Ollama X Z Image Turbo

What This Skill Does

The Ollama X Z Image Turbo skill enables your OpenClaw agent to leverage the power of the x/z-image-turbo model hosted locally via Ollama. It provides a seamless pipeline for generating high-quality images directly from natural language commands. When a user requests a visual creation, the skill triggers a specialized Python script that interfaces with the Ollama engine, processes the prompt, and renders the result as a PNG file. Once generated, the agent utilizes the built-in WhatsApp integration to deliver the visual asset directly to the user's mobile device. This skill bridges the gap between conversational AI and multimedia generation, allowing for rapid artistic experimentation without leaving the OpenClaw environment.

Installation

To install this skill, ensure your local Ollama server is running at http://127.0.0.1:11434 and that you have pulled the x/z-image-turbo model using ollama pull x/z-image-turbo:latest. Once the prerequisites are met, run the following command in your terminal: clawhub install openclaw/skills/skills/eric51/ollama-x-z-image-turbo. The skill will automatically configure the required file paths in your workspace, specifically within the ~/.openclaw/workspace/tmp/ directory to handle image buffering before transmission.

Use Cases

• Creative Content Generation: Rapidly prototype visuals, concept art, or social media graphics.
• Visual Summarization: Generate images that represent complex topics discussed in chat logs.
• Interactive Entertainment: Engage users by dynamically responding to requests for custom artwork or illustrations.
• Prototyping: Quickly generate placeholders for UI/UX design elements using AI-assisted image generation.

Example Prompts

• "Génère une image d'un chat astronaute explorant une planète faite de fromage."
• "Crée-moi une illustration détaillée d'un coucher de soleil cybernétique sur une ville futuriste."
• "Dessine un robot minimaliste lisant un livre dans une forêt ancienne."

Tips & Limitations

To get the best results, remember that image generation is a resource-intensive process. Always ensure the pty=true flag is set when executing the runner script, as the Ollama engine requires a TTY environment to operate correctly. We recommend a timeout of at least 120 seconds to prevent the process from being killed mid-render. You can further customize outputs by adjusting the --width, --height, and --steps parameters in your command configuration. Note that high resolution and higher step counts will significantly increase generation time. Ensure sufficient disk space is available in your /workspace/tmp/ directory, as the skill periodically overwrites the ollama_image.png file.