@musashishao/agent-kit 1.9.0 → 1.9.1

package/.agent/workflows/gamekit-tasks.md

@@ -0,0 +1,208 @@
+---
+description: "Generate executable task list from approved game plan."
+description_vi: "Tạo danh sách task thực thi từ kế hoạch game đã được duyệt."
+description_en: "Generate executable task list from approved game plan."
+---
+
+# /gamekit.tasks - Game Task List Generation
+
+Generate an executable task list with checkboxes, session tracking, and progress overview from an approved implementation plan.
+
+---
+
+## 🎯 Purpose
+
+Transform PLAN into trackable TASKS:
+1. Convert task table to checkbox format
+2. Add verification steps for each task
+3. Include session logging capability
+4. Generate progress overview
+
+---
+
+## Prerequisites
+
+- `PLAN-{game}.md` exists in `docs/plans/`
+- PLAN has been reviewed and approved by user
+
+---
+
+## 📋 Workflow Steps
+
+### Step 0: Read PLAN
+
+```markdown
+🔍 **Reading implementation plan...**
+
+**PLAN file:** docs/plans/PLAN-{game}.md
+**Total phases:** X
+**Total tasks:** Y
+```
+
+### Step 1: Parse Task Table
+
+Extract from PLAN:
+- All task IDs and names
+- Agent assignments
+- Dependencies
+- INPUT → OUTPUT → VERIFY for each
+
+### Step 2: Generate Checkbox Format
+
+Transform each task:
+
+```markdown
+### [Task ID] [Task Name]
+- [ ] **Task:** [Description]
+- **Agent:** `[agent-name]`
+- **Depends on:** [dependencies or "None"]
+- **Input:** [What's needed to start]
+- **Output:** [What's produced when done]
+- **Verify:** [How to confirm completion]
+- **Estimated:** [time estimate]
+```
+
+### Step 3: Add Progress Overview
+
+```markdown
+## 📊 Progress Overview
+
+| Phase | Status | Progress | Estimated |
+|-------|--------|----------|-----------|
+| Phase 1: Core Systems | ⏳ | 0/X | X hours |
+| Phase 2: Gameplay | ⏳ | 0/X | X hours |
+| Phase 3: UI & Polish | ⏳ | 0/X | X hours |
+| Phase 4: Audio | ⏳ | 0/X | X hours |
+| Phase X: Verification | ⏳ | 0/X | X hours |
+| **Total** | | 0/Y | X hours |
+```
+
+### Step 4: Add Session Log Template
+
+```markdown
+## 📝 Session Log
+
+### [DATE] - Session 1
+**Focus:** [What was worked on]
+**Completed:**
+- [x] Task X.X
+- [x] Task X.X
+
+**Blockers:**
+- [None / Description]
+
+**Next Session:**
+- [ ] Task X.X
+- [ ] Task X.X
+
+---
+```
+
+### Step 5: Add Phase X Verification
+
+```markdown
+## Phase X: Verification
+
+### Build Verification
+- [ ] **Compiles without errors**
+  - Command: `[build command]`
+  - Expected: Success
+
+- [ ] **Build size acceptable**
+  - Target: < [from PLAN] MB
+
+### Performance Verification
+- [ ] **60 FPS on target device**
+  - Test device: [from SPEC]
+  - Method: 5-minute play session
+
+### Feature Verification (from SPEC Acceptance Criteria)
+- [ ] AC1: [criterion]
+- [ ] AC2: [criterion]
+- [ ] AC3: [criterion]
+
+### Quality Verification
+- [ ] No critical bugs
+- [ ] All audio plays
+- [ ] Save/Load works
+```
+
+### Step 6: Generate TASKS File
+
+```bash
+# Generate TASKS file
+# Location: docs/plans/TASKS-{game-slug}.md
+```
+
+---
+
+## 📁 Output Files
+
+| File | Location | Content |
+|------|----------|---------|
+| TASKS | `docs/plans/TASKS-{game}.md` | Executable task list |
+
+---
+
+## ⏭️ Next Steps
+
+```markdown
+✅ Task list generated!
+
+**File:** `docs/plans/TASKS-{game}.md`
+**Total tasks:** X tasks
+**Estimated time:** Y hours
+
+**How to use:**
+1. Open TASKS file
+2. Work through tasks in order (respect dependencies)
+3. Check off completed tasks
+4. Log progress in Session Log section
+5. When all done, complete Phase X verification
+
+**Start implementation:**
+→ Use `/game-hybrid` to begin
+→ Or `/orchestrate` for multi-agent parallel work
+```
+
+---
+
+## 🔗 Integration
+
+| Skill | Usage |
+|-------|-------|
+| `spec-ecosystem` | TASKS template |
+| `project-planner` | Task parsing logic |
+
+---
+
+## Task Status Icons
+
+| Icon | Meaning | When to Use |
+|------|---------|-------------|
+| ✅ | Completed | Task verified done |
+| 🔄 | In Progress | Currently working |
+| ⏳ | Waiting | Not started/blocked |
+| ❌ | Blocked | Dependency issue |
+| ⚠️ | Warning | Needs attention |
+
+---
+
+## AI Task Tracking
+
+When AI works on tasks:
+
+```markdown
+📋 **Current Session**
+
+**Working on:** Task 2.3 - Player Movement
+**Status:** 🔄 In Progress
+
+**Progress:**
+- [x] Created PlayerController.cs
+- [x] Added input handling
+- [ ] Testing movement feel
+- [ ] Tuning via ScriptableObject
+
+**Will update TASKS file when complete.**
+```

