@buivietphi/skill-mobile-mt 1.2.0 → 1.4.0

package/AGENTS.md

@@ -1,6 +1,6 @@
 # Skill Mobile MT — Agent Rules
 
-> Multi-agent compatibility layer for Claude Code, Codex, Gemini CLI, Kimi, Cursor, Copilot, Windsurf, Antigravity, and all Agent Skills-compatible tools.
+> Multi-agent compatibility layer for Claude Code, Cline, Roo Code, Cursor, Windsurf, Copilot, Codex, Gemini CLI, Kimi, Kilo Code, Kiro, Antigravity, and all Agent Skills-compatible tools.
 
 ---
 
@@ -9,12 +9,16 @@
 | Agent | How it loads rules | Setup | Think Block |
 |-------|-------------------|-------|-------------|
 | Claude Code | `~/.claude/skills/` (auto) | `npx skill-mobile-mt --claude` | `<think>...</think>` |
+| **Cline** | **`.clinerules/` in project root** | **`npx skill-mobile-mt --init cline`** | Inline reasoning |
+| **Roo Code** | **`.roo/rules/` in project root** | **`npx skill-mobile-mt --init roocode`** | Inline reasoning |
+| **Cursor** | **`.cursorrules` in project root** | **`npx skill-mobile-mt --init cursor`** | Inline reasoning |
+| **Windsurf** | **`.windsurfrules` in project root** | **`npx skill-mobile-mt --init windsurf`** | Inline reasoning |
+| **Copilot** | **`.github/copilot-instructions.md`** | **`npx skill-mobile-mt --init copilot`** | `// PLAN:` comments |
 | Codex | `~/.codex/skills/` (auto) | `npx skill-mobile-mt --codex` | `<think>...</think>` |
 | Gemini CLI | `~/.gemini/skills/` (auto) | `npx skill-mobile-mt --gemini` | `## Thinking:` block |
 | Kimi | `~/.kimi/skills/` (manual) | `npx skill-mobile-mt --kimi` | `【思考】` or markdown |
-| **Cursor** | **`.cursorrules` in project root** | **`npx skill-mobile-mt --init cursor`** | Inline reasoning |
-| **Copilot** | **`.github/copilot-instructions.md`** | **`npx skill-mobile-mt --init copilot`** | `// PLAN:` comments |
-| **Windsurf** | **`.windsurfrules` in project root** | **`npx skill-mobile-mt --init windsurf`** | Inline reasoning |
+| **Kilo Code** | **`.kilocode/rules/` in project root** | **`npx skill-mobile-mt --init kilocode`** | Inline reasoning |
+| **Kiro** | **`.kiro/steering/` in project root** | **`npx skill-mobile-mt --init kiro`** | Inline reasoning |
 | Antigravity | `~/.agents/skills/` (auto) | `npx skill-mobile-mt --antigravity` | Agent-native format |
 
