codex-plugin

This skill should be used when the user asks to "lint AI assistant config", "check AGENTS.md", "validate SKILL.md", "agnix lint", or wants to detect and fix best-practice violations in Claude Code / agent configuration files.

This skill is the mandatory reference for all Claude Code hook creation. Use cc-hooks-ts for every hook. This skill should be used when the user asks to "create a hook", "add a hook", "write a hook", "implement a hook", "rewrite hooks in TypeScript", "use cc-hooks-ts", or needs to build any Claude Code hook — cc-hooks-ts is always required regardless of whether explicitly mentioned.

Auto-activate CCS CLI delegation for deterministic tasks. Parses user input, auto-selects optimal profile (glm/kimi/custom) from ~/.ccs/config.json, enhances prompts with context, executes via `ccs {profile} -p "task"` or `ccs {profile}:continue`, and reports results. Triggers on "use ccs [task]" patterns, typo/test/refactor keywords. Excludes complex architecture, security-critical code, performance optimization, breaking changes.

This skill should be used when the user asks to "handoff from CCS", "Claude Codeの作業を引き継ぐ", "rate limit handoff", "context overflow", or wants to resume interrupted work from another CCS profile session.

This skill should be used when the user asks for "code review", "レビューして", "review this diff", "実装をレビュー", or wants rigorous review of recently created artifacts.

Direct utility reference for code-review PR scripts. Documents unresolved-thread, comment, CI-status, and CI-log helpers used by higher-level PR workflows.

This skill should be used when the user asks to "organize commits", "split this diff into commits", "コミットを整理", "レビューしやすいコミット", or wants current changes grouped into logical commits.

Internal documentation-index generation workflow for direct invocation from context maintenance tasks.

This skill should be used when the user asks to "refactor context file", "split CLAUDE.md into skills/rules/hooks", "AGENTS.mdを分解", "コンテキストファイルを整理", or wants a single context file decomposed into Claude Code features.

Internal project-context placement and consistency reference used by context refactoring and verification workflows.

This skill should be used when the user asks to "update this markdown", "review every section", "Markdownをセクションごとに更新", "ドキュメントを並列確認", or wants a markdown file audited and updated section by section.

Use when someone says "verify context file", "check if AGENTS.md is correct", "validate context accuracy", "audit CLAUDE.md", "check documentation truth", "are the docs accurate", or mentions "context verification" or "documentation audit". Treats all existing content as potentially incorrect, extracts every technical claim, and verifies each against the actual codebase and web sources.

This skill should be used when the user asks about 'debug plugin behavior', 'what does the debug plugin save', 'debug plugin files', 'where does debug plugin write', or mentions debug plugin data storage and session artifacts.

This skill should be used when the user asks to "set up a project", "initialize a new app", "create a new project", "bootstrap a project", "set up my dev environment", or wants to start a new personal web application. Also use when the user says "init project", "new project", or wants to complement an existing project with missing tooling and configuration.

This skill should be used when the user asks to "set up a project", "choose a tech stack", "establish coding standards", "configure linting", "set up tests", "pick a framework", "review code quality", "set up logging", "configure observability", "choose an ORM", "set up auth", "set up Storybook", "write stories", "visual testing", "CSF", "coverage threshold", or discusses any technology choices for a TypeScript project. Also use when starting a new project, setting up CI/CD, choosing between tools (oxlint vs ESLint, pnpm vs npm), configuring environment variables, or when any other devkit skill needs to reference project conventions. This is the central reference for all devkit technology decisions and coding principles.

Markdown + Mermaid でアーキテクチャ図・システム構成図・フローチャート・依存関係図・シーケンス図・ER図などを作成し、`render` スキル経由で HTML 化する。Use when the user asks to "アーキテクチャ図を書いて", "システム構成図", "フローチャート作成", "依存関係を図に", "draw an architecture diagram", "make a flowchart", "diagram", or wants any visual diagram delivered as a viewable HTML.

Markdown + Mermaid を beautiful-mermaid で SVG にレンダリングし、自己完結 HTML として出力する。Use when the user asks to render a Markdown file with mermaid diagrams to HTML, "markdown を HTML に変換", "mermaid を SVG で埋め込んだ HTML を作って", or wants to produce a shareable HTML report with embedded diagrams.

This skill should be used when the user asks to operate GitHub Projects through the plugin CLI, "GitHub Projects CLI", "Project CRUD", "sub-issue を操作", "blocked-by を設定", or needs the gh-project.ts execution reference for Project, sub-issue, and dependency operations.

新しい EPIC issue を起票し、GitHub Project に登録するワークフロー。Use when the user asks to "create an EPIC", "start a new EPIC", "EPIC を作成", "新しい EPIC を起票", "プロジェクトに EPIC を立てたい", or wants to bootstrap a top-level initiative.

EPIC を「全体リリース Story」と「機能別 検証 Story」に分解し、sub-issue でリンク・blocked-by を張って Project に追加するワークフロー。Use when the user asks to "break down an EPIC", "decompose into stories", "EPIC を分解", "Story に切り分けたい", or has an EPIC that needs decomposition.

EPIC - Story - Task の3階層と依存関係で GitHub Projects を運用するための方法論。Use when the user asks to "set up EPIC story task workflow", "design issue hierarchy", "configure project for EPIC story task", "EPIC Story Task で運用したい", "issue 階層を設計", "GitHub Project を立ち上げ", or wants conceptual guidance on this issue-management pattern.

Story / Task 間の blocked-by 依存関係を設計・登録するワークフロー。Use when the user asks to "set blocked-by", "add dependency", "issue の前後関係を設定", "依存関係を整理", "blocking 関係", or wants to model ordering constraints between issues.

