Comfyui Api

The ComfyUI API skill enables AI agents to interact with ComfyUI workflows programmatically, allowing for automated image generation and workflow management without manual interface interaction. It serves as a bridge between text-based agent commands and the complex node-based architecture of ComfyUI.

When to use it

• Automate repetitive image generation tasks by triggering predefined workflows via API calls.
• Integrate custom nodes and external models directly into an AI agent's execution pipeline.
• Manage workflow states, inputs, and outputs dynamically within a serverless or containerized environment.
• Scale generative art production by decoupling the user interface from the backend processing logic.

Key capabilities

• Programmatic access to ComfyUI node graph execution.
• Support for custom node integration and extension.
• Dynamic input parameter handling for image generation.
• Workflow state management and result retrieval.

Example prompts

• "Execute a specific ComfyUI workflow using the provided prompt and seed values."
• "Generate an image by calling the comfyui-api with custom parameters for resolution and style."
• "Retrieve the output of the last generated image from the active ComfyUI session."

Tips & gotchas

Ensure your ComfyUI instance is running as a server with API access enabled before attempting to connect. Verify that any custom nodes required by your workflow are installed on both the agent's configuration and the target ComfyUI instance to prevent execution errors.