package/.agent/workflows/marketing.md

@@ -1,5 +1,7 @@
 ---
 description: Growth marketing and campaign planning workflow. Sets up analytics, crafts copy, and plans the launch.
+description_en: Growth marketing and campaign planning workflow. Sets up analytics, crafts copy, and plans the launch.
+description_vi: Workflow lập kế hoạch marketing tăng trưởng và chiến dịch. Thiết lập phân tích, viết nội dung và lên kế hoạch ra mắt.
 ---
 
 # 🚀 /marketing - Campaign Planning Workflow

package/.agent/workflows/next.md

@@ -1,6 +1,7 @@
 ---
 description: Suggest the next steps for the developer.
 description_vi: Gợi ý các bước tiếp theo cho nhà phát triển.
+description_en: Suggest the next steps for the developer.
 ---
 
 # /next - What Should I Do Next?

package/.agent/workflows/orchestrate.md

@@ -1,6 +1,7 @@
 ---
 description: Coordinate multiple agents for complex tasks. Used for multi-perspective analysis, comprehensive reviews, or tasks requiring different expertise.
 description_vi: Điều phối nhiều Agent cho các tác vụ phức tạp. Được sử dụng để phân tích đa chiều, đánh giá toàn diện hoặc các tác vụ yêu cầu chuyên môn khác nhau.
+description_en: Coordinate multiple agents for complex tasks. Used for multi-perspective analysis, comprehensive reviews, or tasks requiring different expertise.
 ---
 
 # Multi-Agent Orchestration
@@ -37,6 +38,17 @@ $ARGUMENTS
 
 ---
 
+## 🔴 VIBES PRE-FLIGHT (MANDATORY)
+
+Before orchestration begins:
+1. **V** - Set success evidence: `ak memory set --key success_criteria --value "X"`
+2. **I** - Capture full intent using `@[skills/intent-capture]`
+3. **B** - Identify bifurcations using `@[skills/bifurcation-analysis]`
+4. **E** - Define emergence boundaries (scope/deps)
+5. **S** - Confirm safety layers are active
+
+> ⚠️ **HALT** if VIBES check is incomplete.
+
 ## Pre-Flight: Mode Check
 
 | Current Mode | Task Type | Action |

package/.agent/workflows/pentest.md

@@ -1,5 +1,7 @@
 ---
 description: Penetration testing and security audit workflow. Systematic vulnerability discovery and exploitation.
+description_en: Penetration testing and security audit workflow. Systematic vulnerability discovery and exploitation.
+description_vi: Workflow kiểm thử xâm nhập và đánh giá bảo mật. Phát hiện lỗ hổng hệ thống và khai thác thử nghiệm.
 ---
 
