Skill UI

Speeds up LLM inference by combining speculative decoding with Medusa multiple heads and lookahead techniques so chatbots, code assistants, and other real-time workloads can run with 1.5-3.6× lower latency while preserving target-model quality.

This skill optimizes the performance of TwinMind's meeting AI operations. It fine-tunes transcription accuracy and speed using the advanced Ear-3 model, managing audio quality, and implementing caching strategies. Ideal for developers or system administrators needing to integrate, troubleshoot, or optimize high-volume meeting transcription workflows.

A comprehensive guide to building and deploying Vercel Edge Functions. These functions run on Vercel's global Edge Network (V8 isolates), providing ultra-low latency and zero cold starts. Learn how to use Web Standard APIs for critical tasks such as authentication, geolocation routing, A/B testing, and high-performance API middleware, drastically improving application speed and user experience.

This skill provides AI-assisted performance profiling within the Windsurf IDE. It analyzes collected profiling data to pinpoint performance bottlenecks, suggest specific optimizations, and predict the impact of code changes. Use it when optimizing application speed, refining complex algorithms, or general performance tuning in a coding environment.

Launches a Rapid Serial Visual Presentation (RSVP) speed reader. This tool displays text one word at a time, using ORP (Optimal Recognition Point) highlighting, which is designed to enhance comprehension and reading pacing. It is ideal for users who need a quick, focused reading aid for large blocks of text, such as prior conversation responses or provided arguments, rather than requiring document summary or formatting transformation.

This comprehensive tool performs a deep audit of image assets for maximum SEO performance and web speed. It checks critical elements like descriptive alt text, optimal file sizing, usage of modern formats (WebP/AVIF), proper responsive image implementation, and adherence to Core Web Vitals (CLS, LCP, lazy loading). Essential for developers and content teams aiming for top search rankings.

Autoresearch Agent runs an autonomous optimization loop that edits a target file, invokes a provided evaluation command, keeps metric-improving commits, drops failures, and repeats indefinitely so engineers can optimize speed, bundle size, prompts, or other measurable goals overnight.

Opinionated Docker development assistant that analyzes Dockerfiles, docker-compose setups, multi-stage builds, and container security to shrink images, speed builds, and enforce production-ready practices when users need optimized containers.

Bun is a high-performance, all-in-one JavaScript runtime and toolchain. It functions as a runtime, package manager, bundler, and test runner, offering significant speed improvements over traditional Node.js setups. It is ideal for modern, fast-paced development, new projects, and deployment environments like Vercel.

A comprehensive guide detailing Next.js 16+ usage with Turbopack. Turbopack, written in Rust, is an incremental bundler designed to drastically speed up development startup and Hot Module Replacement (HMR) by utilizing file system caching. This document guides developers on best practices, differentiating between the default Turbopack usage and when to revert to Webpack for debugging or compatibility.

This tool allows users to systematically evaluate and rank the outputs of multiple agents within an AgentHub session. It supports metric-based evaluation (running custom commands for measurable metrics like speed or resource usage), or an advanced LLM Judge Mode. The LLM Judge compares results based on criteria like correctness, simplicity, and overall quality, helping users determine the best-performing agent for the task.

Astro is a modern web framework designed specifically for building high-performance, content-heavy websites like blogs, documentation sites, and portfolios. Its core innovation is the Islands Architecture, which ships zero JavaScript by default, drastically improving performance and Core Web Vitals. It supports multi-framework components (React, Vue, Svelte) in a single project, making it ideal for content developers prioritizing speed and flexibility.