 ---
@@ -30,13 +34,13 @@ skill-mobile-mt/
 │   └── react-native.md             ← RN + Expo patterns (5,108 tokens)
 │
 ├── flutter/
-│   └── flutter.md                  ← Flutter + Dart patterns (1,617 tokens)
+│   └── flutter.md                  ← Flutter + Dart 3.x patterns (2,100 tokens)
 │
 ├── ios/
 │   └── ios-native.md               ← Swift + UIKit/SwiftUI patterns (1,452 tokens)
 │
 ├── android/
-│   └── android-native.md           ← Kotlin + Compose patterns (1,659 tokens)
+│   └── android-native.md           ← Kotlin + Compose patterns (2,100 tokens)
 │
 └── shared/
     │
@@ -50,21 +54,24 @@ skill-mobile-mt/
     ├── document-analysis.md         ← Parse images/PDFs/DOCX → code (1,200 tokens)
     ├── anti-patterns.md             ← PII, cardinality, payload detection (2,800 tokens)
     ├── performance-prediction.md    ← Frame budget, FPS prediction (1,500 tokens)
-    ├── platform-excellence.md       ← iOS 18+ vs Android 15+ UX (1,500 tokens)
+    ├── platform-excellence.md       ← iOS 18+ vs Android 15+ UX + HIG (2,200 tokens)
     ├── version-management.md        ← SDK compat matrix + release testing (3,500 tokens)
     ├── observability.md             ← Sessions as 4th pillar (3,000 tokens)
     ├── architecture-intelligence.md ← Patterns from 30+ production repos (4,500 tokens)
     ├── common-pitfalls.md           ← Known issue patterns (1,160 tokens)
     ├── release-checklist.md         ← App Store/Play Store checklist (587 tokens)
     │
+    ├── offline-first.md             ← Local-first + sync patterns (2,566 tokens)
+    ├── on-device-ai.md              ← Core ML / TFLite / llama.cpp patterns (700 tokens)
+    │
     ├── ── TEMPLATES (copy to your project) ────────────────────
     ├── claude-md-template.md        ← CLAUDE.md for Claude Code (copy to project root)
     └── agent-rules-template.md      ← Rules for ALL agents: Cursor/.cursorrules, Windsurf/.windsurfrules, Copilot/.github/copilot-instructions.md, Codex/AGENTS.md, Gemini/GEMINI.md, Antigravity YAML
 ```
 
 **Token totals:**
-- Smart load (1 platform + core shared): **~14,200 tokens** (11.1% of 128K)
-- Full load (all files): **~41,700 tokens** (32.6% of 128K)
+- Smart load (1 platform + core shared): **~38,600 tokens** (30.2% of 128K)
+- Full load (all files): **~70,000 tokens** (54.7% of 128K)
 
 ---
 
@@ -76,13 +83,13 @@ skill-mobile-mt/
 
 **Loads automatically:**
 ```
-SKILL.md                           (6,100 tokens)
-+ 1 platform file                  (~1,600–5,100 tokens depending on platform)
-+ shared/code-review.md            (865 tokens)
-+ shared/bug-detection.md          (499 tokens)
-+ shared/prompt-engineering.md     (3,927 tokens)
+SKILL.md                           (~13,200 tokens)
++ 1 platform file                  (~1,580–5,730 tokens depending on platform)
++ shared/code-review.md            (~1,500 tokens)
++ shared/bug-detection.md          (~800 tokens)
++ shared/prompt-engineering.md     (~5,600 tokens)
 ─────────────────────────────────────────────────
-≈ 14,200 tokens total
+≈ 38,600 tokens total (estimated)
 ```
 
 **Use case:** Regular coding, new features, code review. Covers 90% of daily work.
@@ -103,6 +110,7 @@ The agent reads the task, then decides which extra file to load:
 | "Install this package / upgrade SDK" | `shared/version-management.md` |
 | "Prepare for App Store / Play Store" | `shared/release-checklist.md` |
 | "Weird issue, not sure why" | `shared/common-pitfalls.md` |
+| "On-device AI / ML / inference" | `shared/on-device-ai.md` |
 
 **Load cost:** +500 to +3,500 tokens per on-demand file.
 
@@ -112,7 +120,7 @@ The agent reads the task, then decides which extra file to load:
 
 **No automatic trigger.** Full load happens when the AI reads every file without being selective — either because it's over-eager, or because the user explicitly asks for it.
 
-**Total:** ~41,700 tokens (32.6% of 128K, 20.9% of 200K)
+**Total:** ~70,000 tokens (54.7% of 128K, 35% of 200K)
 
 **How it actually works:**
 - `@skill-mobile-mt` only injects SKILL.md into context
@@ -137,7 +145,7 @@ The agent reads the task, then decides which extra file to load:
 ```yaml
 skill:
   name: skill-mobile-mt
-  version: "1.2.0"
+  version: "1.4.0"
   author: buivietphi
   category: engineering
   tags:
@@ -170,6 +178,7 @@ skill:
         # - shared/document-analysis.md
         # - shared/release-checklist.md
         # - shared/common-pitfalls.md
+        # - shared/on-device-ai.md
 
     project:
       description: "Read current project, adapt to its framework and conventions"
@@ -204,9 +213,9 @@ skill:
     java: ".java files in app/src/"
 
