Skill UI

Behavioral rules for LLMs writing or reviewing code, emphasizing caution, simplicity, surgical edits, and goal-driven execution to avoid common mistakes during implementation tasks, bug fixes, or refactors.

A comprehensive toolkit for evaluating the performance of code generation models. It supports benchmarking against industry standards like HumanEval, MBPP, and MultiPL-E across multiple languages. Use this tool to quantitatively compare different LLM coding abilities and measure code synthesis quality using pass@k metrics.

This guide details advanced methods for merging multiple specialized LLMs (e.g., combining math, coding, and chat capabilities) without the need for expensive retraining. Learn various techniques such as Linear Merging, SLERP, Task Arithmetic, TIES-Merging, and DARE. Use mergekit to create high-performing, specialized model variants quickly, significantly improving performance and reducing computational costs.

A comprehensive collection of battle-tested, structured prompts designed to elevate interactions with Large Language Models (LLMs). Covers advanced use cases including expert coding reviews, system architecture design, performance analysis, technical writing, and creative brainstorming, ensuring high-quality, actionable AI output.

Speeds up LLM inference by combining speculative decoding with Medusa multiple heads and lookahead techniques so chatbots, code assistants, and other real-time workloads can run with 1.5-3.6× lower latency while preserving target-model quality.

A powerful tool to scaffold and ship production-ready Model Context Protocol (MCP) servers directly from OpenAPI specifications. It supports Python and TypeScript, ensuring schema validation, safe evolution, and robust tool definition generation, eliminating the need for manual wrapper coding. Use it to expose existing REST APIs to LLM agents or build shared tool integrations.

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 provides methods for reliably extracting typed, validated, and structured data from Large Language Models (LLMs), moving beyond simple text parsing. It covers best practices across major APIs (OpenAI's response_format, Anthropic's tool_use, Google's responseSchema) and local models. Key workflows include defining schemas (Pydantic/Zod), implementing constrained decoding, and building robust retry loops for production-grade data pipelines.

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.

This tool assesses whether coding agents adhere to defined skills, rules, or agent definitions. It automatically generates scenarios with varying prompt strictness, executes the agent, and generates a comprehensive report detailing the compliance rate of tool calls against predefined workflow specifications using advanced LLM classification.

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.

The Phoenix CLI is a powerful tool for debugging complex LLM and agent workflows. It allows users to fetch detailed traces, spans, and sessions, analyze errors, and structure findings using open and axial coding techniques. It is essential for investigating complex LLM failures, building robust failure taxonomies, and guiding the next steps in application evaluation after initial instrumentation.