comfypet-jmcai-skill

What This Skill Does

The comfypet-jmcai-skill acts as a robust bridge between your OpenClaw AI agent and the JMCAI Comfypet desktop application. It allows you to trigger sophisticated image and video generation workflows directly from your AI agent interface. Instead of manual configuration in ComfyUI, this skill leverages pre-configured workflows, enabling you to query available processes, inspect parameters, submit execution tasks, and poll for final assets. It is specifically designed to handle the complexity of workflow orchestration, ensuring that your AI agent can generate high-quality visual content while maintaining standard parameter schemas.

Installation

To integrate this skill, ensure you have the JMCAI Comfypet desktop application running on your machine. Install the skill via the command line: clawhub install openclaw/skills/skills/allen-jmc/comfypet-jmcai-skill After installation, perform an initial setup by running the diagnostic tools:

1. python scripts/jmcai_skill.py doctor (Verifies connection to the desktop application).
2. python scripts/jmcai_skill.py registry --agent (Syncs the available workflows and their parameters with your agent).

Use Cases

This skill is perfect for creative and production-focused tasks, including:

• Generating cinematic images from descriptive text prompts.
• Creating short-form videos using advanced ComfyUI animation workflows.
• Automating batch generation of assets for creative projects where consistency is key.
• Interacting with local desktop hardware to offload heavy computational tasks to your GPU-enabled machine while keeping the agent interface clean.

Example Prompts

• "Can you generate a cinematic landscape image of a snowy mountain range using my pre-configured landscape workflow?"
• "I need a 5-second video of a futuristic car driving in the rain; use the best available video generation workflow."
• "What workflows are currently available in my Comfypet registry, and which one is best for artistic portrait generation?"

Tips & Limitations

• Strict Adherence to Schema: You must only use parameters exposed in the alias schema. Do not attempt to modify workflow nodes or target bindings, as this will lead to failures.
• Path Handling: You can pass local file paths for images. If you are using a bridge to another machine, the skill handles image uploads automatically.
• Error Management: If a process is queued, do not panic; the skill status command will track its progress until completion. If doctor fails, ensure the Comfypet app is open and your environment variables are set correctly. Always relay raw error messages from the CLI to ensure accurate troubleshooting.