ComfyUI-skill-public transforms text into AI workflows
ComfyUI-skill-public is a new open-source tool that connects an OpenClaw agent directly to ComfyUI for natural language control. Users can describe exactly what they want, and the system handles the workflow creation, eliminating the need for manual node wiring or writing Python scripts.
It was designed to solve the inefficiencies of batch processing and repetitive tasks. The developer built it to handle requests like generating multiple image variations or comparing prompts, allowing the agent to manage the queue automatically without constant user oversight.
Streamlined control and batch processing
- Translates natural language requests into executable workflow JSON.
- Handles batch operations such as seed variations and resolution comparisons.
- Auto-discovers installed nodes and models to ensure compatibility.
- Diagnoses broken graphs and fixes errors before submission.
- Operates across local machines, remote servers, or cloud deployments.
Users who frequently run A/B tests or need to generate large asset libraries will find this tool helpful. It removes the friction of setting up complex node chains for every new idea, making it easier to experiment with different models and settings quickly.
Early release and setup details
While the project is still in its early phases, the developer states that the core workflow functions are stable. The system works by taking a user request, translating it into the necessary JSON format, and posting it to the ComfyUI API to execute the job.
A major advantage is that it requires no changes to your existing ComfyUI setup. The developer said,
'Still pretty early but the core workflow stuff is working.'
Users simply need to clone the repository into their OpenClaw skills environment and configure the connection to their server to get started.
Get ComfyUI-skill-public on GitHub.
