Comprehensive guide for Zod 4+ schema validation patterns, covering form validation, API response validation, environment variables, and TypeScript type inference with modern syntax.

Create engaging Vietnamese stories including fairy tales, fables, and children's narratives with automated tools for story generation, character development, and language validation.

End-to-end type safety with Zod runtime validation, tRPC type-safe APIs, Prisma ORM, and TypeScript 5.7+ features. Build fully type-safe applications from database to UI with exhaustive type checking and branded types.

Deep multi-framework reasoning tool using Gemini for complex problem analysis, idea evaluation, and decision-making through structured thinking frameworks like First Principles, Systems Thinking, and Pre-Mortem analysis.

A comprehensive skill for developing Roblox games using TypeScript with roblox-ts compiler. Covers type-safe Roblox API usage, project setup, package management, TypeScript-to-Luau compilation, and best practices for building performant games.

Intelligent auto-completion system for prompt_toolkit applications, supporting word completion, fuzzy matching, nested commands, file paths, and custom completers with async support.

Develop and optimize AI prompt strategies for code generation across multiple models (Claude, GPT, Gemini, DeepSeek, Grok). Specializes in creating model-specific optimization techniques, prompt quality improvement, and visual layout to code conversion.

Convert audio files to text using NVIDIA's Parakeet ASR model optimized for Apple MLX framework. Provides fast, high-quality on-device speech-to-text transcription with timestamps in SRT format, ideal for podcasts, interviews, and meetings.

A comprehensive guide for creating output-styles that transform Claude's behavior and personality. Helps with authoring, designing, and reviewing persona modes, style files, and behavior modifications using best practices and templates.

Enterprise AI platform for building, customizing, and deploying generative AI models and agents. Includes NeMo Retriever for RAG pipelines, NeMo Customizer for fine-tuning, NeMo Guardrails for safety, and tools for data curation, evaluation, and multi-agent orchestration.

Analyze SAE (Sparse Autoencoder) decoder weights to understand feature output influence, importance, and similarity. Reveals what outputs a feature promotes or suppresses, providing complementary insights to activation analysis for interpretable AI.

Comprehensive LLM model evaluation and ranking system that compares language models across performance, cost, and technical dimensions. Provides benchmark-based rankings, task-specific recommendations, cost-effectiveness analysis, and helps users select the best model for their needs among GPT-4, Claude, Gemini, Llama, and other LLMs.

Analyzes and explains IC10 code behavior in plain English. Breaks down Stationeers scripts into understandable explanations covering device setup, execution flow, and key behaviors.

Reviews Go REST API projects using Gin, Echo, Fiber, or Chi frameworks. Analyzes router organization, middleware patterns, request validation, error handling, and API documentation to ensure best practices and code quality.

Designs evidence-based practice exercises for programming education using cognitive science principles like retrieval practice, spaced repetition, and interleaving. Creates varied exercise types (fill-in-blank, debug-this, build-from-scratch, AI-collaborative) with difficulty progression, test cases, and assessment rubrics aligned to learning objectives.

Expert guidance for building high-performance, type-safe web applications with ElysiaJS framework on Bun runtime. Covers REST APIs, GraphQL, WebSocket, TypeBox validation, Eden client, authentication, official plugins, testing, and production deployment.

Build AI-powered features in Directus including chat interfaces, content generation, smart suggestions, and copilot functionality using OpenAI, Anthropic Claude, and other AI providers with real-time communication, vector search, and RAG systems.

Automatically analyzes Claude Code conversation history to provide RAG-powered semantic search, intelligent insight reports, and pattern detection. Processes JSONL conversation files to extract development patterns, file hotspots, tool usage analytics, and knowledge highlights with optional interactive dashboard visualization.

A multi-agent collaborative discussion system featuring 10 AI experts working together to provide diverse perspectives on complex topics. Ideal for technical architecture discussions, product strategy planning, and brainstorming sessions requiring multi-angle expert insights.

Generates production-ready BAML applications from natural language requirements. Creates complete type definitions, functions, clients, tests, and framework integrations for data extraction, classification, RAG, and agent workflows with 95%+ compilation success.

Comprehensive guide for building LLM applications with LangChain, LangGraph, and FastAPI. Covers chain orchestration, RAG systems, prompt engineering, agent development, ML model integration, streaming responses, and best practices for production AI deployment.

AI-powered shadcn/ui component library expert with Context7 integration for building modern React applications. Provides intelligent design system architecture, theme customization, accessibility compliance, and performance optimization for enterprise-grade UI development.

A comprehensive Gemini 3 Pro client with Thinking mode enabled. Supports text queries, multi-format file analysis (MP4, PDF, images), YouTube video analysis, image generation/editing, and Google Search grounding. Uses browser cookies for authentication without requiring API keys.

Automatically intercepts and optimizes prompts using APE, OPRO, and DSPy patterns with embedding-indexed learning. Improves complex tasks, technical outputs, and reusable content through iterative refinement and performance tracking.

Designs digital experiences optimized for ADHD brains using neuroscience research and UX principles. Specializes in reducing cognitive load, solving time blindness, creating dopamine-driven engagement, and implementing compassionate design patterns that work with executive dysfunction.

Downloads YouTube video transcripts using yt-dlp with automatic fallback to Whisper AI transcription. Supports manual subtitles, auto-generated captions, and audio transcription with deduplication and format conversion.

Automated YouTube video caching system with semantic search capabilities. Automatically detects YouTube URLs, verifies cache status, and provides operations for ingestion, search, and metadata retrieval using Qdrant vector database and archive storage.

Transcribes audio and video files to text using OpenAI's Whisper CLI with contextual grounding from markdown files to improve accuracy for technical terms, proper nouns, and domain-specific vocabulary.

Supervised fine-tuning toolkit for instruction-following models using SFTTrainer. Supports chat template mapping, multi-turn conversation synthesis, and standardized dataset formatting (ShareGPT, Alpaca, OpenAI) with optimized processing.

Memory-efficient Direct Preference Optimization for aligning language models with human preferences using paired chosen/rejected data, without requiring a separate reference model. Optimized for low VRAM environments with FP8 support.