VictorAurelius

Dùng khi user nói 'start new session', 'bắt đầu session mới', '/start-session', 'what's the state?', 'tình trạng hiện tại', hoặc khi cần load context sau khi /clear. Load: CLAUDE.md digest, current wave, open PRs, failing CI, active blockers. Output: single summary block. Cũng check session-lock conflicts khi chạy parallel agents.

Dùng khi user nói 'plan wave', 'cluster gaps', 'wave-pack', 'next wave', hoặc khi cần group ≥3 disjoint gaps thành 1 wave-pack chạy parallel agents. Output: wave plan markdown (foundation PR-ready) + file-overlap matrix + 3-5 agent prompts. Codify pattern từ Wave Observability 2026-04-28 (3 gaps trong ~75 min vs ~6h serial = 5x speedup).

Use when PR touches kitehub-email templates (*.html under templates/emails/), kiteclass-gateway email templates, any new customer-facing email, phrases 'review email', 'kiểm tra email template', 'gửi email cho tenant', 'trial-expiry email', 'welcome email'. Enforces brand/legal/i18n/mobile/GDPR-PDPL gate before customer send per output-review-mandate §5.7.

Dùng khi 'kit parity', 'kit production parity', 'review Track 2 port', 'port theo ui kit', 'kit vs production', 'back-port kit', hoặc trước khi merge PR port kit → production / production → kit. Verifies 4-layer V-model parity (要件定義 / 基本設計 / 詳細設計 / コンポーネント設計) giữa HTML kit prototype và production Next.js code. Bidirectional: kit→production port AND production→kit back-port (production led). Output report /4-layer to documents/04-quality/audits/parity/.

Dùng khi PR thay đổi AI Branding behavior — model swap (Ollama → Bedrock, Llama → Gemma), prompt template change, AIClient/Step interface, ContentModerationService, InstanceQualityReviewer logic, 5 *QualityCheck classes. Auto-trigger qua audit-gate khi files match `kiteclass/kiteclass-core/src/main/java/com/kiteclass/core/module/{ai,branding,instance,quality,moderation,provisioning}/**/*.java`. Manual checklist mode (5 sections × scoring) + delta-vs-baseline rubric. Output: `documents/04-quality/audits/ai-branding/YYYY-MM-DD-<change>.md`.

Dùng khi user nói 'security audit', 'pentest', 'kiểm tra bảo mật', 'dependency scan', hoặc trước production deploy. Deep security check /100 — vượt xa quality-audit §2. v2 audit format mandatory per GAP-564: mỗi control bind với evidence block (Command run + Output + Verdict + Evidence artifact ID).

Dùng khi user nói 'audit prototype', 'review ui kits', 'check landing parity', 'kiểm tra prototype HTML', hoặc trước khi merge wave UI kits. Extends quality/ui-review for static HTML prototype path (vs Next.js dev server). Runs 3 scripts: link-checker (no broken hrefs in kit HTML) + landing-parity (stricter than Tier 1: cards + scores + persona pills + screen counts) + state-coverage (default/loading/empty/error/success/dark per dossier §4). Codifies 2026-04-29 GAP-263 Phase 2 + closes GAP-264.

Dùng khi user nói 'end session', 'đóng session', 'kết thúc session', '/end-session', 'finish work', 'wrap up', hoặc trước khi /clear cho session quan trọng. Working-tree clean + sync gate (Step 0a: no dirty/uncommitted, main synced OR in-flight branch pushed+handed-off) + docs-sync 5-target (CSV/ROADMAP/wave-history/MEMORY/session-handoff per session-end-context-check.md §4.5) + auto-write session-handoff note + archive session-lock + 1-line summary. Symmetric counterpart cho /start-session.

Use khi user nói 'làm khóa luận', 'thesis V1', 'audit thesis', 'thesis ready to defense?', 'render thesis docx', hoặc trước khi ship academic deliverable. Workflow: planning → draft chapters → 3-agent outside-in audit → fix → assemble DOCX. Backed by thesis-content-standard.md rubric /100.

Dùng khi user nói 'extract citations', 'trích dẫn luận văn', 'audit thesis cite', 'verify bibliography', 'orphan citation check', 'kiểm tra cite luận văn', hoặc khi review thesis chapter trước defense. Parse `[N]` cite keys từ thesis chapters + verify chéo với `documents/<thesis-dir>/references/bibliography.md` → báo 3 bucket: matched / orphan-body (cite không có entry) / orphan-bib (entry không cite).

Dùng khi user nói 'figure curation', 'thesis figures', 'caption', 'đánh số hình', 'INDEX figure', 'curate figure cho Chapter N', 'add figure to thesis', 'kiểm tra caption hình', 'audit figure thesis'. Codifies selection criteria (figure vs table vs prose), per-chapter numbering N.M, Vietnamese caption format, per-chapter INDEX generation. Pair với thesis-docx pipeline.

