Skill UI

This skill defines the mandatory conventions and architectural rules for developing maintainable, cross-platform applications using Avalonia UI and the Zafiro toolkit. It enforces a functional-reactive MVVM approach, prioritizing explicit error handling via Result types, and mandates leveraging Zafiro's existing abstractions to ensure code quality and predictability.

A specialist guide for designing, implementing, and optimizing high-performance C++ applications using modern standards (C++20/23). Expertise covers advanced topics like concepts, coroutines, template metaprogramming, and zero-overhead abstractions. Use this when tackling complex memory management, concurrency bottlenecks, or integrating build systems via CMake, ensuring code meets industrial quality standards through rigorous sanitization and profiling.

This orchestrator skill provides comprehensive core principles for game development, covering topics like the Game Loop, design patterns (State Machine, ECS), performance budgeting, and input abstraction. It guides users by routing them to specialized sub-skills based on required platforms (Web, Mobile, PC), dimensions (2D, 3D), and specialty areas (AI, Networking, Art). Ideal for structuring complex game projects from concept to implementation.

A comprehensive guide for migrating Large Language Model (LLM) integrations from providers like OpenAI or Anthropic to Groq. It details technical steps including SDK replacement, model ID mapping, building provider abstraction layers, and implementing zero-downtime traffic shifting using feature flags. This ensures developers can leverage Groq's superior speed while maintaining service continuity.

This blueprint provides a scalable, production-grade reference architecture for advanced LLM applications using LangChain. It guides the implementation of layered design, provider abstraction (allowing seamless switching between LLMs like OpenAI and Anthropic), dynamic chain registry, robust RAG pipelines, and complex multi-agent orchestration. Ideal for building enterprise-level, maintainable, and highly flexible AI systems.

This guide provides comprehensive principles for modern game development, covering engine selection (Unity, Unreal, Godot), platform integration (Steam features, console compliance), input abstraction, and performance optimization techniques. It helps developers make informed decisions when building cross-platform games for PC and console environments.

Expert skill for writing, reviewing, and debugging idiomatic Rust code, focusing on memory safety and zero-cost abstractions. Specializes in advanced concepts like ownership patterns, explicit lifetime management, trait-based API design, and building concurrent async applications using Tokio. Use this when developing high-performance, reliable, systems-level software in Rust.

Expert skill for building high-performance, memory-safe systems, web services, and complex libraries using modern Rust (1.75+). Specializes in advanced async programming with Tokio, zero-cost abstractions, and mastering ownership and lifetime rules. Ideal for implementing scalable backend APIs, low-level networking, and performance-critical components where memory safety is paramount.

torchforge is Meta's advanced PyTorch-native library designed for agentic Reinforcement Learning (RL). It fundamentally separates core RL algorithms from complex distributed infrastructure concerns. This allows researchers to rapidly prototype and experiment with novel RL methods (like GRPO) using clean, PyTorch-native abstractions, while automatically handling large-scale tasks like distributed training across thousands of GPUs using Monarch and TorchTitan.

A comprehensive guide to idiomatic Rust patterns and best practices for building safe, high-performance, and maintainable applications. Covers core concepts like ownership and borrowing, robust error handling using thiserror and anyhow, state modeling with enums, zero-cost abstraction via traits and generics, and safe concurrent programming practices.

This skill acts as an anti-over-engineering guardrail, instructing AI coding agents to maintain strict scope control and minimalism. It guides the AI to only modify the files and lines explicitly requested by the user, avoiding unnecessary abstractions, unrequested refactors, or scope creep. Use it when you prefer the simplest, most direct solution and want to prevent the agent from 'grinding' the code base.

Run Hugging Face models directly in JavaScript/TypeScript for text, vision, and speech tasks without backend servers, using Transformers.js in Node.js or modern browsers with WebGPU, quantization, and pipeline abstractions for efficient inference.