A high-fidelity AI animation skill that transfers expressions and head poses from a source video to static face images.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install dreamact
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 dreamact 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 DreamAct Face Animation?
The DreamAct skill is a specialized tool designed to perform video-driven face animation by mapping facial dynamics from a source video onto static images. As a key component of the Openclaw Skills ecosystem, it allows developers to generate realistic lip movements, expressions, and head poses programmatically. By leveraging the NewportAI backend, this skill enables the creation of high-quality digital avatars and synchronized video content with minimal configuration.
Built for developers who require precise control over facial performance, this skill bridges the gap between static imagery and dynamic video. Whether you are building an automated content pipeline or enhancing an AI agent's visual output, integrating this into your Openclaw Skills workflow ensures access to state-of-the-art animation models.
DreamAct Face Animation Use Cases
- Create realistic talking head videos from a single portrait and a driving reference.
- Batch animate game character avatars using synchronized facial expressions via Openclaw Skills.
- Automate the production of lip-synced marketing materials for social media influencers.
- Generate consistent facial animations across multiple distinct target images using the same driving source.
How DreamAct Face Animation Works
- The user inputs a driving video URL and an array of target face images into the skill interface.
- The system validates the input requirements, ensuring the video duration is under one minute.
- A POST request is dispatched to the NewportAI DreamAct API to initiate an asynchronous animation task.
- The skill manages the polling process, checking the status from pending to completion while handling potential execution timeouts.
- Once processed, the skill returns the final URL of the animated video for use in your Openclaw Skills projects.
DreamAct Face Animation Setup
To get started with DreamAct, you must first obtain an API key from the Dreamface developer portal. Once you have your credentials, configure your environment using the following command:
openclaw config patch --json '{"skills": {"entries": {"dreamact": {"env": {"DREAMACT_API_KEY": "your-api-key-here"}}}}}'
DreamAct Face Animation Data Schema & Taxonomy
The skill organizes input parameters and metadata to ensure reliable animation outputs. Below is the schema used for processing:
| Field | Type | Description |
|---|---|---|
| video | String (URL) | The source driving video in MP4 format (max 1 minute). |
| images | Array (URLs) | List of images (JPG, PNG, WEBP, GIF) to be animated. |
| seed | Integer | Optional seed for reproducible generation (default: 42). |
| status | Integer | The lifecycle state (0=pending, 1-2=processing, 3=completed, 4=timeout). |
DreamAct Face Animation Advanced Features
- Multi-image processing: Animate an array of different faces simultaneously from a single driving video.
- Reproducibility: Use the seed parameter to maintain consistency across different animation runs.
- Asynchronous Polling: Built-in logic to handle long-running tasks that may exceed 3 minutes of execution time.
- OSS Integration: Support for obtaining upload policies to handle local files before passing them to the Openclaw Skills workflow.
SKILL.md
Loading
Related Openclaw Skills
Wallet Overview
A powerful tool for monitoring multi-chain wallet balances and activity history through a single interface.
Agentic Evaluation Patterns
Agentic Evaluation provides robust patterns and techniques for building AI agents that evaluate, critique, and refine their own outputs through iterative feedback loops.
AI Text Detector
An intelligent linguistic analyzer that estimates the probability of text being AI-generated based on structural and stylistic features.
mergeiq
A provider-agnostic complexity scoring engine that quantifies the depth of code changes using a multi-dimensional framework.
Dream Head Tail to Video (Wan2.1)
An AI-powered video generation tool that creates seamless 4-second animations between a provided first and last frame image using the Wan2.1 model.
Daily Digest (Gitee)
A Gitee-integrated automation tool that aggregates notifications, pull requests, and issues into a comprehensive daily work summary.
Featured*
