liblib-ai-gen

What This Skill Does

The liblib-ai-gen skill empowers OpenClaw to interface directly with the LiblibAI API, providing high-fidelity image and video generation capabilities. By leveraging cutting-edge models like Seedream4.5 for static imagery and the Kling suite for dynamic video synthesis, this skill allows users to create professional-grade visual assets via natural language commands. The skill manages the end-to-end lifecycle of a generation request: it handles the authentication via environment variables, constructs the API payloads, initiates the async generation tasks, and provides an automatic polling mechanism to retrieve result URLs once rendering is complete.

Installation

To integrate this skill into your OpenClaw environment, ensure you have your LiblibAI credentials (LIB_ACCESS_KEY and LIB_SECRET_KEY) ready. Execute the following command in your terminal:

clawhub install openclaw/skills/skills/xtaq/liblib-ai-gen

Once installed, verify your environment variables are configured correctly in your system profile or .env file. The skill relies on the scripts/liblib_client.py executable, which requires a standard Python 3 environment.

Use Cases

This skill is ideal for creative professionals, content creators, and developers needing to generate assets on the fly. Common use cases include:

• Generating high-resolution conceptual illustrations using the advanced Seedream4.5 model.
• Creating short-form video content from descriptive text prompts.
• Animating static photography or digital art using the Kling Image-to-Video capabilities.
• Automating batch asset generation for prototyping or rapid design iteration.

Example Prompts

1. "Generate a high-resolution 2048x2048 image of a futuristic cyberpunk city at sunset using the Seedream model."
2. "Create a 5-second video of a rocket launching into space using the Kling v2-6 model with sound enabled."
3. "Take this image [URL] and animate it into a 5-second video where the character turns their head."

Tips & Limitations

• Performance: The API enforces a 1 request-per-second limit for task submission and supports up to 5 concurrent tasks. Plan your batch jobs accordingly.
• Image Quality: For Seedream4.5, note the minimum pixel requirement of 3,686,400 pixels for optimal results.
• Expiration: Be aware that generated assets stored on LiblibAI servers will expire after 7 days; ensure you download your assets promptly.
• Advanced Features: Always specify the pro mode when using turbo models like kling-v2-5-turbo or kling-v2-6 to avoid errors, as these models are optimized for high-performance processing.