An advanced image-to-video automation skill that leverages official RunningHub standard-model APIs to create cinematic animations from static images.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install runninghub-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 runninghub-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 RunningHub Video?
RunningHub Video is a specialized technical tool within the Openclaw Skills ecosystem designed for professional image-to-video generation. It provides a streamlined interface to interact with industry-leading models like Kling, Seedance, and Wan 2.2 through the RunningHub official API. This skill automates the entire video production lifecycle, from binary image uploads and task submission to status polling and local file synchronization.
By using this skill, developers and creators can bypass the complexity of manual API orchestration. It handles transient link expiration by immediately downloading results and provides fine-grained control over camera movement, transitions between start/end frames, and audio generation, ensuring a stable and efficient creative workflow.
RunningHub Video Use Cases
- Creating cinematic transitions between two static images (start and end frames).
- Automating bulk video generation from local image assets for marketing or social media.
- Converting AI-generated characters or landscapes into moving scenes with motion prompts.
- Integrating high-end video generation models like Kling v3.0 into local AI agent workflows.
- Generating high-resolution videos with synchronized audio using the Seedance 2.0 global model.
How RunningHub Video Works
- The skill identifies the user's intent to use RunningHub or specific models like Kling, Wan, or Seedance.
- Local image files are automatically uploaded to the RunningHub binary upload endpoint to generate temporary public URLs.
- A video generation task is submitted to the chosen model endpoint with custom prompts and parameters.
- The skill enters a polling loop, checking the task status via the query API at regular intervals.
- Upon task completion, the skill fetches the video URL and downloads the media file to a local directory before the source link expires.
RunningHub Video Setup
To use this Openclaw Skills component, ensure you have the RunningHub API credentials configured. Run the helper script from your terminal:
# Standard image-to-video generation
python "scripts/runninghub_video.py" \
--image "/path/to/source.png" \
--prompt "slow camera zoom, character smiling" \
--out-dir "./outputs"
RunningHub Video Data Schema & Taxonomy
The skill manages data through a structured CLI interface and local storage strategy:
| Data Type | Organization / Description |
|---|---|
| Inputs | Accepts local file paths, public URLs, or data URIs for images. |
| Task Metadata | Tracks taskId, model name, and parameter configurations for every job. |
| Binary Uploads | Uses RunningHub's /openapi/v2/media/upload/binary for local file handling. |
| Storage | Results are saved to a user-defined --out-dir to prevent data loss from expired URLs. |
| Polling | Status is monitored via /openapi/v2/query with configurable timeouts. |
RunningHub Video Advanced Features
- Support for start-to-end frame transition control for precise motion directing.
- Intelligent model shortcuts for Kling v3.0-std, Seedance 2.0, and Wan 2.2.
- Automated binary media handling for seamless local-to-cloud workflows.
- Configurable audio generation, resolution scaling, and real-person mode switches.
- Submit-only mode for asynchronous task management without blocking the agent.
SKILL.md
Loading
Related Openclaw Skills
Ipsen Biopharma Intelligence
A specialized technical knowledge skill providing deep-dive intelligence on Ipsen's biopharmaceutical operations, oncology portfolio, and market positioning.
Meiji Business Intelligence
Analyze Meiji Holdings' unique food-pharmaceutical business model and dominant market position in the Japanese confectionery industry.
Landing Page Conversion Copy Pro
A professional copywriting engine that generates complete, framework-based landing page copy to maximize conversion rates.
Eastman Chemical Intelligence Skill
A specialized data skill for AI agents focusing on Eastman Chemical's molecular recycling technologies and specialty material innovations.
LiteLLM Vertex AI for OpenAI Codex
Bridge the gap between OpenAI Codex CLI and Google Vertex AI Gemini models using LiteLLM as a protocol-translating proxy.
Compliant Review Reply Coach
A specialized service skill that coaches users on generating professional, tone-calibrated responses to customer reviews across all major e-commerce platforms.
Featured*
