tomada-claude-plugins

Create ASCII wireframes, user flow diagrams, and cross-cutting specifications for UI/UX visualization. Use PROACTIVELY when user mentions wireframe, UI design, UX flow, screen layout, screen design, user flow, or asks to visualize screens. Examples: <example>Context: User needs screen design user: 'Create wireframes for this feature' assistant: 'I will use designing-wireframes skill' <commentary>Triggered by wireframe request</commentary></example> <example>Context: After requirements are detailed user: 'Let me design the screens' assistant: 'I will use designing-wireframes skill' <commentary>Triggered by screen design request</commentary></example>

Plan and create GitHub Issues with agile/scrum methodology. Analyze requirements, identify parallel work, manage dependencies, and suggest git worktree strategies. Use PROACTIVELY when creating tickets, planning sprints, breaking down features, organizing issues, identifying parallel tasks, managing dependencies, or working with GitHub Issues, scrum, agile planning, ticket breakdown, worktree planning. Examples: <example>Context: User wants to plan implementation user: 'チケットを作っていこう' assistant: 'I will use agile-ticket-planner skill' <commentary>Triggered by ticket creation request</commentary></example> <example>Context: User has requirements user: 'この要件をIssueに分割して' assistant: 'I will use agile-ticket-planner skill' <commentary>Triggered by issue breakdown request</commentary></example>

Clarify ambiguous requirements through structured questioning and produce detailed, implementation-ready specifications. Acts as PdM to identify unclear points and resolve them via AskUserQuestion. Use PROACTIVELY when user mentions requirements, specs, PRD, refine, detail, clarify requirements, or asks to detail/refine app ideas. Also trigger when the user is about to jump into implementation with a vague or incomplete spec — even if they haven't explicitly asked to "refine" anything. Examples: <example>Context: User has rough idea user: 'Help me refine this app spec' assistant: 'I will use refining-requirements skill' <commentary>Triggered by spec refinement request</commentary></example> <example>Context: User jumps to implementation with vague spec user: 'Let me start building this feature' assistant: 'I will use refining-requirements skill first to clarify the requirements' <commentary>Proactively triggered before ambiguous implementation</commentary></example> <example>Context: Planning phase user: 'Let

Design UI/UX concepts for apps and web services through systematic research and questioning. Use PROACTIVELY when designing app interfaces, determining visual direction, creating design systems, choosing color schemes, or establishing UX patterns. Examples: <example>Context: User wants to design an app user: 'UI/UXのデザインコンセプトを決めたい' assistant: 'I will use ui-ux-designing skill' <commentary>Triggered by design concept request</commentary></example> <example>Context: User building new feature user: 'この機能の見た目どうしよう' assistant: 'I will use ui-ux-designing skill' <commentary>Triggered by visual design question</commentary></example>

Author a concise, self-contained prompt to hand to Claude Code's /goal command in a separate, unattended session. Researches the target project (CLAUDE.md, dev rules, baseline build/test state, existing patterns) and drafts a goal with measurable done-criteria, transcript-verifiable checks, scope + anti-cheat constraints, and stop rules. Outputs to chat by default; externalizes bulky context to ~/.claude/goal-prompts/<slug>/ only when inlining would bloat it. Decides autonomously, asking only when a goal-defining axis is genuinely ambiguous. Use PROACTIVELY when the user mentions /goal, goal prompt, ゴールプロンプト, /goal に渡す, 自走/無人実行させたい, unattended run, 完了条件を決めて goal を回す, or setting up a long-running autonomous Claude session. Examples: <example>Context: User wants an unattended run user: '/goal に渡すプロンプトを作って' assistant: 'I will use authoring-goal-prompts skill' <commentary>goal-prompt authoring request</commentary></example>

Claude Code ベストプラクティスの対話形式セットアップウィザード。Use PROACTIVELY when setting up new projects, configuring Claude Code, adding agents/rules/hooks, or when user mentions "bootstrap", "setup claude", "configure claude code", "add agents", "add rules", "セットアップ", "初期設定", "ブートストラップ". Examples: <example>Context: User wants to set up Claude Code user: 'Claude Codeをセットアップしたい' assistant: 'I will use bootstrapping-claudecode skill' <commentary>Triggered by setup request</commentary></example>

Capture Claude Code terminal output and interactive UI screens via tmux. Launch a single Claude Code pane, send commands or keystrokes, capture screen output, and save to markdown files. Use PROACTIVELY when capturing Claude Code output, recording terminal sessions, screen capture, UI navigation capture, tmux capture, recording Claude responses, キャプチャ, 画面記録, 出力記録. Examples: <example>Context: User wants to capture Claude output user: 'Helloの出力を記録して' assistant: 'I will use capturing-claudecode skill' <commentary>Triggered by capture request</commentary></example>

Create, audit, refactor, or troubleshoot Claude Code skills via explicit command. Modes: new (scaffold a fresh skill), audit (run validate + audit scripts and write a report), convert (turn an existing doc into a skill), troubleshoot (diagnose activation, frontmatter, or link issues), add-subagents (introduce multi-phase orchestration). Accepts a skill name or path as the second argument.