Use when user says 'review UI', 'check design', 'audit screenshots', or after any UI/UX code change. Auto-runs after every frontend PR. Captures before/after screenshots, scores per-screen on /128 scale.

Dùng khi user nói 'extract citations', 'trích dẫn luận văn', 'audit thesis cite', 'verify bibliography', 'orphan citation check', 'kiểm tra cite luận văn', hoặc khi review thesis chapter trước defense. Parse `[N]` cite keys từ thesis chapters + verify chéo với `documents/<thesis-dir>/references/bibliography.md` → báo 3 bucket: matched / orphan-body (cite không có entry) / orphan-bib (entry không cite).

Dùng khi user nói 'figure curation', 'thesis figures', 'caption', 'đánh số hình', 'INDEX figure', 'curate figure cho Chapter N', 'add figure to thesis', 'kiểm tra caption hình', 'audit figure thesis'. Codifies selection criteria (figure vs table vs prose), per-chapter numbering N.M, Vietnamese caption format, per-chapter INDEX generation. Pair với thesis-docx pipeline.

"Use when user says '{trigger phrase 1}', '{trigger phrase 2}', or {situation when skill applies}. {What the skill does in one sentence}."

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

Dùng khi user nói 'api audit', 'contract check', 'kiểm tra API', 'endpoint match docs?', 'breaking change?', hoặc trước release. Verify API endpoints match api-contract.md /100.

Dùng khi user nói 'audit', 'quality check', 'kiểm tra chất lượng', 'điểm chất lượng', 'ready to merge?', 'persona coverage', hoặc trước khi merge một wave/feature lớn vào main. Chấm điểm 11 categories /110 điểm (10 tech + 1 persona coverage).

Dùng khi user nói 'business audit', 'logic check', 'kiểm tra business logic', 'code đúng rules chưa', hoặc trước GA release. Verify code implement đúng rules.md + use-cases.md. 5 categories /100.

Dùng khi user nói 'design pattern audit', 'pattern check', 'kiểm tra design patterns', 'God service đâu', 'anti-pattern hotspot', hoặc trước một refactor planning cycle. Score code against `.claude/rules/design-patterns.md` §3 BANNED list. Output: hotspot list (file path + LOC + violated pattern) + score /100.

Dùng khi user nói 'ops audit', 'production ready?', 'kiểm tra ops', 'deploy checklist', 'monitoring check', hoặc trước GA release. Production operations readiness /100.

Dùng khi user nói 'perf audit', 'performance check', 'load test', 'kiểm tra hiệu năng', 'N+1', 'bundle size', hoặc trước production deploy. Baseline performance metrics /100.

Dùng khi user nói 'security audit', 'pentest', 'kiểm tra bảo mật', 'dependency scan', hoặc trước production deploy. Deep security check /100. v2 audit format: mỗi control bind với evidence block (Command run + Output + Verdict + Evidence artifact ID).

Dùng khi user nói 'plan wave', 'cluster gaps', 'wave-pack', 'next wave', hoặc khi cần group ≥3 disjoint tasks/gaps thành 1 wave-pack chạy parallel agents. Output: wave plan markdown (foundation PR-ready) + file-overlap matrix + 3-5 agent prompts. Codifies cluster-then-spawn parallel-agent methodology (group disjoint work → spawn worktree-isolated agents → ~5x speedup vs serial PR queue).

Run the migrated source command flashcard. Use when the user asks for /flashcard or wants Anki CSV cards for certification topics, vocabulary, or wrong answers.

Run the migrated source command giai-thich. Use when the user asks for /giai-thich or wants a Vietnamese explanation of Java, Oracle SQL, Spring, or AWS certification concepts.

Run the migrated source command kiem-tra. Use when the user asks for /kiem-tra, mock exams, practice questions, or interactive certification testing for Java, SQL, Spring, or AWS.

Run the migrated source command lich-su-sai. Use when the user asks for /lich-su-sai, wants to log wrong answers, or wants analysis of weak certification topics.

Run the migrated source command on-tap. Use when the user asks for /on-tap or wants a focused review session for Java, SQL, Spring, or AWS certification topics.

Review tiến độ ôn thi + gợi ý bước học kế tiếp

Run the migrated source command tong-hop. Use when the user asks for /tong-hop or wants English technical material summarized into Vietnamese for the certification roadmap.

Run the migrated source command tu-vung. Use when the user asks for /tu-vung or wants to learn technical English vocabulary for Oracle, Java, Spring, or AWS certifications.