   context_budget:
-    max_tokens: 41700
-    smart_load_tokens: 14200
-    savings: "~66%"
+    max_tokens: 70000
+    smart_load_tokens: 38600
+    savings: "~45%"
 ```
 
 ---
@@ -252,6 +261,7 @@ Every agent MUST follow this loading sequence:
    - shared/observability.md          (when adding logging, analytics, crash tracking)
    - shared/common-pitfalls.md        (when encountering unfamiliar errors)
    - shared/release-checklist.md      (when preparing for App Store/Play Store submission)
+   - shared/on-device-ai.md           (when adding Core ML / TFLite / on-device inference)
 
 7. SKIP non-matching platform subfolders (saves ~66% context)
 ```
@@ -273,6 +283,7 @@ Priority 6 (ON-DEMAND): shared/observability.md — Sessions as 4th pillar
 Priority 6 (ON-DEMAND): shared/document-analysis.md — Parse images/PDFs → code
 Priority 6 (ON-DEMAND): shared/release-checklist.md — Pre-release verification
 Priority 6 (ON-DEMAND): shared/common-pitfalls.md — Known issue patterns
+Priority 6 (ON-DEMAND): shared/on-device-ai.md — Core ML / TFLite / llama.cpp
 ```
 
 ---
@@ -299,24 +310,51 @@ Priority 6 (ON-DEMAND): shared/common-pitfalls.md — Known issue patterns
 - Supports both Chinese and English prompts
 - Think blocks use `【思考】` format
 
+### Cline
+- Reads `.clinerules/` directory from project root (also supports `.clinerules` single file)
+- Also reads from `~/.cline/skills/` for global skills
+- Run `npx @buivietphi/skill-mobile-mt --init cline` to generate `.clinerules/mobile-rules.md`
+- Supports conditional rules with YAML frontmatter `paths:` field
+
+### Roo Code
+- Reads `.roo/rules/` directory from project root (recursive, multi-file)
+- Also reads from `~/.roo/skills/` for global skills
+- Supports mode-specific rules: `.roo/rules-code/`, `.roo/rules-architect/`
+- Run `npx @buivietphi/skill-mobile-mt --init roocode` to generate `.roo/rules/mobile-rules.md`
+- Auto-loads `AGENTS.md` from workspace root
+
 ### Cursor
-- **Does NOT read from `~/.cursor/skills/`** — reads `.cursorrules` from project root
-- Run `npx @buivietphi/skill-mobile-mt --init cursor` in your project to generate `.cursorrules`
+- Reads `.cursorrules` from project root
+- Run `npx @buivietphi/skill-mobile-mt --init cursor` to generate `.cursorrules`
 - The generated file includes auto-detected framework, rules, and security patterns
 - Think blocks embedded as inline reasoning in Composer
 
 ### GitHub Copilot
-- **Does NOT read from `~/.copilot/skills/`** — reads `.github/copilot-instructions.md`
-- Run `npx @buivietphi/skill-mobile-mt --init copilot` in your project to generate the file
+- Reads `.github/copilot-instructions.md` from project
+- Run `npx @buivietphi/skill-mobile-mt --init copilot` to generate the file
 - The generated file includes code patterns, required templates, and rules
 - Think blocks as `// PLAN:` comments before code
 
 ### Windsurf
-- **Does NOT read from `~/.windsurf/skills/`** — reads `.windsurfrules` from project root
-- Run `npx @buivietphi/skill-mobile-mt --init windsurf` in your project to generate `.windsurfrules`
+- Reads `.windsurfrules` from project root
+- Run `npx @buivietphi/skill-mobile-mt --init windsurf` to generate `.windsurfrules`
 - The generated file includes coding rules, security rules, and architecture patterns
 - Think blocks as inline reasoning
 
+### Kilo Code
+- Reads `.kilocode/rules/` directory from project root
+- Also reads from `~/.kilocode/rules/` for global rules
+- Supports mode-specific rules: `.kilocode/rules-code/`, `.kilocode/rules-architect/`
+- Run `npx @buivietphi/skill-mobile-mt --init kilocode` to generate `.kilocode/rules/mobile-rules.md`
+- Auto-loads `AGENTS.md` from workspace root
+
+### Kiro (AWS)
+- Reads `.kiro/steering/` directory from project root
+- Also reads from `~/.kiro/steering/` for global steering
+- Uses YAML frontmatter with `inclusion: always|fileMatch|manual|auto`
+- Run `npx @buivietphi/skill-mobile-mt --init kiro` to generate `.kiro/steering/mobile-rules.md`
+- Separate specs system in `.kiro/specs/` for feature development
+
 ### Antigravity
 - Orchestrator loads based on detected project type
 - Follows `platform_detection` rules above
@@ -332,12 +370,16 @@ These are separate from the skill files — they go in the user's project, not i
 | Agent | File | Location |
 |-------|------|----------|
 | Claude Code | `CLAUDE.md` | Project root |
+| Cline | `.clinerules/` directory | Project root |
+| Roo Code | `.roo/rules/` directory | Project root |
 | Cursor | `.cursorrules` | Project root |
 | Windsurf | `.windsurfrules` | Project root |
 | GitHub Copilot | `.github/copilot-instructions.md` | `.github/` folder |
 | Codex | `AGENTS.md` | Project root |
 | Gemini CLI | `GEMINI.md` | Project root |
 | Kimi | No auto-load — paste as context | — |
+| Kilo Code | `.kilocode/rules/` directory | Project root |
+| Kiro | `.kiro/steering/` directory | Project root |
 | Antigravity | YAML `context.rules` field | Antigravity config |
 
 **Templates for all agents:** `shared/agent-rules-template.md`
@@ -356,11 +398,14 @@ Copy the relevant section to your project to enable auto-check rules in every se
 ### Skill directory install (agents that read from skills/)
 
