コード変更に影響を受ける md ドキュメントを検出し、ユーザー承認後に更新する。 md ファイルの YAML frontmatter に宣言された depends-on（ファイルパス・glob）と 本文中の Markdown リンクから依存グラフを構築し、git diff との突合で影響範囲を特定する。 /doc-check で起動、またはコード変更を含むタスク完了時に実行を検討する。

md ドキュメントの陳腐化・欠落・矛盾を 4 Layer 構造で検出し、ユーザー承認後に修正する。 depends-on 検証、coverage チェック、business-rule/architecture の未文書化知識検出、 readme/CLAUDE.md のメタ整合検査を行う。/doc-audit で起動、または大規模なコード変更後に ドキュメント整合性を確認したい時に使用する。

前セッションの project-state.json から未完了タスク（in_progress/blocked）を特定し、 ユーザー承認後に作業を再開する。/continue の明示呼び出し、"continue from handover" のような継続指示、handover.md への言及を検出した時に使用する。worktree 切り替えと Pipeline Detection も含む。

スキルを作成・更新・プロンプト改善するためのメタスキル。 **collaborative**モードでユーザーと対話しながら共創し、 抽象的なアイデアから具体的な実装まで柔軟に対応する。 **orchestrate**モードでタスクの実行エンジン（Claude Code / Codex / 連携）を選択。 Anchors: • Continuous Delivery (Jez Humble) / 適用: 自動化パイプライン / 目的: 決定論的実行 • The Lean Startup (Eric Ries) / 適用: Build-Measure-Learn / 目的: 反復改善 • Domain-Driven Design (Eric Evans) / 適用: 戦略的設計・ユビキタス言語・Bounded Context / 目的: ドメイン構造の明確化 • Clean Architecture (Robert C. Martin) / 適用: 依存関係ルール・層分離設計 / 目的: 変更に強い高精度スキル • Design Thinking (IDEO) / 適用: ユーザー中心設計 / 目的: 共感と共創 Trigger: 新規スキルの作成、既存スキルの更新、プロンプト改善を行う場合に使用。 スキル作成, スキル更新, プロンプト改善, skill creation, skill update, improve prompt, Codexに任せて, assign codex, Codexで実行, GPTに依頼, 実行モード選択, どのAIを使う, IPC Bridge統一, API統一パターン, safeInvoke/safeOn, Preload API標準化, IPC handler registration, Preload API integration, contextBridge, Electron IPC pattern

Google Forms APIを使用してフォームを自動生成するスキル。 ユーザーとの対話形式でヒアリングを行い、Markdown形式で構成確認後、 APIでフォームを作成する。全11種類の質問タイプ、クイズ機能、 条件分岐、Drive/Sheets連携に対応。

Production-tested patterns for maximizing Claude Code effectiveness. Covers CLAUDE.md setup, subagent design, hooks, permissions, parallel sessions, and MCP integration. Not for general coding or debugging.

Analyze Codex/agent review comments on one or more PRs, triage what should be implemented, and provide a clear implementation recommendation list for the user.

Verify implemented Factory plans against original source requirements, run all required validation lanes, create a repair branch when gaps exist, and fix gaps until all requirements are satisfied.

Implement any valid Factory execution plan story-by-story using the active repo profile, planned changelog/versioning fields, and required validation lanes.

Convert incident artifacts into deterministic regression fixtures and CI-ready outputs. Use when users ask to reproduce a failure, build repeatable graders, or emit junit evidence.

Evaluate Gait policy changes safely with deterministic fixture tests and rollout simulation. Use when asked to validate allow or block behavior, check reason codes, compare policy outcomes, or plan enforce-by-stage rollout.

Convert a Gait run artifact into a deterministic regression workflow. Use when asked to initialize fixtures from run_id or runpack path, run graders, produce CI-friendly outputs, or summarize drift and failures.

Capture and verify deterministic Gait runpacks from normalized run input. Use when asked to record a run, produce run_id or runpack artifacts, generate ticket-ready proof, or validate artifact integrity before handoff.

Update the local Factory submodule pointer to latest factory/main through a small PR before running Factory-backed skills.

Generate portable evidence receipts from run artifacts. Use when users ask for ticket-ready proof, receipt footers, or integrity-backed handoff metadata.

Resolve changelog-derived semver, finalize the release changelog through Factory scripts, tag from the default branch, monitor release/UAT, and run bounded hotfix loops when failures are actionable.

Commit scoped changes, push a branch, open or update a PR, and keep driving profile-defined CI, passive review, merge, and post-merge loops until success or a hard blocker remains.

Perform a profile-driven full-repository or scoped code review with severity-ranked findings focused on contracts, boundaries, fail-closed safety, determinism, portability, release integrity, and docs correctness.

Triage CI failures using artifact-first checks. Use when users need fast root-cause isolation from failing runs, integrity verification, and deterministic reruns.

Run a profile-driven evidence-based OSS/product readiness audit across product model, onboarding, architecture, DX, security, release readiness, and market wedge.

Convert user-provided recommendations into a timestamped, execution-ready backlog plan using the active repo profile. Use when a repo needs a one-off plan that should not overwrite PLAN_NEXT.

How to write SKILL.md files that trigger reliably and teach effectively. Use when creating, improving, or reviewing skills for any tool — SKILL.md is the cross-tool open spec (agentskills.io), read identically by Claude Code, Codex CLI, and Antigravity.

How to write .claude/rules/ files that Claude actually follows. Use when creating, improving, or reviewing project rules.

How to write effective system prompts for any LLM. Universal prompt engineering -- role clarity, structured output, injection resistance, few-shot examples. Use when writing prompts, system instructions, or AI configuration.