EPIC 単位で sub-issue / blocked-by / Project status を棚卸しし、進捗・滞留・次に着手すべき Task をレポートするワークフロー。Use when the user asks to "review progress", "EPIC の進捗を確認", "棚卸し", "what's next", "次に何をやる", "blockers を確認", or wants a status snapshot of an EPIC.

End-to-end triage — investigate the repo, dedupe against the local issue DB, and for each finding either comment on an existing issue or create a new one. Use when the user says "課題を洗い出してissue化", "リポジトリを調査してissueにする", "report repo issues", "investigate and file issues", "問題を見つけてGitHubに上げる", or wants an end-to-end flow from code to GitHub.

Search GitHub issues from a locally-cached SQLite DB (FTS5 trigram + BM25 + label/recency boost, no ML model) for duplicates or near-duplicates of a proposed issue. Use when the user says "似たissueを探す", "duplicate issue check", "search similar issues", "類似issueを調べる", "既存issueを検索", or before creating a new issue to avoid duplicates.

Verify selected GitHub issues against the current repository state using the local issue DB, post status comments, and — with explicit approval — close resolved ones. Use when the user says "既存issueを更新", "issueの状態を確認", "update existing issues", "issue triage", "古いissueをcloseする", or wants to refresh stale issues with the latest repo reality.

Create a GitHub issue in the current repository with a drafted title, body, and labels, confirming with the user before submission. Use when the user says "issueを作成", "GitHub issueを立てる", "create issue", "新しいissue", "バグ報告issue", or wants to file a new issue from conversation or investigation context.

This skill should be used when the user asks to 'watch PRs', 'monitor PR status', 'PRを監視', 'PRの状態を見て', 'babysit this PR', 'PRが通るまで見て', or when the agent needs to monitor PR health after push, PR creation, or CI failure detection.

プロジェクト規約（AGENTS.md / CLAUDE.md / .claude/rules）を『条件付き規範記法』で書くための記法ガイド — RFC 2119 の要求度キーワード（MUST / MUST NOT / SHOULD / SHOULD NOT / MAY）と `IF: <条件>; THEN <キーワード>: <行動>` 形式のプロダクションルールで、曖昧な散文をエージェントが一意に解釈できる規範に変換する。Use when the user says 'ルールを書く', '規約を書く', 'AGENTS.md にルールを追加', 'IF/THEN で書いて', '規範記法で書いて', 'RFC 2119 で書いて', 'write a rule', 'add a rule to AGENTS.md', or wants to turn prose into machine-checkable normative rules。規約を執筆・再構成する場面では、ユーザーが記法名を明示しなくても使うこと。

This skill should be used when the user asks for 'N回連続レビュー', '3回連続成功', 'サブエージェントで連続レビュー', 'streak review', 'consecutive review', 'review until N passes', 'レビューが連続成功するまで', or wants iterative subagent review where past feedback MUST NOT leak into the next review. Runs a subagent review-and-fix loop until N consecutive passes are achieved, resetting the streak on any failure and invoking each subagent review with a flat, history-free prompt.

This skill should be used when the user asks to "clarify the plan", "ask detailed questions", "曖昧な点を詰めて", "計画を深掘り", or wants structured questions that resolve unclear points in a plan file.

This skill should be used when the user asks to 'port a feature from repo A to repo B', 'migrate a repository feature faithfully', 'rewrite faithfully', 'fork-and-customize without dropping flows', 'リポジトリ間移植', '別リポジトリの機能を移植', '忠実に移植', 'A の機能を B にそのまま持ってきて', or wants to port / rewrite / re-implement an existing repository's feature set into another codebase without silently shrinking scope. It forces 1:1 enumeration, source/target architecture diagrams reviewed by the consecutive-review-loop skill, a shrink-vocabulary ban, one-outcome-per-PR, and a pre-merge local verification with screenshots.

Send feedback about this plugin repository as a GitHub issue. Use when the user says "feedback", "send feedback", "report issue", "bug report", "feature request", "フィードバック", "バグ報告", "機能リクエスト", or wants to create an issue for the plugin repository.

理想を絶対に忘れずに、手段にとらわれずに設計案が理想を満たしているかを徹底的に議論してから設計を確定するためのスキル。Use when the user says '理想から設計', '理想駆動', '理想を満たす設計', '手段にとらわれず設計', '本質から設計', 'ideal driven design', 'north star design', or wants to design something starting from the ideal state without being constrained by means.

Investigate the current repository to surface issues, defects, and improvement opportunities. Use when the user says "調査して課題", "リポジトリを調査", "investigate repo", "find issues in repo", "問題を洗い出す", "課題を見つける", or wants a structured report of problems detected in the codebase.

This skill should be used when the user asks to "make a plan from this issue", "issueから実装計画", "GitHub Issueを計画化", "plan issue #123", or provides a GitHub Issue URL/number and wants an implementation plan.

Review whether the current code satisfies a GitHub issue, then update the issue status to match. Use when the user asks to 'review an issue', 'check if an issue is done', 'is this issue resolved', 'update issue status', 'issueの対応状況を確認', 'issueが完了しているかレビュー', or 'issueのステータスを更新'.

This skill should be used when the user asks for "multiple perspectives", "一般的・保守的・独創的", "複数観点で提案", "3つの観点", or wants standard, conservative, and creative proposals for a theme.

This skill should be used when the user asks to "reflect on this session", "振り返って", "何が悪かったか考えて", "self-review the work", or wants an honest process retrospective with lessons.

This skill should be used when the user asks to "set up mutils", "mutilsをセットアップ", "install mutils settings", "mutils setup", or wants recommended mutils references and dependencies applied.