 # 🕵️ /pentest - Security Audit Workflow

package/.agent/workflows/plan.md

@@ -1,6 +1,7 @@
 ---
 description: Create a project plan using the project-planner agent. Only creates plan files - does not write code.
 description_vi: Tạo kế hoạch dự án bằng Agent project-planner. Chỉ tạo file kế hoạch - không viết code.
+description_en: Create a project plan using the project-planner agent. Only creates plan files - does not write code.
 ---
 
 # /plan - Project Planning Mode
@@ -18,8 +19,32 @@ $ARGUMENTS
 
 ---
 
+## 🔴 VIBES PRE-FLIGHT (MANDATORY)
+
+Before planning:
+1. **V** - Define success: "Plan approved when user confirms all phases"
+2. **I** - Capture full intent using `@[skills/intent-capture]`
+3. **B** - Identify architecture bifurcations using `@[skills/bifurcation-analysis]`
+4. **E** - Set scope boundaries to prevent creep
+5. **S** - Confirm safety layers for Tier 2/3 actions
+
+> ⚠️ **HALT** if VIBES check is incomplete.
+
+---
+
 ## Task
 
+## 💾 Git Checkpoint (Safety First)
+
+Before starting work:
+1. [ ] **Clean working tree?** (Run `git status`)
+2. [ ] **On correct branch?** (Not `main` or `master`)
+3. [ ] **Latest pulled?** (Run `git pull`)
+
+> 🔴 **STOP** if git state is dirty. Ask user to commit or stash first.
+
+---
+
 Use the `project-planner` agent with this context:
 
 ```
@@ -88,3 +113,20 @@ Next steps:
 /plan mobile app for fitness tracking
 /plan SaaS dashboard with analytics
 ```
+
+---
+
+## 📝 Decision Logging (MANDATORY)
+
+After EVERY clarifying question is answered, log to `.agent/memory/decisions.md`:
+
+```markdown
+## Session: [YYYY-MM-DD HH:MM]
+
+### Q: [Question asked]
+**A:** [User's answer]
+**Impact:** [How this affects plan]
+```
+
+> 🔴 Log decisions BEFORE writing plan. Use logged decisions as source of truth.
+

package/.agent/workflows/preview.md

@@ -1,6 +1,7 @@
 ---
 description: Manage preview server - start, stop, and status check. Manages local development server.
 description_vi: Quản lý máy chủ xem trước - bắt đầu, dừng và kiểm tra trạng thái. Quản lý máy chủ phát triển cục bộ.
+description_en: Manage preview server - start, stop, and status check. Manages local development server.
 ---
 
 # /preview - Preview Management

package/.agent/workflows/quality.md

@@ -1,6 +1,7 @@
 ---
 description: Optimize context and output quality for Codex CLI. Used for complex tasks requiring high-quality results.
 description_vi: Tối ưu hóa ngữ cảnh và chất lượng đầu ra cho Codex CLI. Được sử dụng cho các tác vụ phức tạp đòi hỏi kết quả chất lượng cao.
+description_en: Optimize context and output quality for Codex CLI. Used for complex tasks requiring high-quality results.
 ---
 
 # /quality - Quality Optimization Workflow

package/.agent/workflows/saas.md

@@ -1,5 +1,7 @@
 ---
 description: SaaS launch workflow. From landing page to billing integration.
+description_en: SaaS launch workflow. From landing page to billing integration.
+description_vi: Workflow ra mắt sản phẩm SaaS. Từ trang landing page đến tích hợp thanh toán.
 ---
 
