Skill UI

Provides best practices for configuring Large Language Model (LLM) access, specifically for Anthropic Claude, across development, staging, and production environments. This guide details setting up separate API keys, selecting appropriate models (e.g., Haiku, Sonnet, Opus), and managing unique rate limits for robust, scalable, and secure LLM integration in enterprise systems.

This guide details the process of deploying production-grade inference services on CoreWeave using KServe. It covers setting up autoscaling, GPU scheduling, and implementing scale-to-zero functionality for efficient ML model serving endpoints. Ideal for MLOps engineers deploying large language models (LLMs) or other complex AI models in a production environment.

This guide details how to set up and execute large-scale, distributed GPU training jobs on the CoreWeave platform. It covers both single-node multi-GPU setups and multi-node training utilizing PyTorch DDP and Kubernetes resources (A100/H100). This workflow is ideal for fine-tuning large language models (LLMs) or running complex deep learning models that require high-performance, shared-storage compute clusters.

This guide demonstrates how to deploy GPU workloads on the CoreWeave platform using Kubernetes (kubectl). It covers deploying vLLM inference services for large language model (LLM) testing and running batch CUDA jobs for benchmarking. This is ideal for users running their first GPU workloads, testing inference pipelines, or verifying cluster access on bare-metal GPU nodes (A100, H100).

This skill acts as an expert Design Systems Lead, analyzing complex technical assets (HTML, CSS, screenshots) from existing projects. It synthesizes a comprehensive 'Semantic Design System' into a structured DESIGN.md file. This output serves as the authoritative 'source of truth' for AI prompting, ensuring that any new UI generated by tools like Stitch adheres perfectly to the established visual language, color palette, and component standards.

This skill automatically converts structured data specifications (such as CSV, JSON schemas, SQL DDL, or Protobuf definitions) into functional, production-ready code. It generates complete models, validation logic, full CRUD API endpoints, and unit tests in multiple languages (Python, TypeScript, Go, etc.), significantly reducing boilerplate coding time and ensuring type safety.

A comprehensive evaluation suite designed to benchmark the performance of Large Language Models (LLM) and Vision-Language Models (VLM) specifically for home security applications. It covers 143 diverse tests across 16 critical domains, including tool use, prompt injection resistance, contextual reasoning, alert triage, and detailed scene analysis, ensuring robustness for smart home AI systems.

This benchmark evaluates Vision-Language Models (VLMs) on detecting potential safety hazards within indoor environments. It uses static camera frames, simulating real-world fixed security camera monitoring. The test covers 40 scenarios across 5 critical categories, including fire/smoke, electrical risks, trips/falls, child safety, and falling objects, providing a rigorous measure of VLM's practical safety inspection capability.

Lambda is a compact, machine-native communication protocol specifically designed for agent-to-agent (A2A) messaging. It functions as a shared, unambiguous vocabulary built from 'atoms' (2-character codes), completely bypassing the need for human natural language interpretation. It is ideal for orchestrators, task delegation, and logging structured coordination signals (like heartbeats or state changes) in multi-agent systems. Crucially, Lambda is intended only for machine communication, not for user-facing dialogue.

A comprehensive index of 46 production-grade test automation skills, covering E2E, unit, mobile, BDD, and cloud testing across 15+ languages. This skill equips AI coding assistants to act as senior QA architects, generating structured, real-world test code for major frameworks like Playwright, Selenium, Cypress, and Jest. It supports project setup, CI/CD integration, and cross-platform debugging.

This guide provides advanced patterns for composing robust, multi-step Large Language Model (LLM) workflows using LangChain 1.0 core runnables. Learn to implement parallel retrievals (RunnableParallel), conditional routing (RunnableBranch), and state management. It also addresses common pitfalls like silent KeyErrors due to mismatched dictionary shapes, providing solutions using debug probes and type annotations for reliable RAG, classification, and extraction pipelines.

This advanced skill generates high-quality, natural speech from text entirely offline using the VoxCPM2 model. Supporting 30 languages, it offers three modes: standard text-to-speech, descriptive voice design, and advanced voice cloning from reference audio. All processing runs on the local device (optimized for Apple Silicon), ensuring zero API calls, zero cost, and maximum privacy. Ideal for creating professional voiceovers, podcasts, and video narration without internet dependency.