Skill UI

This tool is designed to help optimize and refine prompts used with Large Language Models (LLMs). By structuring inputs, it ensures that the instructions given to AI systems are clear, comprehensive, and contextually rich, resulting in significantly higher quality and relevance of the generated output. Useful for developers, writers, and researchers.

Build a production-ready, AI-powered data collection agent for any public source (job boards, prices, news, etc.). This agent automatically scrapes data, enriches it using free LLMs (like Gemini Flash), and stores the results in databases such as Notion, Sheets, or Supabase. It runs on a free schedule using GitHub Actions and learns from user feedback, making it ideal for continuous data monitoring and tracking.

This skill helps detect NTLM relay attacks (T1557.001) by analyzing Windows Security Event ID 4624 (logon type 3) for NTLMSSP usage. It flags critical anomalies such as IP-to-hostname mismatches, evidence of LLMNR/NBT-NS poisoning, rapid cross-domain authentication, and verifies SMB signing status. It is essential for proactive threat hunting and security incident investigation in Active Directory environments.

PydanticAI is a robust Python framework that integrates Pydantic's type-checking and validation capabilities into Large Language Models (LLMs). It allows developers to build production-ready AI agents that guarantee structured, validated outputs, moving beyond simple raw strings. Key features include supported tool calling, dependency injection for testability, and multi-model compatibility across major providers like OpenAI, Anthropic, and Gemini. Use this when reliability and correct data structure are paramount.

This guide assists developers in designing and architecting Model Context Protocol (MCP) servers that integrate deeply with LLMs like Claude. It is not a coding skill, but a comprehensive consultation process that guides users through use case interrogation, determines the optimal deployment model (e.g., Remote HTTP, MCPB), and selects the appropriate tool-design pattern for seamless, robust integration.

A comprehensive security tool designed to detect prompt injection attacks targeting Large Language Models (LLMs). It employs a multi-layered approach combining regex signature matching, structural anomaly scoring, and advanced DeBERTa transformer classification. It is essential for securing input validation layers in chatbots, AI agents, and RAG pipelines, and for retrospectively auditing LLM interaction logs against OWASP standards.

This sophisticated tool generates publication-quality AI illustrations and diagrams for academic papers. It implements a multi-stage, iterative workflow, using advanced LLMs (like Claude and Gemini) to parse user requests, optimize complex layouts, verify academic styling (e.g., CVPR/NeurIPS standards), and render high-fidelity figures. Ideal for researchers needing professional, complex visuals for their manuscripts.

Collection of battle-tested PyTorch training recipes covering architecture selection, optimizers, LR scheduling, precision tips, and debugging to train or fine-tune LLMs, vision, diffusion, and biomedical models with high GPU throughput.

Behavioral guidelines for writing, reviewing, and refactoring code with LLMs, emphasizing minimalism, clear assumptions, surgical edits, and verifiable success criteria to avoid common machine-assisted coding mistakes.

This skill provides comprehensive management capabilities for AI Integrations within the Arize platform. Users can create, list, retrieve, and update credentials for various Large Language Model (LLM) providers, including OpenAI, Anthropic, Azure OpenAI, AWS Bedrock, and Vertex AI. These integrations are crucial for enabling Arize's advanced monitoring and evaluation features that require invoking external LLMs.

Describes how to analyze a repository, prioritize key docs, and author an llms.txt entry so LLMs immediately understand the project scope, documentation clusters, and configuration artifacts they should consult.

This skill guides the end-to-end process of building automated evaluation pipelines for Python applications powered by Large Language Models (LLMs). It teaches how to define evaluation criteria, instrument the application, build golden datasets, and run real-world evaluations using tools like pixie, ensuring the app's logic (not just the LLM) is rigorously tested. Ideal for QA, benchmarking, and improving LLM-based services.