CAPHTECH

デザインシステムの構築・運用を体系的に行うスキル。ドキュメント（要件定義・ブランドガイドライン）からDesign Tokens・コンポーネント体系・ガバナンスまでを一貫して設計する。思考フレームワーク（Double Diamond・Atomic Design・Design Tokens 3層）の適用、暗黙的デザイン判断の形式知化（DDR/QOC/RFC）、UIインベントリ収集（手動+MCP自動化）を統合する。ゼロから体系的に構築する用途向け。既存の単発UIから再利用可能な構造を抽出するにはrelational-design-pluginのdesign-system-backflowスキルを使用すること。Use when: 「デザインシステムを構築して」「デザインシステムを設計して」「デザイントークンを定義して」「デザイン原則を策定して」「UIインベントリを作って」「デザインガバナンスを設計して」「デザイン判断を記録して」と言われた時。

iOS/Androidモバイルアプリのデザインを体系的に行うスキル。Apple HIG・Material Design 3などプラットフォームガイドラインに準拠し、ナビゲーション・レイアウト・コンポーネント・モーション・アクセシビリティを適切に設計する。デザイン判断の観察・関係・仮説・撤回可能性をtraceとして残したい場合はrelational-design-pluginを使用すること。Use when: 「モバイルアプリをデザインして」「iOSアプリのUIを設計して」「Androidアプリの画面を作って」「モバイルのナビゲーションを設計して」「タッチUIを改善して」「アプリのアクセシビリティを対応して」と言われた時。

Webアプリの個別画面・コンポーネントのデザインを体系的に行うスキル。デザインプロセス・レイアウト・コンポーネント設計・インタラクション・アクセシビリティなどWebデザインの確立された手法を適用する。デザインシステム全体の構築・運用はdesign-system-builderスキルを使用すること。デザイン判断の観察・関係・仮説・撤回可能性をtraceとして残したい場合はrelational-design-pluginを使用すること。Use when: 「Webアプリをデザインして」「UIを設計して」「画面をデザインして」「レスポンシブ対応して」「アクセシビリティを改善して」「コンポーネントを設計して」「カラーパレットを決めて」と言われた時。

Use this skill when a design assumption, user segment, business goal, constraint, or product requirement has changed and you need to analyze which design hypotheses, decisions, artifacts, copy, or components should be retracted or revised. Requires a Relational Design trace session with recorded observations/relations/hypotheses/decisions to analyze against — not a general "redo this design" request.

Use this skill to critique an existing UI, frontend implementation, wireframe, flow, or visual design through Relational Design relations: user state, business intent, action risk, trust, information density, reversibility, accessibility, and implementation constraints. This is a relation-based critique, not a general design-methodology review — for platform-guideline or design-system compliance review, use design-plugin instead.

Use this skill after creating or reviewing a design artifact to extract reusable design-system knowledge: semantic tokens, components, variants, interaction rules, copy patterns, accessibility constraints, and governance notes. This only backflows structure out of an artifact that already exists — to build or govern a design system from scratch (Design Tokens layers, Atomic Design, DDR/QOC/RFC, UI inventory), use design-plugin's design-system-builder instead.

Use this skill for UI/UX, frontend visual design, product flow, landing page, dashboard, component, or design-system-aware work where design reasoning should be traceable, isolable, and retractable. It orchestrates context reading, relation mapping, hypothesis generation, artifact design, critique, and trace archival. This is about the reasoning process, not visual output methodology — for HIG/Material Design/Atomic Design-driven UI or design-system construction without a trace/retraction workflow, use design-plugin's web-app-designer, mobile-app-designer, or design-system-builder instead.

