neodomain-ai

What This Skill Does

The neodomain-ai skill serves as a powerful interface for the Neodomain AI platform, enabling OpenClaw users to generate high-quality visual content directly from their command-line or automated workflows. It provides a robust suite of tools for text-to-image, text-to-video, image-to-video, and sophisticated motion-controlled video generation. By integrating this skill, users can bypass the manual constraints of web-based AI platforms, allowing for batch processing, parameter-driven consistency, and reference-based generation using character or style sheets. It is designed for creators, developers, and researchers who require programmatically accessible AI generation capabilities.

Installation

To integrate this skill, use the ClawHub installer. Ensure you have your environment set up correctly:

1. Execute: clawhub install openclaw/skills/skills/bandwhite/neodomain-ai
2. Set your API credentials: export NEODOMAIN_ACCESS_TOKEN="your_access_token_here"
3. Optionally, verify installation by checking available models using the provided image_models.py or video_models.py scripts located in your local skill directory.

Use Cases

• Prototyping: Rapidly generate concept art or visual storyboards for game and film development using specific prompts and aspect ratios.
• Consistency Tracking: Utilize reference images to maintain character consistency across multiple generations.
• Automated Content Creation: Build pipelines that automatically turn user-submitted text prompts into branded video content or marketing assets.
• Batch Processing: Generate large quantities of images with fixed seeds or varied aspect ratios for dataset creation.

Example Prompts

1. "OpenClaw, generate a 2K resolution image of a cyberpunk city street at night, use the doubao-seedream-4-0 model, and output it in PNG format."
2. "Create a 16:9 video of a flowing waterfall using this reference image [URL] and apply a slight panning motion effect."
3. "Generate 4 variations of a futuristic office space with a clean aesthetic, excluding any clutter or low-quality artifacts."

Tips & Limitations

• Reference Images: You can chain up to 10 reference images for character consistency. Always provide high-quality source images for best results.
• Model Selection: Not all models support all features (e.g., specific video motion controls). Use the video_models.py script to verify model capabilities before starting long render jobs.
• Cost & Rate Limits: As an external API service, large batch requests (especially video) may consume quota quickly. Monitor your API logs to avoid reaching provider limits.
• Guidance Scale: Adjusting the --guidance-scale allows you to balance prompt adherence against creative freedom; use values around 7.5 for a standard, realistic output.