Runway, and Seedance.
  • Text-to-Video & Image-to-Video: Generate clips from descriptive prompts or animate a reference image with tail-frame guidance and aspect ratio control.
  • Provider Selection & Fallbacks: Uses a video_selector approach to compare availability, cost, and quality and fall back automatically when a provider is unavailable.
  • Use Case: Quickly produce social media verticals, cinematic trailers, or prototype motion clips by specifying prompt, provider preferences, and aspect ratio.

Quick Start

Use the ai-video-gen skill to generate a 16:9 cinematic clip by providing a descriptive prompt, optionally a reference image URL, and then poll the execution until the returned asset is ready.

Frequently Asked Questions about ai-video-gen

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

FAQPage Schema
How do I generate AI video from text prompts and reference images?

To generate AI video from text prompts, you provide a descriptive prompt and optionally a reference image URL. The Skill processes these inputs through HeyGen or fal.ai gateways, handling execution polling until your video clip is ready for download.

Can I use HeyGen and fal.ai for both text-to-video and image-to-video generation?

Yes, HeyGen and fal.ai support both text-to-video and image-to-video generation workflows. You can animate a reference image with tail-frame guidance or create clips entirely from descriptive text prompts using these provider gateways.

Do I need API keys to start generating AI videos with HeyGen or fal.ai?

Yes, you need either a HEYGEN_API_KEY or FAL_KEY environment variable configured. These API keys are required to authenticate requests and access models like VEO, Kling, MiniMax, Runway, and Seedance through the provider gateways.

What is the best way to handle provider selection when generating AI videos?

Provider selection is handled automatically using a video_selector approach that compares availability, cost, and quality across HeyGen and fal.ai. If your preferred provider is unavailable, the system falls back to an alternative gateway seamlessly.

Does AI video generation support custom aspect ratios for social media content?

Yes, AI video generation supports custom aspect ratios for social media verticals and cinematic trailers. You can specify your desired aspect ratio alongside your text prompt or reference image to ensure the output matches your target platform format.

Why does my AI video generation request keep polling without returning a result?

Polling continues until the provider gateway finishes processing your video generation request. Delays can occur due to provider availability or queue times, but the automatic fallback mechanism should switch to an alternative provider if the initial gateway becomes unresponsive.