mulerouter

What This Skill Does

The mulerouter skill acts as a powerful interface for the MuleRouter and MuleRun multimodal APIs, enabling OpenClaw agents to interface with state-of-the-art generative AI models. It acts as an abstraction layer for complex tasks including Text-to-Image, Image-to-Image, Text-to-Video, and Image-to-Video generation. Beyond simple generation, it supports advanced editing features like video keyframe interpolation and VACE, allowing for sophisticated media manipulation workflows. By supporting industry-leading models such as Wan2.6, Veo3, and Nano Banana Pro, the skill provides a unified way to handle all your visual media production requirements within the OpenClaw environment.

Installation

Installation is streamlined through the ClawHub platform. Run the following command in your terminal: clawhub install openclaw/skills/skills/misaka43fd/mulerouter. Once installed, you must configure your environment variables. Ensure you create a .env file containing your MULEROUTER_API_KEY and either MULEROUTER_BASE_URL or MULEROUTER_SITE. Verify your installation by running the command uv run python -c "from core.config import load_config; load_config(); print('Configuration OK')". Ensure that uv is installed and the virtual environment is synced using uv sync to handle all model dependencies.

Use Cases

This skill is perfect for creators and developers who need to integrate AI media generation into their automated workflows. Use it for rapid prototyping of visual assets, creating promotional videos for digital marketing, generating consistent image sets for UI/UX design, or applying complex video transformations such as stylistic frame interpolation. Whether you are generating high-fidelity landscapes from text descriptions or transforming static images into cinematic video clips, mulerouter provides the infrastructure to execute these tasks at scale.

Example Prompts

1. "Generate a 5-second video of a cyberpunk city street with rain falling, using the Wan2.6 model."
2. "Transform this static landscape photo into a cinematic panning shot using keyframe interpolation."
3. "Create a 4k image of a futuristic robot holding a rose, using the Midjourney configuration in MuleRouter."

Tips & Limitations

Always check your model parameters before generating large media files to avoid unnecessary costs or long processing times. Use the list_params flag for each model script to ensure you are passing the correct arguments. Remember that MULEROUTER_BASE_URL takes precedence over MULEROUTER_SITE in your configuration. Because this skill interacts with external high-compute APIs, ensure your internet connection is stable to avoid partial downloads or timed-out API requests. Large video generations may take considerable time depending on the backend capacity.