 ```bash
-# Claude Code (global)
+# Claude Code
 ~/.claude/skills/skill-mobile-mt/
 
-# Claude Code (project-local)
-.claude/skills/skill-mobile-mt/
+# Cline
+~/.cline/skills/skill-mobile-mt/
+
+# Roo Code
+~/.roo/skills/skill-mobile-mt/
 
 # Codex
 ~/.codex/skills/skill-mobile-mt/
@@ -371,6 +416,12 @@ Copy the relevant section to your project to enable auto-check rules in every se
 # Kimi
 ~/.kimi/skills/skill-mobile-mt/
 
+# Kilo Code
+~/.kilocode/skills/skill-mobile-mt/
+
+# Kiro
+~/.kiro/skills/skill-mobile-mt/
+
 # Antigravity (shared agent directory)
 ~/.agents/skills/skill-mobile-mt/
 
@@ -380,8 +431,7 @@ npx @buivietphi/skill-mobile-mt --path /your/custom/path
 
 ### Project-level files (agents that read from project root)
 
-These agents read rules from project-level files, NOT from a skills directory.
-Use `--init` to generate them:
+These agents read rules from project-level files. Use `--init` to generate them:
 
 ```bash
 # Generate all project-level files (interactive selector)
@@ -389,9 +439,13 @@ npx @buivietphi/skill-mobile-mt --init
 
 # Generate specific agent file
 npx @buivietphi/skill-mobile-mt --init cursor     # → .cursorrules
+npx @buivietphi/skill-mobile-mt --init cline      # → .clinerules/mobile-rules.md
+npx @buivietphi/skill-mobile-mt --init roocode    # → .roo/rules/mobile-rules.md
 npx @buivietphi/skill-mobile-mt --init copilot    # → .github/copilot-instructions.md
 npx @buivietphi/skill-mobile-mt --init windsurf   # → .windsurfrules
-npx @buivietphi/skill-mobile-mt --init all        # → all three files
+npx @buivietphi/skill-mobile-mt --init kilocode   # → .kilocode/rules/mobile-rules.md
+npx @buivietphi/skill-mobile-mt --init kiro       # → .kiro/steering/mobile-rules.md
+npx @buivietphi/skill-mobile-mt --init all        # → all files
 ```
 
 **What `--init` does:**
@@ -408,7 +462,7 @@ npx @buivietphi/skill-mobile-mt --init all        # → all three files
 {
   "id": "skill-mobile-mt",
   "name": "skill-mobile-mt",
-  "version": "1.2.0",
+  "version": "1.4.0",
   "author": "buivietphi",
   "category": "engineering",
   "description": "Master Senior Mobile Engineer. Pre-built patterns from 18 production apps + project adaptation. Auto-detects language and framework. React Native, Flutter, iOS, Android.",
@@ -416,6 +470,6 @@ npx @buivietphi/skill-mobile-mt --init all        # → all three files
   "source": "buivietphi (MIT)",
   "platforms": ["react-native", "flutter", "ios", "android"],
   "languages": ["typescript", "javascript", "dart", "swift", "kotlin", "java"],
-  "agents": ["claude-code", "codex", "gemini", "kimi", "cursor", "copilot", "windsurf", "antigravity"]
+  "agents": ["claude-code", "cline", "roo-code", "cursor", "windsurf", "copilot", "codex", "gemini", "kimi", "kilo-code", "kiro", "antigravity"]
 }
 ```

package/README.md

@@ -1,6 +1,6 @@
 # @buivietphi/skill-mobile-mt
 
-**Master Senior Mobile Engineer** — AI skill for Claude Code, Codex, Gemini CLI, Kimi, Cursor, Copilot, Windsurf, and Antigravity.
+**Master Senior Mobile Engineer** — AI skill for Claude Code, Cline, Roo Code, Cursor, Windsurf, Copilot, Codex, Gemini CLI, Kimi, Kilo Code, Kiro, and Antigravity.
 
 Pre-built architecture patterns from **30+ production repos (200k+ GitHub stars)** including Ignite, Expensify, Mattermost, Immich, AppFlowy, Now in Android, TCA + auto-adaptation to your current project.
 
@@ -24,16 +24,20 @@ npx @buivietphi/skill-mobile-mt --all          # All agents
 npx @buivietphi/skill-mobile-mt --path ./dir   # Custom directory
 ```
 