 # 🚀 /saas - SaaS Launch Workflow

package/.agent/workflows/spec.md

@@ -1,6 +1,7 @@
 ---
 description: Create specification documents before planning. Used for complex features requiring clear requirements before implementation.
 description_vi: Tạo tài liệu đặc tả trước khi lập kế hoạch. Được sử dụng cho các tính năng phức tạp yêu cầu yêu cầu rõ ràng trước khi triển khai.
+description_en: Create specification documents before planning. Used for complex features requiring clear requirements before implementation.
 ---
 
 # /spec - Specification Writing Mode
@@ -36,8 +37,32 @@ Create a clear, testable specification document that serves as the "source of tr
 
 ---
 
+## 🔴 VIBES PRE-FLIGHT (MANDATORY)
+
+Before writing spec:
+1. **V** - Define success: "Spec approved when user marks status as Approved"
+2. **I** - Extract explicit/implicit needs using `@[skills/intent-capture]`
+3. **B** - Identify scope bifurcations: MVP vs Full Feature
+4. **E** - Define emergence boundaries (what's OUT of scope)
+5. **S** - Ensure testable acceptance criteria
+
+> ⚠️ **HALT** if VIBES check is incomplete.
+
+---
+
 ## Task
 
+## 💾 Git Checkpoint (Safety First)
+
+Before starting work:
+1. [ ] **Clean working tree?** (Run `git status`)
+2. [ ] **On correct branch?** (Not `main` or `master`)
+3. [ ] **Latest pulled?** (Run `git pull`)
+
+> 🔴 **STOP** if git state is dirty. Ask user to commit or stash first.
+
+---
+
 Use the `spec-writing` skill with this context:
 
 ```
@@ -188,3 +213,20 @@ Next steps:
 /spec real-time chat with WebSocket
 /spec admin dashboard with analytics
 ```
+
+---
+
+## 📝 Decision Logging (MANDATORY)
+
+After EVERY clarifying question is answered, log to `.agent/memory/decisions.md`:
+
+```markdown
+## Session: [YYYY-MM-DD HH:MM]
+
+### Q: [Question asked]
+**A:** [User's answer]
+**Impact:** [How this affects spec]
+```
+
+> 🔴 Log decisions BEFORE writing spec. Use logged decisions as source of truth.
+

package/.agent/workflows/status.md

@@ -1,6 +1,7 @@
 ---
 description: Display project and agent status. Track progress and status boards.
 description_vi: Hiển thị trạng thái dự án và Agent. Theo dõi tiến độ và bảng trạng thái.
+description_en: Display project and agent status. Track progress and status boards.
 ---
 
 # /status - Show Status

package/.agent/workflows/test.md

@@ -1,6 +1,7 @@
 ---
 description: Create and run tests. Create and execute tests for the code.
 description_vi: Tạo và chạy các bài kiểm tra. Tạo và thực thi kiểm tra cho mã nguồn.
+description_en: Create and run tests. Create and execute tests for the code.
 ---
 
 # /test - Test Generation and Execution
@@ -27,6 +28,19 @@ This command generates tests, runs existing tests, or checks test coverage.
 
 ---
 
+## 🔴 VIBES PRE-FLIGHT (MANDATORY)
+
+Before testing:
+1. **V** - Define success: "All tests pass with exit code 0"
+2. **I** - Capture intent: What behavior must be verified?
+3. **B** - Bifurcation: Unit vs Integration vs E2E?
+4. **E** - Emergence: Check for phantom dependencies in test stubs
+5. **S** - Safety: Ensure tests don't have side effects
+
+> ⚠️ **HALT** if VIBES check is incomplete.
+
+---
+
 ## Behavior
 
 ### Generate Tests

package/.agent/workflows/ui-ux-pro-max.md

@@ -1,6 +1,7 @@
 ---
 description: Professional UI/UX planning and implementation.
 description_vi: Lập kế hoạch và triển khai UI/UX chuyên nghiệp.
+description_en: Professional UI/UX planning and implementation.
 ---
 
 # UI/UX Pro Max - Design Intelligence