core-vitals-fixer

What This Skill Does

The core-vitals-fixer skill is an automated performance audit agent designed to optimize web application metrics, specifically targeting Core Web Vitals (LCP, CLS, and FID). By leveraging the GPT-4o-mini engine, the tool performs a deep scan of your project's source directory, identifying specific antipatterns in your codebase that contribute to poor loading performance, visual instability, and sluggish input responsiveness. Instead of just highlighting issues, it provides actionable code-level refactoring suggestions, allowing you to resolve complex performance bottlenecks without manually debugging audit logs. It is purpose-built to streamline the transition from failing Lighthouse scores to high-performance, SEO-friendly web experiences.

Installation

To integrate this skill into your environment, use the OpenClaw CLI tool. Run the following command in your terminal to install it directly from the OpenClaw repository:

clawhub install openclaw/skills/skills/lxgicstudios/core-vitals-fixer

Ensure you have your OPENAI_API_KEY set in your environment variables, as the agent requires this to perform the intelligent analysis of your code.

Use Cases

This skill is ideal for several development scenarios:

• SEO Rescue: When your site is experiencing ranking drops due to poor performance scores in Google Search Console.
• Performance Auditing: When conducting routine maintenance on a production codebase to ensure continued adherence to modern web standards.
• New Site Optimization: When building a high-traffic landing page and you want to verify that your component architecture follows performance best practices before deployment.
• Refactoring: When legacy codebases have accumulated heavy dependencies or inefficient rendering patterns that cause "layout shift" or slow interactive states.

Example Prompts

1. "Fix the CLS issues in my ./src/components folder, specifically looking for images without explicit aspect ratios."
2. "Analyze the ./app/ directory for LCP bottlenecks and give me a priority list of code changes to reduce the time to render the main hero section."
3. "Help me improve the FID score for my user dashboard; identify which JS bundles might be blocking the main thread during interaction."

Tips & Limitations

To get the most out of the core-vitals-fixer, prioritize fixing LCP issues first, as these usually provide the highest impact on user perception and search rankings. Always ensure your code is under version control before applying suggestions, as automated refactors should always be verified in a staging environment. Note that this tool requires external access to the OpenAI API and performs deep reads of your source files to generate its recommendations. It does not replace browser-based synthetic monitoring but acts as a powerful companion for static analysis.