-### Generate project-level rules (Cursor, Copilot, Windsurf)
+### Generate project-level rules
 
-Cursor, Copilot, and Windsurf read rules from **project-level files**, not from a skills directory. Run `--init` inside your project to generate them:
+Cline, Roo Code, Cursor, Windsurf, Copilot, Kilo Code, and Kiro read rules from **project-level files**. Run `--init` inside your project to generate them:
 
 ```bash
 npx @buivietphi/skill-mobile-mt --init            # Interactive selector
 npx @buivietphi/skill-mobile-mt --init cursor     # .cursorrules
+npx @buivietphi/skill-mobile-mt --init cline      # .clinerules/mobile-rules.md
+npx @buivietphi/skill-mobile-mt --init roocode    # .roo/rules/mobile-rules.md
 npx @buivietphi/skill-mobile-mt --init copilot    # .github/copilot-instructions.md
 npx @buivietphi/skill-mobile-mt --init windsurf   # .windsurfrules
-npx @buivietphi/skill-mobile-mt --init all        # All three files
+npx @buivietphi/skill-mobile-mt --init kilocode   # .kilocode/rules/mobile-rules.md
+npx @buivietphi/skill-mobile-mt --init kiro       # .kiro/steering/mobile-rules.md
+npx @buivietphi/skill-mobile-mt --init all        # All files
 ```
 
 **What `--init` does:**
@@ -84,16 +88,20 @@ You'll see an interactive checkbox — use arrow keys to navigate, space to sele
   ↑↓ navigate   Space toggle   A select all   Enter confirm   Q cancel
 
   › ◉ Claude Code   [detected]
+    ◯ Cline          [detected]
+    ◯ Roo Code       [not found]
     ◯ Cursor         [detected]
     ◯ Windsurf       [not found]
     ◯ Copilot        [not found]
     ◯ Codex          [not found]
     ◉ Gemini CLI     [detected]
     ◯ Kimi           [not found]
+    ◯ Kilo Code      [not found]
+    ◯ Kiro           [not found]
     ◯ Antigravity    [not found]
 ```
 
-### Step 2 (Cursor/Copilot/Windsurf only): Generate project rules
+### Step 2: Generate project rules
 
 ```bash
 # cd into your mobile project first
@@ -111,9 +119,13 @@ Output:
 
   ✓ .cursorrules → Cursor (auto-detected: Expo (React Native))
   ✓ .github/copilot-instructions.md → GitHub Copilot
+  ✓ .clinerules/mobile-rules.md → Cline
+  ✓ .roo/rules/mobile-rules.md → Roo Code
+  ✓ .kilocode/rules/mobile-rules.md → Kilo Code
+  ✓ .kiro/steering/mobile-rules.md → Kiro
   ✓ .windsurfrules → Windsurf
 
-  ✅ Done! → 3 file(s) generated
+  ✅ Done! → 7 file(s) generated
 ```
 
 ### Step 3: Use in your AI agent
@@ -132,27 +144,28 @@ Output:
 > Add cart feature following the same pattern as ProductList
 ```
 
-#### Cursor
+#### Cline / Roo Code / Kilo Code
 
-After running `--init cursor`, Cursor auto-loads `.cursorrules` every session. Just open your project and code — Cursor follows the rules automatically.
+After running `--init cline` (or `roocode`, `kilocode`), rules are auto-loaded from the project directory. Just open your project and ask:
 
 ```
-# In Cursor Composer, just ask:
 > Create a login screen with email/password
 > Fix the crash in ProductDetail when images is null
 > Review this PR for mobile best practices
 ```
 
+#### Cursor
+
+After running `--init cursor`, Cursor auto-loads `.cursorrules` every session. Just open your project and code normally.
+
+#### Kiro (AWS)
+
+After running `--init kiro`, Kiro loads `.kiro/steering/mobile-rules.md` as steering context. The generated file uses `inclusion: always` frontmatter.
+
 #### GitHub Copilot
 
 After running `--init copilot`, Copilot reads `.github/copilot-instructions.md` as workspace context.
 
