SkillAgentSearch skills...

creatify-ai-skill

A reusable AI Agent Skill and Python SDK Wrapper for Creatify AI - programmatic video generation.

Install / Use

claude mcp add alphaparkinc -- npx -y github:alphaparkinc/creatify-ai-skill

If the server publishes to npm under a different name, use that package instead — check the repo README.

About this skill
🔌

MCP Server

Model Context Protocol server

Quality Score

71/100

Supported Platforms

Claude Code
Claude Desktop

creatify-ai-skill

GenPark AI Agent Skill -- # Creatify AI Skill - Python SDK & Developer Integration Guide

This repository contains the Creatify AI Developer Skill — a comprehensive Python client SDK wrapper and developer documentation designed to automate the generation of high-impact video ads, AI avatars, text-to-speech, and asset generation using the official Creatify AI REST API (v1 and v2).


🚀 Product Capabilities & Overview

Creatify AI is an advanced AI-powered marketing asset platform that automates the end-to-end creation of video ads. Key technical features include:

  • URL-to-Video API: Automatically parses any product link (Shopify, Amazon, Etsy, etc.), extracts visual assets, writes compelling marketing scripts, synthesizes voiceovers, and generates high-converting short-form videos.
  • AI Avatars & Lipsync (v1 & v2): Converts text or audio files into lifelike talking-head videos featuring highly realistic avatars.
    • v1: Simple single-segment actor with a single voice and script.
    • v2: Multi-scene narrative configuration allowing scene-by-scene editing of actors, voices, backgrounds, and caption styling.
  • Text-to-Speech (TTS) & Voice Cloning: Converts text into ultra-realistic spoken audio in multiple accents or clones specific voices.
  • Custom Templates: Enables mass-personalization by programmatically injecting dynamic content into pre-made video templates.
  • Aurora (Single-Image Talking Head): Converts a single face photo and voice clip into an expressive talking-head video.
  • IAB Banner Images: Instantly generates IAB-compliant ad banners in multiple standardized sizes from a single product image.

💳 Credit Usage & Cost Matrix

Creatify AI operates on a subscription-based credit system. API-based calls consume credits as follows:

| Endpoint Feature | Credit Cost | Description | | :--- | :--- | :--- | | Script Generation | 1 credit | Generates ad copy/scripts from a title or description. | | Product Image Preview | 1 credit | Generates a preview image for a product. | | Video Preview (URL/Avatar) | 1 credit | 30s quick video preview (fast, cost-effective). | | Video Rendering (URL/Avatar) | 4 credits | 30s final high-resolution video render. | | Full Video Generation (Direct) | 5 credits | 30s video created directly without preview steps. | | Text-to-Speech | 1 credit | Generates 30s of high-quality speech audio. | | IAB Banner Images | 2 credits | Generates multi-size banner ads from one image. |


🛠️ Installation & Setup

  1. Clone this repository:

    git clone https://github.com/<your-username>/creatify-ai-skill.git
    cd creatify-ai-skill
    
  2. Install dependencies:

    pip install -r requirements.txt
    
  3. Obtain your API Credentials from the Creatify Dashboard (Workspace Settings -> API Settings) and set them as environment variables:

    • Windows (CMD/PowerShell):
      $env:X_API_ID="your_api_id"
      $env:X_API_KEY="your_api_key"
      
    • Linux/macOS:
      export X_API_ID="your_api_id"
      export X_API_KEY="your_api_key"
      

💻 SDK Usage Reference

The creatify_ai.py client SDK wraps the REST endpoints cleanly. Here are the core code snippets:

1. Initialize Client & Check Balance

from creatify_ai import CreatifyClient

client = CreatifyClient(api_id="your_api_id", api_key="your_api_key")

# Check remaining credits
info = client.get_remaining_credits()
print(f"Remaining credits: {info['credits']}")

2. URL-to-Video Generation Workflow

The typical API workflow is asynchronous:

# Step 1: Create a link object from product page URL
link_obj = client.create_link("https://www.amazon.com/dp/B08P2H5LW2")
link_id = link_obj["id"]

# Step 2: Trigger Async Video Generation
job = client.generate_video_from_link(
    link_id=link_id,
    visual_style="modern",
    aspect_ratio="9:16",
    language="en"
)
job_id = job["id"]

# Step 3: Poll for final output MP4 (with built-in polling helper)
result = client.poll_job(job_id=job_id, job_type="video")
print(f"Finished Video URL: {result['video_output']}")

3. Multi-Scene AI Avatar (Lipsync V2) Workflow

Structure your video scene-by-scene with custom scripts, voices, backgrounds, and actors:

scenes = [
    {
        "script": "Introducing our brand new marketing skill wrapper!",
        "avatar_id": "avatar-uuid-1",
        "voice_id": "voice-uuid-1",
        "background_url": "https://example.com/bg1.jpg",
        "caption_setting": {"style": "normal-black"}
    },
    {
        "script": "It allows developers and AIs to automate video generation instantly.",
        "avatar_id": "avatar-uuid-1",
        "voice_id": "voice-uuid-1",
        "background_url": "https://example.com/bg2.jpg"
    }
]

# Generate lipsync v2 video job
job = client.generate_lipsync_v2(scenes=scenes, aspect_ratio="9:16")
job_id = job["id"]

# Poll for completed video URL
result = client.poll_job(job_id=job_id, job_type="lipsync", version="v2")
print(f"AI Avatar Video URL: {result['video_output']}")

🔬 API Endpoint Specifications

All endpoints use https://api.creatify.ai as their base URL and require X-API-ID and X-API-KEY headers.

| Method | Endpoint | Description | | :--- | :--- | :--- | | GET | /api/workspace/remainingcredits/ | Get current credit balance. | | POST | /api/links/ | Scrapes a webpage and creates a link resource. | | PUT | /api/links/{id}/ | Refines and updates link text/image assets. | | POST | /api/link_to_videos/ | Triggers a 5-credit video generation job. | | POST | /api/link_to_videos/preview/ | Triggers a fast 1-credit preview video. | | POST | /api/link_to_videos/render/ | Renders a final video using preview ID. | | GET | /api/link_to_videos/{id}/ | Polls status of a video job. | | GET | /api/personas/ | Gets list of available AI avatar actors. | | GET | /api/voices/ | Gets list of all available voices. | | POST | /api/lipsyncs/ | Generates simple 1-scene avatar video (v1). | | POST | /api/lipsyncs_v2/ | Generates complex multi-scene avatar video (v2). | | GET | /api/lipsyncs_v2/{id}/ | Polls status of a lipsync v2 job. | | POST | /api/text-to-speech/ | Synthesizes text into high-quality audio file. |


📜 License

This project is licensed under the MIT License.

Related Skills

View on GitHub
GitHub Stars9
CategoryDevelopment
Updated1mo ago
Forks0

Languages

Python

Security Score

75/100

Audited on Jul 10, 2026

1 medium2 low