product-video-ads

Generates product advertisement videos from images via background removal, scripting, AI video generation, and audio synthesis.

53|10|Updated Jan 26, 2026
One-click install
npx skills add https://github.com/Chase-Zzx/product-ad-video-generator --skill product-video-ads-chase-zzx
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: product-video-ads
Source: https://github.com/Chase-Zzx/product-ad-video-generator
Command: npx skills add https://github.com/Chase-Zzx/product-ad-video-generator --skill product-video-ads-chase-zzx

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires rembg, moviepy, Pillow, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve? Creating professional product advertisement videos traditionally requires video editing skills, copywriting, voiceover recording, and expensive production resources. This Skill automates the entire pipeline from a single product image to a finished ad video. ## Core Features & Use Cases - End-to-End Video Pipeline: Removes image backgrounds with rembg, identifies product attributes, writes ad scripts at 280-320 characters per minute, designs storyboards, generates video segments via an external video API, and merges everything with TTS narration and BGM using moviepy. - Product Fidelity Control: Enforces strict prompt constraints so the product's shape, structure, and appearance remain identical to the reference image, using camera movement and lighting instead of product deformation. - Use Case: An e-commerce seller uploads a smartwatch photo and requests a 15-second tech-style ad; the Skill cuts out the product, writes a script, designs 3-4 storyboard shots, generates each video segment, produces one continuous TTS narration, and outputs the final merged MP4. ## Quick Start Upload a product image and ask the agent to generate a 30-second advertisement video for it, specifying your target audience and preferred style.

Frequently Asked Questions about product-video-ads

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

FAQPage Schema
How do I generate a product ad video from an image?

Upload a product image and specify the desired duration, style, and target audience. The workflow removes the background with rembg, writes a script and storyboard, generates video segments through a video generation API, then merges them with TTS narration and BGM using moviepy.

What Python libraries are needed for AI video ad generation?

This workflow requires rembg 2.0.50 for background removal, moviepy 1.0.3 for video concatenation and audio mixing, and Pillow 10.0.0 for image handling. Install them with pip before running the scripts.

How do I configure the video generation API?

Copy .env.example to .env and set VIDEO_API_KEY and VIDEO_API_URL for your video generation provider. The generate_video.py script reads these credentials and supports both synchronous responses and asynchronous task polling.

Why must TTS narration be generated as a single audio file?

Generating TTS in one pass keeps the narration speed uniform at 280-320 characters per minute across the whole video. Splitting TTS per storyboard segment and concatenating causes uneven pacing and audible jumps between segments.

How does the workflow prevent product distortion in generated videos?

Video prompts are restricted to visual elements only and must declare that the product's shape, structure, and appearance remain exactly as shown in the reference image. Products stay static while camera movements like push-in, rotation, and orbit provide the motion.

What video lengths and formats does the ad generator support?

The workflow supports 15, 30, and 60 second advertisements with script word counts scaled accordingly. Final output is an MP4 file encoded with libx264 video and AAC audio, with optional BGM mixed at adjustable volume.