Orchestrate multiple Claude Code sessions via tmux. Self-contained with setup/cleanup scripts, queue-based task tracking, and report-based completion detection. Use PROACTIVELY when running parallel tasks, multi-agent workflows, orchestrating Claude instances, tmux automation, or when user mentions orchestration, parallel execution, multi-agent, オーケストレーション, 並列実行, マルチエージェント. Examples: <example>Context: User wants parallel execution user: '2つのタスクを並列で実行して' assistant: 'I will use tmux-orchestrating skill' <commentary>Triggered by parallel execution request</commentary></example>

Convert HTML, Markdown, or plain text to WordPress Gutenberg block format for SWELL theme. Use PROACTIVELY when converting content to WordPress, SWELL theme formatting, Gutenberg blocks, WordPress記事変換, SWELL形式, ブログ記事変換, or working with wp:paragraph, wp:heading, wp:list, swl-marker. Examples: <example>Context: User has markdown or HTML to convert user: 'Convert this article to WordPress format' assistant: 'I will use converting-to-wordpress-swell skill' <commentary>Triggered by WordPress conversion request</commentary></example>

とまだ式の技術記事執筆・改善スキル。Use PROACTIVELY when user mentions「とまだ式」「記事を書いて」「Zenn記事」「技術記事執筆」「リライト」or requests article writing, rewriting, improvement in Japanese. Plan modeで構成ヒアリング後に執筆開始。Pre-Writing Phaseで構成案を事前評価し、サブエージェントによる7軸並列評価で95点以上を目指す。 Examples: <example>Context: User requests article user: '記事を書いて' assistant: 'I will use tomada-writing skill' <commentary>Triggered by article writing request</commentary></example> <example>Context: Rewrite request user: 'この記事をリライトして' assistant: 'I will use tomada-writing skill' <commentary>Triggered by rewrite request</commentary></example>

プロジェクトのClaude Code harness設定を監査し、ハーネスエンジニアリングのベストプラクティスに基づいて改善提案を生成する。CLAUDE.md、Hooks、リンター、ADR、プリコミット設定を包括的にチェック。明示的に呼び出された場合のみ実行。Use when user mentions 'harness audit', 'harness review', 'ハーネス監査', 'harness engineering', 'ハーネスエンジニアリング', 'harness check', 'ハーネスチェック', 'プロジェクト監査', 'harness改善', 'hooks見直し', 'CLAUDE.md監査'. Examples: <example>Context: User wants to audit project harness user: 'このプロジェクトのharnessを監査して' assistant: 'I will use ccl-harness-engineering skill' <commentary>Triggered by harness audit request</commentary></example> <example>Context: User wants best practice check user: 'ハーネスエンジニアリングに沿っているか確認' assistant: 'I will use ccl-harness-engineering skill' <commentary>Triggered by best practice review</commentary></example>

Example multi-phase skill that reviews a pull request by spawning two specialist sub-agents in parallel — one for security, one for performance — each primed with a numbered checklist. Demonstrates the orchestration patterns documented in claudecode-skill-creating. NOT a real working skill — read it for structure, do not invoke it.

Agent Teamsで3体のチームメイト（提案者・批評者・統合者）による3ラウンドのマルチ視点ディスカッションを実施。オープンなテーマや質問に対して直接対話と引用反論を通じた合意形成で実行可能な提案を生成し、指示があれば修正も実行。Use PROACTIVELY when ディスカッション、議論、検討、意見を聞きたい、複数視点で議論、レッドチーミング、チーム議論、提案がほしい、妥当性検証、是非を議論、メリデメ検討、方針を決めたい、構成の妥当性、設計レビュー

Analyze code structure, patterns, and complexity without making modifications. Use when reviewing code, planning refactoring, or conducting read-only code analysis.

Generate friendly greetings in multiple languages with cultural context. Use when creating welcome messages, international greetings, or multilingual user interfaces.

Explain HTTP status codes with examples and best practices. Use when working with APIs, debugging HTTP responses, or learning about REST API design and HTTP protocol.

Validate project structure, dependencies, and configuration files. Use when setting up projects, checking project health, or troubleshooting configuration issues.