A local image generation skill for Openclaw Skills that enables private, high-performance visual creation using ComfyUI workflows.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install comfy-ui
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install comfy-ui using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is ComfyUI Local Image Generation?
The ComfyUI local skill allows developers to harness the power of node-based image generation directly within their agent workflows. By connecting to a local ComfyUI instance, this skill ensures that all image generation remains private and utilizes your own hardware resources. This is a critical addition to the ecosystem of Openclaw Skills for users who require high-quality SDXL output without the costs or privacy concerns of external cloud APIs.
Whether you are using the default Juggernaut XL setup or your own complex API JSON workflows, this skill provides a seamless bridge. It handles the complexities of WebSocket handshakes and prompt queuing, allowing your agent to focus on the creative prompt logic while the local server handles the heavy lifting of diffusion models.
ComfyUI Local Image Generation Use Cases
- Generating private, high-resolution visual assets for development projects.
- Automating bulk image creation using custom SDXL workflows.
- Prototyping UI/UX designs with local AI-generated imagery.
- Integrating professional-grade image generation into local-first AI agent environments using Openclaw Skills.
How ComfyUI Local Image Generation Works
- The skill identifies the local server address via the COMFYUI_SERVER_ADDRESS environment variable.
- When a generation request is made, a Python helper script initiates a WebSocket and HTTP connection to the ComfyUI API.
- The skill prepares the workflow, injecting the user prompt and a random seed into the designated nodes.
- The prompt is queued on the ComfyUI server, and the skill monitors the execution status in real-time.
- Once the generation is complete, the resulting image is downloaded and saved to the local directory for immediate use.
ComfyUI Local Image Generation Setup
To get started with this skill for Openclaw Skills, follow these steps:
- Ensure ComfyUI is running on your network and Enable Dev mode is turned on in the settings.
- Set your environment variable to point to your ComfyUI server:
export COMFYUI_SERVER_ADDRESS="http://192.168.1.119:8189"
- You can test the generation using the internal script:
python3 scripts/comfy_gen.py "a workspace with neon lighting" $COMFYUI_SERVER_ADDRESS
- For custom workflows, place your exported API JSON files in the
workflows/directory.
ComfyUI Local Image Generation Data Schema & Taxonomy
The skill manages data across the following structures:
| Component | Location | Description |
|---|---|---|
| Workflows | /workflows/ |
Stores ComfyUI API JSON files for custom generation logic. |
| Output Images | /image-gens/ |
Directory where final generated PNG/JPG files are stored. |
| Configuration | Environment | Uses COMFYUI_SERVER_ADDRESS for networking and local pathing for script execution. |
| Metadata | Internal JSON | Maps prompt and seed inputs to specific node IDs within the workflow. |
ComfyUI Local Image Generation Advanced Features
- Custom Workflow Injection: Seamlessly swap between different node-based configurations by referencing JSON files in the workflows folder.
- SDXL Default Pipeline: Optimized for the Juggernaut XL model to provide professional-grade results out of the box.
- Automatic Seed Management: Handles random seed generation to ensure variety in every request unless specified otherwise.
- Local Sync Capabilities: Configurable to sync generated assets directly to local document folders for better file management within Openclaw Skills.
- Prompt Injection: Automatically identifies text nodes to insert dynamic agent-generated prompts into the local workflow.
SKILL.md
Loading
Related Openclaw Skills
Pushover Notify
A streamlined integration for sending instant push notifications to mobile devices via the Pushover API during AI-driven workflows.
Security Skill Scanner
An automated security auditing tool designed to scan Openclaw Skills for malicious patterns, credential theft, and unauthorized network exfiltration.
YouTrack Project & Billing Manager
A comprehensive integration for managing YouTrack projects, issues, and automated billing through time tracking data.
Prayer Times
A comprehensive tool for retrieving accurate Islamic prayer times and setting automated background reminders based on global calculation methods.
MolTunes
MolTunes is a decentralized marketplace where AI agents can publish, discover, and install skills while earning MOLT tokens through a secure, cryptographic economy.
Tube Summary
A powerful tool that searches YouTube for relevant videos and generates detailed textual summaries by analyzing video subtitles.
Featured*
