kaijutale

Run external Ralph Loop (Geoffrey Huntley method) with clean context per iteration. Supports Simple mode (single PROMPT.md) and Full Ecosystem mode (specs/, fix_plan.md, AGENT.md, plan/build prompts). Unlike the official ralph-loop plugin which uses Stop hooks (same session, context pollution), this spawns fresh `claude -p` sessions each iteration. Use when user mentions "camone-ralph-loop", "external ralph", "clean ralph loop", "外部ラルフ", "ラルフループ実行", or wants to run iterative AI development with isolated context windows.

Execute semantic git commits with individual file staging, prefix selection, and split-commit enforcement. Use when committing code, managing git history, or when user mentions git/commit/push.

Detect project tech stack and generate a comprehensive, production-grade .gitignore. Covers 15+ languages, 20+ frameworks, security patterns, modern tooling, and monorepo support.

Install MCP servers into the current project scope using claude mcp add. Use when setting up a new project, when user mentions MCP setup, project MCP, MCPインストール, MCPセットアップ, or /installing-project-mcps.

Create a git worktree via gtr (git-worktree-runner) for a GitHub issue and generate a CLAUDE.local.md with issue context so that Claude Code launched in that worktree knows what to work on. Issue-specific context is written to CLAUDE.local.md (not .claude/CLAUDE.md) to avoid Git conflicts when merging PRs from multiple worktrees. Use when user mentions "gtr issue", "issue worktree", "issueでworktree", "issueに着手", "gtr-issue", or wants to set up an isolated worktree for a specific GitHub issue. Requires: gtr (git-worktree-runner) and gh CLI installed.

Generate detailed, agent-executable feature specifications by interviewing the user with AskUserQuestionTool. Transforms rough ideas into comprehensive SPEC.md files through phased deep interviews (5 phases) covering technical architecture, data design, API design, UI/UX, edge cases, security, performance, and tradeoffs. Produces hybrid specs with Acceptance Criteria and phased implementation tasks with dependencies. Use when user mentions "spec", "仕様書作成", "仕様駆動", "インタビューして", "SPEC.md", "feature spec", "仕様を書いて", "大きな機能を計画", or wants to plan a feature before implementation.

Claude Code workflow tips from Boris Cherny (creator of Claude Code) and the Claude Code team. Use when: setting up Claude Code, optimizing workflows, running parallel sessions, configuring CLAUDE.md, using skills/commands, subagents, hooks, MCP integrations, or learning best practices. Covers: git worktrees, plan mode, verification, permissions, Slack MCP, BigQuery, prompting tips, plugins, custom agents, sandboxing, keybindings, status lines, output styles, and customization.

プロジェクトの .claude/settings.json にPostToolUseフック（prettier, eslint等）を設定する。Use when user mentions hooks設定/フック設定/formatter hooks/PostToolUse/setup-hooks/setup-post-tool-use-hooks.

Project Domain Review — project-essence-orchestrator (master skill) からの明示呼出専用 (Skillツール経由のみ)。 context:fork + subagent:project-domain-reviewer で評価対象への「本プロジェクト固有ドメインルール (.docs/identity/)」観点からの照合分析を孫エージェントに委譲する。 description マッチによる自動誘発は想定しない (project-essence-orchestrator skill 内からのみ起動)。

4 領域 (ハーネス/skill/UI/プロジェクトドメイン) の essence + identity レビューを並列起動して統合判断を返す master skill (プロジェクトローカル)。 既存 essence-reviewing-orchestrator (グローバル 3 領域 = 普遍軸) に固有軸 (project-domain = .docs/identity/) を加えた 4 領域版。 context:fork 経由で各 reviewer を呼出し、メインコンテキストを汚さず純粋な領域別評価を実行する。 「project essence レビュー」「4 領域 essence」「全領域 + ドメインレビュー」「project-essence-orchestrator 起動」等で発動。 Skill ツール経由の明示呼出を推奨 (高コスト fork 起動のため意図せぬ発火を避ける)。

ひ孫エージェント検証用スキル。context:fork で孫コンテキストを切り離し、subagent:great-grandchild-inspector でひ孫を起動する。孫エージェント（recursive-grandchild-inspector）から呼び出すことを想定。

ひ孫検証用の孫起動スキル。context:fork で子コンテキストを切り離し、subagent:recursive-grandchild-inspector で Skillツール持ちの孫を起動する。子エージェント（deep-experiment-coordinator）から呼び出すことを想定。

孫エージェント検証用スキル。context:fork で親コンテキストを切り離し、subagent:grandchild-inspector で孫を起動する。子エージェント（experiment-coordinator）から呼び出すことを想定。

Upload local images to a GitHub PR and embed them in the description or comments. Use when asked to "attach screenshots to PR", "add images to PR", "upload test results to PR", "embed screenshots in PR description", "add before/after images to PR", "attach UI screenshots", "show test results in PR", "add visual evidence to PR", or any request involving images and PRs. Always use this skill when the user wants to visually document changes in a pull request, even if they don't use the word "upload" — phrases like "put the screenshot in the PR" or "show the image in the PR" should trigger this skill. Supports Playwright MCP / Chrome DevTools MCP / agent-browser as browser automation backends.

Extract tryable feature updates from changelog or update summary files and create structured GitHub issues for feature trial tracking. Use when the user mentions "changelog issue化", "feature trial issues", "アプデをissue化", "機能試行issue", "新機能をissueに", "changelog to issues", "試せる機能をissue化", or wants to create GitHub issues from update/changelog content for testing new features.