A unified API skill for generating asynchronous AI videos from text or images using top-tier models like Bytedance Seedance and Google Veo.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install oatda-generate-video
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install oatda-generate-video using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is OATDA AI Video Generation Skill?
This skill leverages the OATDA unified API to provide a seamless interface for generating AI videos through a variety of leading providers. By integrating models such as Alibaba Wan, MiniMax, and OpenAI Sora, it enables developers to create cinematic clips and animations directly from text descriptions or reference images. As a core component of the Openclaw Skills ecosystem, it handles the complexities of multi-model mapping and asynchronous task management, ensuring a robust workflow for AI-driven media creation.
The skill is designed for high-performance environments where asynchronous processing is essential. It allows users to submit a request, receive a task ID, and then poll for the finished video asset, making it ideal for integration into larger automated pipelines or agentic workflows.
OATDA AI Video Generation Skill Use Cases
- Create cinematic 5-10 second video clips from descriptive text prompts.
- Transform static images into dynamic animations using advanced image-to-video capabilities.
- Rapidly prototype visual content across different providers like Google Veo or Bytedance Seedance using a single interface.
- Automate video production workflows within an AI agent environment powered by Openclaw Skills.
- Generate marketing assets or social media clips with specific aspect ratios and resolutions.
How OATDA AI Video Generation Skill Works
- The skill resolves the OATDA API key from environment variables or the local credentials configuration file.
- It maps the user's request to the appropriate provider and model ID, defaulting to Seedance if no specific model is mentioned.
- A POST request is sent to the OATDA asynchronous endpoint including the prompt and optional parameters like resolution or duration.
- The system returns a unique taskId and a polling URL, confirming that the generation process has started.
- The user or agent monitors the task status via the polling endpoint until the video generation is marked as complete and the file URL is provided.
OATDA AI Video Generation Skill Setup
To get started with this component of Openclaw Skills, ensure you have curl and jq installed on your system.
- Obtain an API key from the OATDA dashboard.
- Configure your credentials by setting an environment variable:
export OATDA_API_KEY="your_api_key_here"
- Alternatively, place your key in a credentials file at
~/.oatda/credentials.json. - Verify the setup by querying available video models:
curl -s -X GET "https://oatda.com/api/v1/llm/models?type=video" -H "Authorization: Bearer $OATDA_API_KEY" | jq
OATDA AI Video Generation Skill Data Schema & Taxonomy
The skill manages data through structured JSON payloads and model-specific parameters. Below is a summary of the core data structure:
| Field | Description | Format |
|---|---|---|
| taskId | Unique identifier for tracking the video generation task | String |
| status | Current state of the video (pending, processing, completed) | String |
| provider | The AI service provider (e.g., bytedance, google, minimax) | String |
| model_params | Object containing specific settings like duration, ratio, and audio generation | JSON Object |
| pollUrl | Endpoint used to check the progress of the asynchronous task | String |
For image-to-video tasks, the model_params field must include public URLs for first_frame_image and optionally last_frame_image to define the visual transition.
OATDA AI Video Generation Skill Advanced Features
- Dynamic Model Discovery: Automatically query the latest available video models and their supported parameters via the OATDA API.
- Image-to-Video (I2V) Support: Create smooth transitions by providing start and end frame images as reference URLs.
- Fine-grained Control: Customize resolution (720P, 1080P), aspect ratios (16:9, 9:16, 1:1), and specific model parameters like camera movement and audio generation.
- Robust Error Handling: Integrated logic for handling rate limits, content policy violations, and invalid API keys within the Openclaw Skills framework.
- Asynchronous Polling: Optimized for long-running generation tasks with a reliable task ID and status tracking system.
SKILL.md
Loading
Related Openclaw Skills
Quick Music
A zero-dependency CLI tool for instantaneous music searching and streaming link retrieval.
Beta Backtester
A professional quantitative backtesting framework for validating and optimizing trading strategies across stocks, crypto, and forex.
Command Guardian
A safety-first preflight guard that analyzes shell and infrastructure commands for risks, secret leakage, and destructive side effects before execution.
Feishu MCP
A sophisticated Model Context Protocol integration that enables AI agents to perform full-lifecycle document and table operations within Feishu.
OATDA List Models
A powerful utility for listing and filtering available AI models across 10+ providers including OpenAI, Google, and Anthropic.
OATDA Image Generation
A powerful unified interface for generating AI artwork and illustrations from leading providers like OpenAI, Google, and Alibaba.
Featured*
