Logo

What This Skill Does

The Logo skill provides an end-to-end framework for AI-assisted graphic design, specifically tailored for brand marks and app icons. It moves beyond simple prompting by implementing a rigorous validation loop that ensures outputs are professional, scalable, and practical. The skill integrates specialized prompt structures for different design needs—ranging from minimalist geometric icons to sophisticated iOS Liquid Glass styles—while providing a clear path for refining raw AI outputs into polished final assets. It is designed to act as a design consultant, helping users select the right model (Nano Banana Pro, Ideogram, etc.) based on their specific requirement for text accuracy, artistic flair, or technical layout.

Installation

To integrate this skill into your OpenClaw environment, execute the following command in your terminal: clawhub install openclaw/skills/skills/ivangdavila/logo. Ensure you have the latest version of the OpenClaw agent runtime to support the suggested model switching capabilities.

Use Cases

This skill is ideal for startups needing immediate branding, developers requiring high-fidelity iOS app icons, and content creators looking for rapid iterative logo design. It is particularly effective for users who struggle with AI 'hallucinations' in text rendering, as it provides a structured process for when to use specialized models like Ideogram versus when to manually add text to a raw icon.

Example Prompts

1. 'I need a minimalist logo for a fintech app called 'SwiftPay'. Use the Nano Banana Pro model and ensure it looks clear at 32px size.'
2. 'Generate a polished iOS app icon for a meditation app using the Liquid Glass style. The element should be a stylized lotus flower with a soft blue gradient.'
3. 'Help me brainstorm a logo strategy for a tech company; compare the pros and cons of using GPT Image 1.5 versus Midjourney v7 for a tech-focused geometric mark.'

Tips & Limitations

Always remember that AI output is merely a 'creative draft.' The validation loop is mandatory: if the design is not readable at 32px, it is not finished. For production assets, you must vectorize the output to SVG format and manually refine typography, as AI-generated text often contains artifacts. If the AI fails to produce a coherent result after 7 attempts, pivot your strategy—change the model or simplify the visual prompt elements significantly.