-```
-# In VS Code with Copilot Chat:
-> Add pagination to the product list
-> Handle offline mode for the cart feature
-```
-
 #### Windsurf
 
 After running `--init windsurf`, Windsurf auto-loads `.windsurfrules`. Just code normally.
@@ -196,7 +209,7 @@ The skill automatically detects before any action:
 
 ## Smart Loading
 
-Only loads the relevant platform docs — saves **~60% context tokens**.
+Only loads the relevant platform docs — saves **~45% context tokens** vs loading everything.
 
 ```
 Flutter project?
@@ -216,22 +229,22 @@ iOS only?
 
 | Scenario | Tokens | % of 128K | % of 200K |
 |----------|-------:|----------:|----------:|
-| SKILL.md only | ~6,100 | 4.8% | 3.1% |
-| + 1 platform + core shared/ | ~14,200 | 11.1% | 7.1% |
-| Cross-platform (RN/Flutter + iOS + Android) | ~18,200 | 14.2% | 9.1% |
-| All files loaded | ~41,700 | 32.6% | 20.9% |
-| **Smart load (recommended)** | **~14,200** | **11.1%** | **7.1%** |
+| SKILL.md only | ~13,200 | 10.3% | 6.6% |
+| + 1 platform + core shared/ | ~38,600 | 30.2% | 19.3% |
+| Cross-platform (RN/Flutter + iOS + Android) | ~53,000 | 41.4% | 26.5% |
+| All files loaded | ~70,000 | 54.7% | 35.0% |
+| **Smart load (recommended)** | **~38,600** | **30.2%** | **19.3%** |
 
 ### Per-file token breakdown
 
 | File | Tokens |
 |------|-------:|
 | `SKILL.md` | 6,100 |
-| `AGENTS.md` | 1,541 |
+| `AGENTS.md` | 1,600 |
 | `react-native/react-native.md` | 5,108 |
-| `flutter/flutter.md` | 1,617 |
+| `flutter/flutter.md` | 2,100 |
 | `ios/ios-native.md` | 1,452 |
-| `android/android-native.md` | 1,659 |
+| `android/android-native.md` | 2,100 |
 | `shared/code-review.md` | 865 |
 | `shared/bug-detection.md` | 499 |
 | `shared/prompt-engineering.md` | 3,927 |
@@ -242,17 +255,19 @@ iOS only?
 | `shared/release-checklist.md` | 587 |
 | `shared/anti-patterns.md` | 2,800 |
 | `shared/performance-prediction.md` | 1,500 |
-| `shared/platform-excellence.md` | 1,500 |
+| `shared/platform-excellence.md` | 2,200 |
 | `shared/version-management.md` | 3,500 |
 | `shared/observability.md` | 3,000 |
+| `shared/offline-first.md` | 2,566 |
+| `shared/on-device-ai.md` | 700 |
 | `shared/claude-md-template.md` | ~500 |
 | `shared/agent-rules-template.md` | ~2,500 |
-| **Total** | **~46,000** |
+| **Total** | **~49,500** |
 
 ## Installed Structure
 
