anchi-toolkit 1.0.0

package/.cursor/commands/undo.md

@@ -0,0 +1,166 @@
+---
+description: ⚡⚡ Rollback last workflow or specific changes
+argument-hint: [optional: steps to undo or commit hash]
+---
+
+# /undo - Rollback Changes
+
+**What:** Rollback last workflow hoặc specific changes.
+
+## Usage
+```
+/undo                      # Undo last action
+/undo 3                    # Undo last 3 steps
+/undo "abc1234"            # Undo to specific commit
+/undo --list               # Show undo history
+```
+
+---
+
+## 💰 Model Selection
+
+| Action | Recommended Model | Cost |
+|--------|-------------------|------|
+| Simple undo | GPT-5.1 Codex Mini | 💚 Low |
+| Complex rollback | Sonnet 4.5 | 💛 Medium |
+
+> Undo operations primarily use git - minimal AI needed.
+
+---
+
+## Workflow
+
+```
+┌─────────────────────────────────────────────────────────────┐
+│ ↩️ UNDO WORKFLOW                                            │
+├─────────────────────────────────────────────────────────────┤
+│                                                             │
+│  Step 1: ANALYZE                                            │
+│  [AUTO] Identify last workflow/changes                      │
+│                                                             │
+│  Step 2: SHOW                                               │
+│  Display: What will be undone                               │
+│                                                             │
+│  Step 3: CONFIRM                                            │
+│  User: Confirm or select specific items                     │
+│                                                             │
+│  Step 4: EXECUTE                                            │
+│  [AUTO] Rollback changes                                    │
+│  [AUTO] Update memory                                       │
+│                                                             │
+└─────────────────────────────────────────────────────────────┘
+```
+
+---
+
+## Output
+
+```
+↩️ Undo Available Actions
+
+╔═════════════════════════════════════════════════════════════╗
+║ # │ Time     │ Action            │ Files Changed            ║
+╠═════════════════════════════════════════════════════════════╣
+║ 1 │ 2 min    │ /cook LoginForm   │ 3 files                  ║
+║ 2 │ 10 min   │ /generate:api     │ 2 files                  ║
+║ 3 │ 15 min   │ /theme dark       │ 2 files                  ║
+║ 4 │ 30 min   │ /auto:rapid       │ 8 files                  ║
+╚═════════════════════════════════════════════════════════════╝
+
+📋 Latest Action Details (#1):
+  Command: /cook "Add LoginForm"
+  Time: 2 minutes ago
+  
+  Files to restore:
+  ├── components/LoginForm.tsx (delete)
+  ├── app/login/page.tsx (revert)
+  └── lib/auth.ts (revert)
+
+⚠️ This will:
+  • Delete 1 new file
+  • Revert 2 files to previous state
+  • Remove 1 commit
+
+👉 Undo this action? (ok/select #/cancel)
+```
+
+---
+
+## Undo Options
+
+```
+/undo               # Undo last action (confirmed)
+/undo --force       # Undo without confirmation
+/undo --soft        # Keep files, just update memory
+/undo --list        # Show all undo options
+/undo --all         # Undo entire session
+```
+
+---
+
+## Safety Features
+
+| Feature | Description |
+|---------|-------------|
+| **Backup** | Auto-backup before undo |
+| **Selective** | Choose specific items |
+| **Preview** | See what will change |
+| **Memory Sync** | Update AI memory |
+
+---
+
+## Undo History
+
+```
+/undo --list
+
+📜 UNDO HISTORY (Last 24h)
+
+| # | Command | Files | Commit | Undoable |
+|---|---------|-------|--------|----------|
+| 1 | /cook LoginForm | 3 | abc123 | ✅ |
+| 2 | /generate:api | 2 | def456 | ✅ |
+| 3 | /theme dark | 2 | - | ✅ |
+| 4 | /auto:rapid | 8 | ghi789 | ✅ |
+| 5 | /fix:types | 1 | jkl012 | ✅ |
+
+💡 Tip: /undo 4 to undo back to step 4
+```
+
+---
+
+## Agent & Skill Activation
+
+**Agents:**
+- `.cursor/agents/git-manager.md` - Git operations
+
+**Skills:**
+- `devops/` - Version control
+
+---
+
+## Configuration
+
+```yaml
+undo:
+  max_history: 50
+  auto_backup: true
+  backup_dir: .undo/
+  confirm_before: true
+```
+
+---
+
+##  Dynamic Orchestration
+
+> **AI MUST read orchestration config before execution**
+
+```
+1. READ .cursor/agent-skill-index.yaml
+2. CHECK command_mapping for this command
+3. APPLY skill activation rules
+4. CHECK .cursor/failure-modes.yaml
+5. LOG decision to .ai-audit/
+```
+
+ [docs/ORCHESTRATION_RUNTIME.md](../docs/ORCHESTRATION_RUNTIME.md)

