Skill UI

Guides senior data scientists through designing controlled experiments, engineering structured features, training ML models, and translating results into decisions using Python, R, SQL, and tools like Scikit-learn, XGBoost, and MLflow.

This skill covers the full lifecycle of deploying machine learning models into production environments. It focuses on establishing robust MLOps pipelines, including model deployment workflows (e.g., Docker, Canary releases), feature store management (Feast), and comprehensive drift monitoring. Furthermore, it details advanced integration patterns for Large Language Models (LLMs) and building Retrieval-Augmented Generation (RAG) systems, emphasizing operational stability, performance validation, and cost optimization.

A comprehensive, model-agnostic toolkit for advanced LLM development. It enables iterative prompt optimization through static analysis (token usage, clarity), measures critical RAG metrics (relevance, faithfulness, coverage), and validates complex agent workflows and configurations. Essential for rigorous LLM quality assurance and cost management.

This central skill specializes in performing comprehensive threat modeling using STRIDE and calculating risk via DREAD scoring. It guides users through defining scope, identifying threats, and planning mitigations for new architectures. Crucially, it acts as a security router, directing specialized requests—such as pen-testing, incident response, cloud posture checks, and AI security reviews—to the appropriate dedicated skill.

Sentence Transformers provides 5000+ pre-trained models that generate sentence, text, and image embeddings for RAG, semantic search, clustering, and multilingual production workloads without relying on external APIs.

Automates SIEM rule generation within advanced security workflows, guiding through industry best practices, compliance checks, and threat modeling to deliver production-ready configurations verified against common standards.

SimPO (Simple Preference Optimization) is a state-of-the-art, reference-free method designed for aligning Large Language Models (LLMs) using human preference data. It is an efficient alternative to DPO and PPO, notably outperforming DPO without requiring a separate reference model. It is ideal for practitioners who need faster, simpler, and more resource-efficient fine-tuning for preference alignment.

Automates sklearn pipeline builder tasks within ML training, activating on relevant triggers to guide data prep, model training, hyperparameter tuning, and experiment tracking with production-ready code and standards validation.

SAELens provides a framework for training and analyzing Sparse Autoencoders (SAEs). SAEs decompose the dense, often polysemantic activations of large language models into sparse, monosemantic features. Use this when you need to discover the discrete, interpretable concepts a model has learned, study feature superposition, or analyze specific safety-relevant behaviors (like bias or deception) within deep neural networks.

This workflow provides an AI-powered platform for immersive conversation practice. It uses advanced language models (like GPT-4o) and real-time speech APIs to simulate real-life scenarios (e.g., ordering food, navigating). Users receive immediate, detailed feedback on pronunciation (phoneme level), grammar corrections, and vocabulary notes, making it an ideal core module for building language fluency and simulating professional tutoring experiences.

Automates splitting datasets into training, validation, and test subsets by generating and running Python code so you can quickly prepare data for machine learning model training and evaluation.

A comprehensive guide to leveraging Stable Diffusion models via the HuggingFace Diffusers library. This skill facilitates state-of-the-art image creation, including Text-to-Image generation from prompts, Image-to-Image translation, inpainting, outpainting, and advanced workflow building. Supports various models like SDXL and SD 3.0, ideal for professional creative and computational vision tasks.