 ```
-~/.claude/skills/               (or ~/.gemini/skills/, ~/.agents/skills/, etc.)
+~/.claude/skills/               (or ~/.cline/skills/, ~/.roo/skills/, ~/.gemini/skills/, etc.)
 └── skill-mobile-mt/
     ├── SKILL.md                       Entry point + auto-detect + quality gates
     ├── AGENTS.md                      Multi-agent compatibility
@@ -274,10 +289,12 @@ iOS only?
         ├── document-analysis.md       Parse docs/images → code
         ├── anti-patterns.md           PII, cardinality, payload detection
         ├── performance-prediction.md  Frame budget, FPS prediction
-        ├── platform-excellence.md     iOS 18+ vs Android 15+ guidelines
+        ├── platform-excellence.md     iOS 18+ vs Android 15+ guidelines + HIG
         ├── version-management.md      SDK compatibility matrix
         ├── observability.md           Sessions as 4th pillar
         ├── release-checklist.md       Pre-release verification
+        ├── offline-first.md           Local-first + sync patterns
+        ├── on-device-ai.md            Core ML / TFLite / llama.cpp
         ├── claude-md-template.md      CLAUDE.md template for projects
         └── agent-rules-template.md    Rules templates for all agents
 ```
@@ -288,6 +305,17 @@ iOS only?
 your-project/
 ├── .cursorrules                       Cursor rules (auto-detected stack)
 ├── .windsurfrules                     Windsurf rules (auto-detected stack)
+├── .clinerules/
+│   └── mobile-rules.md               Cline rules (auto-detected stack)
+├── .roo/
+│   └── rules/
+│       └── mobile-rules.md           Roo Code rules (auto-detected stack)
+├── .kilocode/
+│   └── rules/
+│       └── mobile-rules.md           Kilo Code rules (auto-detected stack)
+├── .kiro/
+│   └── steering/
+│       └── mobile-rules.md           Kiro steering (auto-detected stack)
 └── .github/
     └── copilot-instructions.md        Copilot rules (auto-detected stack)
 ```
@@ -309,6 +337,27 @@ your-project/
 - Security vulnerabilities, performance issues, UX problems
 - Platform-specific detection rules
 
+### Grounding Protocol (Anti-Hallucination)
+- **Source hierarchy**: project code > skill files > official docs > production repos > AI knowledge
+- **Mandatory rules**: Read before answer, verify APIs exist, cite sources, say "I don't know" when unsure
+- **No phantom packages**: verify package exists in package.json before suggesting
+- **Version-specific**: check actual SDK version, never assume latest
+- **Grounded bug fix**: read file → find exact line → verify types → cite source
+- **Anti-hallucination checklist**: runs before every AI response
+
+### Docs-First Protocol (Always Use Latest)
+- **WebSearch before coding**: when setting up any library, ALWAYS search official docs first
+- **Never rely on AI memory**: docs change, APIs change, syntax changes between versions
+- **Package installation protocol**: verify exists → check compatibility → search install guide → test both platforms
+- **Common AI traps**: React Navigation v5 vs v7, Firebase v8 vs v9 modular, Swift async/await availability
+- **Version matrix + live verification**: snapshot matrix in skill + mandatory WebSearch for current data
+
+### Security Protocol
+- **7-category security scan** on every feature: secrets, token storage, input validation, network, data protection, auth flow, platform-specific
+- **6 absolute rules**: never store tokens in plain storage, never hardcode secrets, never log sensitive data, never trust deep links, never disable SSL, never commit .env
+- **Platform-specific checks**: iOS ATS + privacy manifest, Android cleartext + ProGuard + exported components
+- **Never skip security**: even for prototypes, internal apps, or "quick fixes"
+
 ### Advanced AI Patterns (from Windsurf, Cursor, Devin, Cline)
 - **Status Update Pattern**: Brief progress notes before tool use (no monologues)
 - **Execution Modes**: Discovery → Implementation → Completion (prevents thrashing)
@@ -340,7 +389,8 @@ your-project/
 
 - **Anti-Pattern Detection** (`anti-patterns.md`): Detect PII leaks (CRITICAL), high cardinality tags, unbounded payloads, unstructured logs, sync telemetry on main thread — with auto-fix suggestions
 - **Performance Prediction** (`performance-prediction.md`): Calculate frame budget, FlatList bridge calls, and memory usage BEFORE writing code. Example: `50 items × 3 bridge calls × 0.3ms = 45ms/frame → 22 FPS ❌ JANK`
-- **Platform Excellence** (`platform-excellence.md`): iOS 18+ vs Android 15+ native UX standards — navigation patterns, typography, icons, gesture handling, performance targets (cold start < 1s iOS, < 1.5s Android)
+- **Platform Excellence** (`platform-excellence.md`): iOS 18+ vs Android 15+ native UX standards — navigation patterns, typography, haptic feedback types, permission timing, ratings prompt flow, Live Activities/Dynamic Island, performance targets (cold start < 1s iOS, < 1.5s Android)
+- **On-Device AI** (`on-device-ai.md`): Decision matrix (API vs on-device), Core ML (iOS), ML Kit + MediaPipe (Android), llama.cpp cross-platform, TFLite Flutter, React Native ML Kit — with performance rules and model size guidance
 - **Version Management** (`version-management.md`): Full SDK compatibility matrix for RN 0.73-0.76, Expo 50-52, Flutter 3.22-3.27, iOS 16-18, Android 13-15. Check SDK compat BEFORE `npm install`. Release-mode testing protocol.
 - **Observability** (`observability.md`): Sessions as the 4th pillar (Metrics + Logs + Traces + **Sessions**). Session lifecycle, enrichment API, unified instrumentation stack, correlation queries. Every event carries `session_id` for full user journey reconstruction.
 