Codex CLI (OpenAI gpt-5.4) に直接相談するスキル。設計判断・実装方針・コードレビュー・デバッグ方針など、別モデルの視点が欲しい時に使用する。「codexに相談」「codexに聞いて」「別の視点が欲しい」「セカンドオピニオン」「codexでレビュー」「codexとペアプロ」「codexとTDD」「codexでペアプログラミング」「ペアプロモード」「adversarial review」と言った時にトリガーする。

Prepare, validate, and ship higher-graphen releases from release-readiness triage through final publication. Use when asked to prepare a release, run release quality gates, fix release blockers, create release tags, draft release notes, publish a GitHub release, or verify post-release health for this repository.

Use when an agent needs the current HigherGraphen Architecture Product direct database access smoke review workflow.

Use when an agent must turn code diffs, tests, schemas, or HigherGraphen workflow surfaces into high-order HG structure before judging test gaps, reviews, or completion candidates.

Use when an agent needs to create, inspect, reason over, review, patch, or validate CaseGraphen cases and workflow graphs through the installed `cg` workspace CLI and the repository-owned `casegraphen` CLI.

Use when an agent needs to run or interpret the bounded `highergraphen ddd` review workflow for domain model, bounded context, aggregate, anti-corruption mapping, evidence-boundary, projection-loss, or closeability diagnostics.

Use when an agent needs to run or interpret repository-owned HigherGraphen CLI workflow reports, including Architecture Product smoke, Feed reader, completion review, PR review target recommendation, and test-gap detector contracts.

Core pce-memory v2 workflow skill. Guides raw observe, distill/promote, intent-aware activate, feedback, rollback, and the narrow upsert escape hatch. Triggered by: 'memory workflow', 'activate context', 'record to memory', 'promotion pipeline'.

Manage CAPHTECH/pce-memory GitHub issue, milestone, and Project

Dedicated pce-memory v2 promotion pipeline skill. Guides distill -> promote -> rollback with candidate lineage, reviewer metadata, and durable repair rules. Triggered by: 'promotion pipeline', 'distill memory', 'promote candidate', 'rollback claim'.

Team knowledge synchronization skill for pce-memory v2. Manages push/pull/status with memory_type-aware export/import and layer-aware sync rules. Triggered by: 'sync knowledge', 'push knowledge', 'pull team knowledge', 'check sync status'.

Knowledge graph operations skill for pce-memory. Create, query, and visualize entities and relations. Triggered by: 'create entity', 'graph relation', 'query knowledge graph', 'map dependencies'.

Use when the user wants multi-step work managed through CaseGraph instead of ad hoc chat state. Trigger on phrases like "manage this in cg", "drive this from the case", "record evidence for compaction", "resume from cg", "verify before close", or whenever durable checkpoints, verification, and guarded closure are needed for implementation, docs, investigation, or review work.

Use when the user wants to review AI-performed work on a CaseGraph case. Trigger on phrases like "case <id> をレビューして", "AI がやった作業を確認して", "review the case", "audit this case", "証跡をチェックして", or whenever a structured read-only inspection of task completion, evidence integrity, decision traceability, and event-log trustworthiness is needed.

Use when integrating CaseGraph with external tools (markdown importer, optional sinks like Todoist / GitHub Issues, workers such as shell / local-llm / code-agent) or when verifying or recovering storage. Trigger on references to `cg import markdown`, `cg sync push|pull`, `cg worker run`, `cg validate`, `cg validate storage`, `cg cache rebuild`, `cg events verify|export`, `cg migrate`, `env_allowlist`, `approval_policy`, or phrases like "import markdown", "sync pull", "run a worker", "rebuild the cache", "verify events", "run migration". For day-to-day graph reading or editing use casegraph; for AI-authored graph changes use casegraph-patch.

Use when proposing AI-driven graph changes to a CaseGraph workspace as a `GraphPatch`. Trigger on references to `cg patch validate|review|apply`, `GraphPatch`, `base_revision`, `patch.applied`, generator metadata, the `casegraph-patch` fenced-block convention, or phrases like "propose as a patch", "wrap this change into a patch", "base_revision is stale", "apply the patch". Encodes ADR-0003 "AI does not own state". For direct manual edits use casegraph; for importer/sink/worker/recovery use casegraph-integrate.

