Claude Skills

Use when you have a spec or requirements for a multi-step task, before touching code

Use PROACTIVELY when setting up end-to-end testing, debugging UI issues, creating visual regression suites, or automating browser testing. Uses Playwright with LLM-powered visual analysis, screenshot capture, and fix recommendations. Zero-setup for React, Next.js, Vue, Node.js, and static sites. Not for unit testing, API-only testing, or mobile native apps.

Multi-layer verification of implementation against spec and intent. Use when: (1) After BUILD phase in 5D-SDD workflow, (2) Implementation is complete and needs validation, (3) User asks to 'test,' 'verify,' or 'check' the implementation, (4) Before considering a feature done. This phase catches errors at multiple levels and routes fixes appropriately.

Convert refined plan into technical specification with architectural decisions. Use when: (1) After REFINE phase in 5D-SDD workflow, (2) User is ready for technical design, (3) Moving from 'what' to 'how,' (4) User asks for 'technical spec,' 'architecture,' or 'design doc.' This phase bridges business intent and implementation detail.

Show what's new in Kata since the installed version, displaying changelog entries, checking for Kata updates, or reviewing recent changes. Triggers include "what's new", "whats new", "show changes", "changelog", "recent changes", and "what changed".

This skill should be used when the user asks to "implement stories", "work on PRD", "autonomous coding", "overnight run", "implement features while I sleep", "AFK mode", "hands-off development", "run unattended", "how does /go work", "batch development", "iterate on stories", or discusses HITL vs AFK approaches. Covers the go loop execution modes (HITL vs AFK), when to use each, and progression from cautious to confident autonomous development.

Use PROACTIVELY when working on multiple branches simultaneously or creating parallel Claude Code sessions. Automates git worktree creation with prerequisite checking, development environment initialization, and safe cleanup. Supports single worktree, batch creation, and worktree removal. Not for non-git projects or branch management without worktrees.

Use this skill when configure kata workflow toggles and model profile. Triggers include "settings".

Use PROACTIVELY when user needs to create a new GitHub repository or set up a project with best practices. Automates repository creation with four modes - quick public repos (~30s), enterprise-grade with security and CI/CD (~120s), open-source community standards (~90s), and private team collaboration with governance (~90s). Not for existing repo configuration or GitHub Actions workflow debugging.

Perl開発支援（Mojolicious/DBI/Carton）。 モダンPerl、perlcritic、Test::Moreでガイド。

Codex CLI を使用してコードやアーキテクチャをレビューし、改善提案や技術的な相談に対応する。コードレビュー、実装方針の相談、バグ調査、リファクタリング提案、解消が難しい問題の調査に使用する。

Use PROACTIVELY when setting up OpenTelemetry monitoring for Claude Code usage tracking, cost analysis, or productivity metrics. Provides local PoC mode (full Docker stack with Grafana) and enterprise mode (centralized infrastructure). Configures telemetry collection, imports dashboards, and verifies data flow. Not for non-Claude telemetry or custom metric definitions.

A comprehensive skill for A/B split test engineering, following Neil Patel's methodology. It covers test prioritization (ICE scoring), statistical significance, sample size, hypothesis formation, variable isolation, multivariate vs. A/B decisions, test documentation, and conversion lift calculation. Use this skill to optimize web pages, ads, and emails for higher conversion rates.

Build production-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use this skill when the user wants to create, scaffold, or implement an MCP server for any API or service, in any language. Triggers include requests to "build an MCP server", "create MCP tools", "implement MCP", "add tools for [service]", "connect [service] via MCP", or any task involving MCP server development, tool design, or MCP architecture decisions.

Use when you have a written implementation plan to execute with continuous task execution

Read, update, and format Google Sheets spreadsheets via API. Use when users need to (1) Read cell data from spreadsheets, (2) Update cell values, (3) Append rows to sheets, (4) Format cells (background color, borders, text style), (5) Create pivot tables, or (6) Export data to JSON/CSV format.

Convert expanded thinking into a concrete, multi-perspective plan. Use when: (1) After SPAR phase in 5D-SDD workflow, (2) User is ready to define what will be built, (3) User asks to 'write up the plan' or 'document what we're building,' (4) Transitioning from exploration to commitment. This phase creates the authoritative description of intent before technical specification.

Skip swarm analysis, minimal memory creation