Skill UI

This skill provides comprehensive patterns for building highly reliable applications that interact with external Large Language Model (LLM) APIs, such as Perplexity Sonar. It implements critical reliability techniques including Circuit Breakers, Model Fallback, Streaming Timeout handling, and Caching layers. Use this to ensure your application maintains high availability and graceful degradation even when external services experience intermittent failures or variable performance.

This skill enables the development of robust, event-driven pipelines interacting with the Perplexity Sonar API. Since Perplexity lacks native webhooks, this system simulates event patterns using advanced techniques: Streaming Server-Sent Events (SSE) for real-time data, job queues (like BullMQ/SQS) for asynchronous batch research, and cron-triggered monitoring for scheduled news and trend alerts. It provides patterns for building complex, reliable, and scalable AI applications.

Expert in designing and implementing advanced Retrieval-Augmented Generation (RAG) pipelines. This role focuses on optimizing the entire information retrieval process—from document chunking and embedding model selection to vector database architectures and hybrid search strategies. The goal is to bridge the gap between raw knowledge documents and accurate LLM understanding, drastically reducing hallucination and enhancing context relevance for enterprise applications.

Runs external LLM CLIs (OpenAI Codex or Google Gemini) to review uncommitted changes, branch diffs, or specific commits, providing an independent second opinion before merges or releases.

Serve production-grade LLM APIs with vLLM, leveraging PagedAttention, continuous batching, quantization, and tensor parallelism to maximize throughput while keeping latency, GPU usage, and scalability under control for OpenAI-compatible deployments.

A comprehensive guide for developing, optimizing, and standardizing skills for AI agents. This resource provides a decision tree for choosing the right architecture (Tier 1-3), incorporating best practices like CSO (LLM SEO) and Anti-Rationalization, and ensuring compliance through rigorous testing and checklists. Essential for robust AI system building.

MarkItDown converts PDFs, DOCX, PPTX, XLSX, HTML, JSON, XML, archives, YouTube links, images via OCR, and audio via transcription into clean Markdown, providing token-efficient text ready for LLMs and documentation workflows.

A comprehensive guide covering patterns and architectures for automating code execution using LLMs. It spans simple sequential pipelines (e.g., daily dev scripts) to advanced, RFC-driven multi-agent Directed Acyclic Graph (DAG) orchestration. Ideal for building CI/CD style continuous development pipelines, managing complex state, and automating quality gates without human intervention.

MolyKit is a comprehensive toolkit designed for building robust, cross-platform AI chat applications. It provides core utilities for integrating large language models (LLMs) like OpenAI, handling real-time SSE streaming responses, and managing complex cross-platform asynchronous patterns (PlatformSend, spawn()). The kit includes ready-to-use widgets for messages, prompt inputs, and avatars, making it ideal for developers creating sophisticated, multilingual chat experiences.

This skill outlines a comprehensive methodology for designing, building, and deploying robust LLM-powered applications. It covers assessing task-model fit, designing modular pipelines (acquire -> prepare -> process -> parse -> render), and managing state using file systems to ensure scalability and reliability in AI systems.

A set of behavioral guidelines derived from Andrej Karpathy's observations to help users writing, reviewing, or refactoring code with Large Language Models (LLMs). It guides developers to avoid overcomplication, make surgical changes, explicitly surface assumptions, and define verifiable success criteria, promoting clean, simple, and robust code development.

This skill guides the development of EarLLM One, a complex Android application. The app captures real-time audio from Bluetooth earbuds, transcribes the voice using STT, sends the text input to a Large Language Model (LLM) via a robust pipeline, and plays the response back using TTS. It requires deep knowledge of Kotlin, Android development, Bluetooth profiles (A2DP/HFP), and advanced audio routing.