Use when reading, manually authoring, or analyzing a CaseGraph workspace through the `cg` CLI. Trigger on references to `.casegraph/`, `cg case`, `cg node`, `cg edge`, `cg task`, `cg frontier`, `cg blockers`, `cg analyze`, `cg case view`, or phrases like "show the case", "what's blocked", "analyze impact", "create a case", "show the frontier". Do NOT use for AI-proposed graph changes (use casegraph-patch) or for importer/sink/worker/storage recovery (use casegraph-integrate).

Guide stable FlutterHelm setup and execution through the default npx-backed plugin workflow. Use when Codex needs to start or operate stable FlutterHelm flows such as workspace selection, run and inspect, tests, profiling, compatibility checks, or native handoff without enabling beta runtime_interaction or native_build paths.

Enable FlutterHelm runtime_interaction beta with the built-in external runtime driver and optional mobile-mcp companion MCP. Use when Codex needs semantic tap, text entry, scroll, richer screenshots, or UI-state verification on a running Flutter session.

Configure and use FlutterHelm native_build beta with a stdio_json nativeBuild provider such as an XcodeBuildMCP-like backend. Use when Codex needs iOS-first native build or launch plus same-session Flutter runtime attachment and correlated native evidence.

Diagnose FlutterHelm failures and degraded states. Use when Codex needs to triage approval_required responses, SESSION_BUSY or WORKSPACE_BUSY, root or fallback problems, degraded delegate/provider health, runtime driver errors, native_build setup problems, or HTTP preview confusion.

Use when an agent needs to run AdvisoryGraphen for evidence-backed technical advisory, architecture review, AI-governed completion review, or case reasoning workflows.

Use when reviewing a pull request, local diff, or agent-made change with AdvisoryGraphen to propose where a human reviewer should focus. This skill builds an evidence-backed advisory snapshot from code/docs/test changes, models review focus as verification requirements, runs AdvisoryGraphen validation/check/projection, and translates obstructions and HigherGraphen correspondence signals into review priorities.

`context-probe` を既存 repo に適用するための YAML 入力整備ワークフロー。`context-probe input を作って` `repo に context-probe を適用して` `scaffold から curated YAML にして` と言われた時に使う。

Run `context-probe` measurements and analyze the results after repo inputs exist. Executes `score.compute`, `report.generate`, `gate.evaluate`, `review.list_unknowns`, and deeper `doc.extract_*` `trace.*` `history.*` or `review.resolve` only when needed. Use when asked to measure a repo, analyze a result, inspect unknowns, or continue after YAML inputs are ready.

Execute state_gate release process including version bumping (major/minor/patch), CHANGELOG updates, git tagging, GitHub release creation, and npm publication. Use when the user asks to "release", "publish a new version", "create a release", "bump version", or mentions releasing state_gate to npm.

Document ID management for Git repositories. Validates, assigns, and tracks immutable doc_ids across Markdown/YAML files. Use when working with doc_id, document IDs, "@see docid" references, document integrity validation, or shirushi commands.

Use engi to diagnose whether Rust, Python, or TypeScript code is structurally recognizable to AI agents and humans, rank hard-to-change units, build evidence-carrying context packs, and produce measurable refactoring plans. Use when asked to assess AI readability, find code an agent will struggle to modify, prepare context for a code change, compare readability before and after refactoring, or evaluate engi itself on a package, project, or workspace.

Write tests for the TemporalKit LTL library WITHOUT reading the source first. Hands over the library-specific facts the bare model gets wrong from memory: the exact normalized()/NNF rewrite forms, PropositionID/proposition construction, the per-directory test framework, and the model-checking setup. Use when adding or editing tests under Tests/TemporalKitTests for LTLFormula, normalized(), LTLModelChecker / KripkeStructure, trace evaluation, or the DSL.