package/.cursor/commands/why.md

@@ -0,0 +1,155 @@
+---
+description: ⚡⚡ Explain AI decisions and reasoning
+argument-hint: [optional: --last | --command <name>]
+---
+
+# /why - AI Decision Explainability
+
+**What:** Giải thích vì sao AI chọn agent, skill, và rules cho quyết định.
+
+## Usage
+```
+/why                    # Explain last decision
+/why --last 3           # Explain last 3 decisions
+/why --command "/do"    # Explain specific command type
+```
+
+---
+
+## 💰 Model Selection
+
+| Action | Recommended Model | Cost |
+|--------|-------------------|------|
+| Explain decision | GPT-5.1 Codex Mini | 💚 Low |
+
+---
+
+## Output
+
+```
+/why
+
+═══════════ DECISION EXPLANATION ═══════════
+
+📋 Last Command: /do "Add user authentication"
+
+🤖 AGENT SELECTION:
+
+   ┌─────────────────────────────────────────────────────────────┐
+   │ PRIMARY: fullstack-developer                                │
+   │ ├── WHY: Task involves both frontend (login page) +        │
+   │ │        backend (API routes, session handling)            │
+   │ └── Confidence: 0.85                                       │
+   ├─────────────────────────────────────────────────────────────┤
+   │ SECONDARY: database-admin                                   │
+   │ ├── WHY: User table schema changes detected                │
+   │ └── Auto-escalated for: schema modification                │
+   ├─────────────────────────────────────────────────────────────┤
+   │ NOT SELECTED: security-auditor                              │
+   │ ├── WHY: Basic auth flow, no payment/sensitive data        │
+   │ └── Would activate if: payment, admin access, API keys     │
+   └─────────────────────────────────────────────────────────────┘
+
+📚 SKILL ACTIVATION:
+
+   ┌─────────────────────────────────────────────────────────────┐
+   │ ✅ ACTIVATED:                                               │
+   │ ├── better-auth/                                            │
+   │ │   └── Trigger: Detected "next-auth" in package.json     │
+   │ ├── databases/                                              │
+   │ │   └── Trigger: prisma/schema.prisma modification needed  │
+   │ └── frontend-development/                                   │
+   │     └── Trigger: Creating login page component             │
+   ├─────────────────────────────────────────────────────────────┤
+   │ ❌ NOT ACTIVATED:                                           │
+   │ ├── payment-integration/                                    │
+   │ │   └── WHY: No payment keywords detected                  │
+   │ └── shopify/                                                │
+   │     └── WHY: Not an e-commerce context                     │
+   └─────────────────────────────────────────────────────────────┘
+
+📜 RULES APPLIED:
+
+   ┌─────────────────────────────────────────────────────────────┐
+   │ CONTRACT RULES:                                             │
+   │ ├── P2: "Use service layer"                                │
+   │ │   └── Applied: Created AuthService                       │
+   │ ├── S1: "No direct DB in routes"                           │
+   │ │   └── Applied: Used UserRepository                       │
+   │ └── C1: "Match existing patterns"                          │
+   │     └── Applied: Followed project's naming conventions     │
+   ├─────────────────────────────────────────────────────────────┤
+   │ PRESET RULES (Professional):                                │
+   │ ├── "Modular structure required" → Created auth module     │
+   │ └── "Tests required" → Added auth.test.ts                  │
+   └─────────────────────────────────────────────────────────────┘
+
+📊 RISK ASSESSMENT:
+
+   ┌─────────────────────────────────────────────────────────────┐
+   │ Risk Level: MEDIUM                                          │
+   │ ├── Factors:                                                │
+   │ │   ├── + Authentication = sensitive                       │
+   │ │   ├── - Using established library (next-auth)            │
+   │ │   └── - Following project patterns                       │
+   │ ├── Confidence: 0.85                                        │
+   │ └── Actions Taken:                                          │
+   │     └── Auto-enabled code review                           │
+   └─────────────────────────────────────────────────────────────┘
+
+💡 ALTERNATIVE APPROACHES CONSIDERED:
+
+   1. JWT tokens → Rejected (project uses session-based)
+   2. Custom auth → Rejected (next-auth already configured)
+   3. OAuth only → Rejected (user requested email/password)
+```
+
+---
+
+## Flags
+
+| Flag | Description |
+|------|-------------|
+| `--last N` | Explain last N decisions |
+| `--command <name>` | Filter by command type |
+| `--detailed` | Show full decision tree |
+| `--quiet` | Summary only |
+
+---
+
+## Use Cases
+
+| Scenario | Command |
+|----------|---------|
+| Debug AI decision | `/why` |
+| Onboard new team member | `/why --last 5` |
+| Audit AI behavior | `/why --command "/do"` |
+| Understand skill selection | `/why --detailed` |
+
+---
+
+## Agent & Skill Activation
+
+**Primary Agents:**
+- `.cursor/agents/docs-manager.md` - Explanation generation
+
+**Skills:**
+- `document-skills/` - Clear communication
+
+> /why reads from audit logs and orchestration config.
+
+---
+
+##  Dynamic Orchestration
+
+> **AI MUST read orchestration config before execution**
+
+```
+1. READ .cursor/agent-skill-index.yaml
+2. CHECK command_mapping for this command
+3. APPLY skill activation rules
+4. CHECK .cursor/failure-modes.yaml
+5. LOG decision to .ai-audit/
+```
+
+ [docs/ORCHESTRATION_RUNTIME.md](../docs/ORCHESTRATION_RUNTIME.md)

