Skill UI

Guides senior architects through asking discovery questions, defining use cases, selecting patterns, and planning folder layout before crafting SKILL.md instructions for consistent agent behavior.

Guides developers through building Claude skills, covering when to trigger a skill, structuring SKILL.md metadata, and organizing bundled scripts, references, and assets for efficient workflow handoffs.

Guides authors through building Claude Code plugin skills by detailing required metadata, structure, resource types, and best practices for scripts, references, assets, and progressive disclosure so Claude can load workflows efficiently.

A comprehensive meta-skill for validating, testing, and scoring the quality of technical skills within an ecosystem. It performs deep structural checks (SKILL.md compliance), executes Python scripts for syntax, runtime, and output validation, and generates multi-dimensional quality scores using letter grades and tier recommendations. Ideal for new skill authorship, auditing existing components, or integrating rigorous QA gates into CI/CD pipelines.

Automates iterative use of the plugin-dev skill-reviewer agent to diagnose and fix Claude Code SKILL.md issues, driving repeated review/fix cycles until standards are met before emitting the completion marker.

This guide outlines a systematic process for transforming any recurring technical pattern, complex debugging solution, or workflow into a standalone, portable skill. It details the multi-step extraction workflow, including pattern identification, structured documentation generation (SKILL.md), and quality control checks to ensure the skill can be reused across different projects.

Coordinates discovery, matching, and workflow orchestration across agent ecosystems, automatically scanning SKILL.md, ranking relevant skills, and executing multi-step plans for complex requests.

Validates SKILL.md files against the agentskills specification and Anthropic best practices, catching structure, semantic, naming, and quality issues in a single read-only pass before publishing or debugging skills.

This comprehensive tool validates Claude Code plugins against official Anthropic specifications and internal enterprise standards. It performs two main types of checks: structural validation (checking `plugin.json` fields, file references, and permissions) and content validation (conducting deep grading using `SKILL.md`). Use this when building, auditing, or preparing plugins for marketplace submission to ensure compliance and quality.

Demonstrates the SKILL.md layout and frontmatter options for creating a user-invoked slash command, outlining argument parsing, tool usage, and documentation so developers can replicate the structure.

This skill analyzes accumulated usage logs within the ARIS research harness. It proposes structural optimizations to skill definitions (SKILL.md), default parameters, prompts, and workflow sequencing. It acts as an outer-loop mechanism, observing user interaction patterns to improve the core research framework itself, ensuring continuous and data-driven refinement of the entire development process.

Provides a minimal skill definition that only includes the required fields, showing how to structure SKILL.md for quick testing or deployment of basic examples.