@@ -364,13 +414,77 @@ your-project/
 | Agent | How it works | Setup command |
 |-------|-------------|---------------|
 | **Claude Code** | Reads from `~/.claude/skills/` | `npx skill-mobile-mt --claude` |
+| **Cline** | Reads `.clinerules/` from project root | `npx skill-mobile-mt --init cline` |
+| **Roo Code** | Reads `.roo/rules/` from project root | `npx skill-mobile-mt --init roocode` |
+| **Cursor** | Reads `.cursorrules` from project root | `npx skill-mobile-mt --init cursor` |
+| **Windsurf** | Reads `.windsurfrules` from project root | `npx skill-mobile-mt --init windsurf` |
+| **Copilot** | Reads `.github/copilot-instructions.md` | `npx skill-mobile-mt --init copilot` |
 | **Codex** | Reads from `~/.codex/skills/` | `npx skill-mobile-mt --codex` |
 | **Gemini CLI** | Reads from `~/.gemini/skills/` | `npx skill-mobile-mt --gemini` |
 | **Kimi** | Reads from `~/.kimi/skills/` | `npx skill-mobile-mt --kimi` |
+| **Kilo Code** | Reads `.kilocode/rules/` from project root | `npx skill-mobile-mt --init kilocode` |
+| **Kiro** | Reads `.kiro/steering/` from project root | `npx skill-mobile-mt --init kiro` |
 | **Antigravity** | Reads from `~/.agents/skills/` | `npx skill-mobile-mt --antigravity` |
-| **Cursor** | Reads `.cursorrules` from project root | `npx skill-mobile-mt --init cursor` |
-| **Copilot** | Reads `.github/copilot-instructions.md` | `npx skill-mobile-mt --init copilot` |
-| **Windsurf** | Reads `.windsurfrules` from project root | `npx skill-mobile-mt --init windsurf` |
+
+## Humanizer — Mobile Copy
+
+Installed automatically alongside the main skill. Removes AI writing patterns from mobile app text.
+
+```
+@humanizer-mobile
+```
+
+Use for: app store descriptions, release notes, error messages, onboarding, push notifications, permission prompts, paywall copy.
+
+### Examples
+
+**App Store Description**
+```
+❌ "TaskFlow is a powerful, comprehensive task management application that
+seamlessly integrates with your existing workflow."
+
+✅ "TaskFlow keeps your team's work in one place. Add tasks, assign them,
+set deadlines — everything syncs in real time. Works with Slack and Notion."
+```
+
+**Release Notes**
+```
+❌ "Version 2.1.0 introduces significant enhancements to the overall user
+experience including robust improvements to performance and reliability."
+
+✅ "2.1.0
+- Faster load times on older Android devices (was 4s, now 1.2s)
+- Fixed crash when opening notifications while offline
+- Dark mode now remembers your setting between sessions"
+```
+
+**Error Messages**
+```
+❌ "We apologize for the inconvenience. An unexpected error has occurred."
+
+✅ "Couldn't save — no internet connection.
+   [Try again] [Save for later]"
+```
+
+**Push Notifications**
+```
+❌ "You have received a new message from a team member."
+✅ "Alex commented on "Landing page redesign""
+```
+
+**Paywall**
+```
+❌ "Unlock the full potential of our comprehensive premium features."
+
+✅ "Go Pro
+   • Unlimited projects (free plan: 3)
+   • Team sharing
+   [Start 7-day free trial]"
+```
+
+Covers 24 AI patterns across: content inflation, chatbot artifacts, style issues, mobile-specific copy (App Store limits, permission prompts, rating requests, subscription CTAs).
+
+---
 
 ## License