package/.cursor/custom-commands/example.md

@@ -0,0 +1,30 @@
+---
+description: Custom command example - Copy and modify this template
+argument-hint: [description]
+---
+
+# My Custom Command
+
+**What it does:** [Describe your command]
+
+## Usage
+```
+/my-command "description"
+```
+
+## Steps
+
+1. Read relevant files
+2. Analyze context
+3. Execute action
+4. Report results
+
+## Example
+
+```
+/my-command "Add authentication to /api/users"
+```
+
+---
+
+> 💡 **Tip:** Create your own commands in `.cursor/custom-commands/`

package/.cursor/failure-modes.yaml

@@ -0,0 +1,148 @@
+# Failure Modes & Self-Restraint
+# When AI must stop, ask, or escalate
+# ═══════════════════════════════════════════════════════════════
+
+# ═══════════════════════════════════════════════════════════════
+# STOP CONDITIONS
+# ═══════════════════════════════════════════════════════════════
+
+stop_conditions:
+  # Confidence too low
+  low_confidence:
+    condition: "confidence < 0.5"
+    action: ask_user
+    message: |
+      ⚠️ I'm not confident about this approach.
+      My confidence level: {confidence}
+      
+      Options:
+      1. Provide more context
+      2. Try a different approach
+      3. Cancel
+
+  # Missing context
+  missing_context:
+    condition: "required_context_missing"
+    action: request_info
+    message: |
+      ❓ I need more information:
+      - {missing_items}
+      
+      Please provide the above, or I can make assumptions.
+
+  # Architecture violation
+  architecture_violation:
+    condition: "violates_preset_contract"
+    action: stop
+    message: |
+      🛑 This would violate the project architecture.
+      
+      Preset: {preset}
+      Rule violated: {rule}
+      
+      Options:
+      1. Modify approach to comply
+      2. Request architecture override (requires lead approval)
+      3. Cancel
+
+  # Security concern
+  security_concern:
+    condition: "potential_security_risk"
+    action: escalate
+    escalate_to: security-auditor
+    message: |
+      🔒 Security concern detected.
+      Escalating to security-auditor for review.
+
+  # Outside expertise
+  outside_expertise:
+    condition: "agent_boundary_exceeded"
+    action: escalate
+    message: |
+      🔄 This requires expertise I don't have.
+      Escalating to: {appropriate_agent}
+
+# ═══════════════════════════════════════════════════════════════
+# SELF-RESTRAINT RULES (Never do these)
+# ═══════════════════════════════════════════════════════════════
+
+restraints:
+  # Credentials & Secrets
+  - id: R001
+    name: "Never guess credentials"
+    triggers: ["password", "api_key", "secret", "token"]
+    action: "Always ask user for credentials, never generate"
+    
+  - id: R002
+    name: "Never expose secrets"
+    triggers: ["console.log", "print", "response.json"]
