Get the FREE Ultimate OpenClaw Setup Guide →

image-generation

Scanned
npx machina-cli add skill ggozad/haiku.skills/image-generation --openclaw
Files (1)
SKILL.md
359 B

Image Generation

Use the generate_image tool to create images from text descriptions.

  • Generate images based on user prompts
  • Return the image as a markdown image reference for inline display
  • Requires a running Ollama instance with an image generation model

Source

git clone https://github.com/ggozad/haiku.skills/blob/main/skills/image-generation/haiku_skills_image_generation/image-generation/SKILL.mdView on GitHub

Overview

Generates images from text prompts using Ollama’s image models. The generate_image tool sends prompts to a running Ollama instance and returns the resulting image. Images are returned as markdown image references for easy inline display, making it ideal for quick concept art and visual brainstorms.

How This Skill Works

You provide a descriptive prompt to the generate_image tool, which communicates with a local Ollama server hosting an image-generation model. The tool returns the generated image, which the skill formats as a markdown image reference for inline display. A running Ollama instance with a compatible image-generation model is required for operation.

When to Use It

  • Idea capture: turn rough concepts into visuals from prompts.
  • UI/UX mockups: generate light concept art for interfaces.
  • Content visuals: create illustrations for blog posts or social posts.
  • Character and worldbuilding: quick concept art for stories or games.
  • Prototype assets: placeholder art for design reviews.

Quick Start

  1. Step 1: Ensure Ollama is running with an image-generation model reachable by the tool
  2. Step 2: Craft a clear, descriptive text prompt including style and lighting
  3. Step 3: Run generate_image with the prompt and insert the returned markdown image reference into your content

Best Practices

  • Write specific prompts with style, era, lighting, and color details.
  • Be aware of model capabilities and output resolution expectations.
  • Iterate by varying adjectives, nouns, and styles to explore alternatives.
  • Adhere to model safety guidelines and content restrictions.
  • Cache useful results and reuse prompts to speed up workflows.

Example Use Cases

  • A serene alpine landscape at sunrise, ultra-detailed, 4k, cinematic lighting
  • Retro-futuristic robot design, chrome surface, neon accents, vector art style
  • Watercolor illustration of a cat wearing a hat in a cozy cafe
  • Product concept: reusable glass bottle with minimal flat-design branding
  • Character concept: fearless explorer, 1920s adventure vibe, ink sketch

Frequently Asked Questions

Add this skill to your agents
Sponsor this space

Reach thousands of developers