+    check: "Does output contain sensitive data?"
+    action: "Sanitize before output"
+
+  # File Operations
+  - id: R003
+    name: "Never delete without confirmation"
+    triggers: ["rm", "delete", "remove", "drop"]
+    action: "Require explicit user confirmation"
+    
+  - id: R004
+    name: "Never modify .env without confirmation"
+    triggers: [".env", "environment"]
+    action: "Show changes and ask for approval"
+
+  # Deployment
+  - id: R005
+    name: "Never deploy without approval"
+    triggers: ["deploy", "publish", "release", "npm publish"]
+    action: "Require explicit lead approval"
+    
+  - id: R006
+    name: "Never push to production"
+    triggers: ["production", "main branch", "master branch"]
+    action: "Stop and require multi-person approval"
+
+  # Security
+  - id: R007
+    name: "Never skip security review for auth/payment"
+    triggers: ["auth", "login", "payment", "checkout"]
+    action: "Always engage security-auditor"
+    
+  - id: R008
+    name: "Never add backdoors"
+    triggers: ["bypass", "skip auth", "admin access"]
+    action: "Stop and explain security implications"
+
+# ═══════════════════════════════════════════════════════════════
+# WARNING CONDITIONS (Proceed with caution)
+# ═══════════════════════════════════════════════════════════════
+
+warnings:
+  - condition: "0.5 <= confidence < 0.7"
+    message: "⚠️ Proceeding with medium confidence. Review recommended."
+    auto_enable_review: true
+    
+  - condition: "no_tests_for_new_code"
+    message: "⚠️ No tests added for new functionality."
+    suggest: "/test to add coverage"
+    
+  - condition: "large_change"
+    threshold: "files > 10 OR lines > 500"
+    message: "⚠️ Large change detected. Breaking into phases recommended."
+
+# ═══════════════════════════════════════════════════════════════
+# RECOVERY ACTIONS
+# ═══════════════════════════════════════════════════════════════
+
+recovery:
+  on_error:
+    - log_error_to_audit
+    - suggest_rollback
+    - offer_debug_session
+    
+  on_user_cancel:
+    - save_progress_to_memory
+    - log_cancellation_reason
+    
+  on_escalation:
+    - transfer_context_to_new_agent
+    - log_escalation_reason

package/.cursor/hooks/.env.example

@@ -0,0 +1,15 @@
+# Claude Code Hooks - Local Environment Variables
+# This file is for hook-specific environment variables (lowest priority)
+# Priority: process.env > .cursor/.env > .cursor/hooks/.env
+# Copy this file to .env and fill in your actual values
+
+# ============================================
+# Discord Notifications
+# ============================================
+DISCORD_WEBHOOK_URL=
+
+# ============================================
+# Telegram Notifications
+# ============================================
+TELEGRAM_BOT_TOKEN=
+TELEGRAM_CHAT_ID=