@su-record/vibe 2.10.2 → 2.12.1

package/skills/vibe.review/SKILL.md

@@ -16,7 +16,7 @@ user-invocable: true
 /vibe.review PR#123                  # Review specific PR
 /vibe.review feature/login           # Review specific branch
 /vibe.review src/api/                # Review specific path
-/vibe.review --race                  # Multi-LLM race mode (GPT + Gemini)
+/vibe.review --race                  # Multi-LLM race mode (GPT + Antigravity)
 /vibe.review --race security         # Race mode for specific review type
 ```
 
@@ -30,11 +30,11 @@ user-invocable: true
 > CODEX_AVAILABLE=$(node "{{VIBE_PATH}}/hooks/scripts/codex-detect.js" 2>/dev/null || echo "unavailable")
 > ```
 >
-> `available`이면 `/codex:review`, `/codex:rescue` 자동 호출. `unavailable`이면 기존 GPT+Gemini Race 모드로 동작.
+> `available`이면 `/codex:review`, `/codex:rescue` 자동 호출. `unavailable`이면 기존 GPT+Antigravity Race 모드로 동작.
 
 ## Race Mode (v2.6.9)
 
-**Multi-LLM competitive review** - Same review task runs on GPT + Gemini in parallel, results are cross-validated.
+**Multi-LLM competitive review** - Same review task runs on GPT + Antigravity in parallel, results are cross-validated.
 
 ### How It Works
 
@@ -43,7 +43,7 @@ user-invocable: true
 
 security-review:
 ├─ GPT Codex  → [SQL injection, XSS]
-└─ Gemini     → [SQL injection, CSRF]
+└─ Antigravity     → [SQL injection, CSRF]
          ↓
    Cross-validation:
    - SQL injection (2/2) → 🔴 P1 (100% confidence)
@@ -73,14 +73,14 @@ security-review:
 ## SECURITY Review (Race Mode)
 
 **Duration**: 3420ms
-**Models**: GPT Codex, Gemini
+**Models**: GPT Codex, Antigravity
 
 ### Model Results
 
 | Model | Issues Found | Duration | Status |
 |-------|--------------|----------|--------|
 | gpt | 3 | 1823ms | OK |
-| gemini | 2 | 2156ms | OK |
+| antigravity | 2 | 2156ms | OK |
 
 ### Cross-Validated Issues
 
@@ -89,7 +89,7 @@ security-review:
 
 #### 🔴 P1 - SQL Injection in user query
 
-- **Confidence**: 100% (gpt, gemini)
+- **Confidence**: 100% (gpt, antigravity)
 - **Severity**: critical
 - **Location**: `src/api/users.ts:42`
 - **Suggestion**: Use parameterized queries
@@ -103,7 +103,7 @@ security-review:
 
 ### Codex Review (Codex 플러그인 활성화 시)
 
-Race Mode에서 GPT+Gemini와 **동시에** Codex review 실행하여 3중 교차 검증:
+Race Mode에서 GPT+Antigravity와 **동시에** Codex review 실행하여 3중 교차 검증:
 
 ```
 /codex:review
@@ -112,7 +112,7 @@ Race Mode에서 GPT+Gemini와 **동시에** Codex review 실행하여 3중 교
 교차 검증 테이블:
 
 ```markdown
-| Issue | GPT | Gemini | Codex | Confidence |
+| Issue | GPT | Antigravity | Codex | Confidence |
 |-------|-----|--------|-------|------------|
 | {이슈} | ✅/❌ | ✅/❌ | ✅/❌ | {%} |
 ```
@@ -129,7 +129,7 @@ Race Mode에서 GPT+Gemini와 **동시에** Codex review 실행하여 3중 교
 | Quick iteration | ❌ Standard review |
 | API cost concerns | ❌ Standard review |
 
-### Tool Invocation (Race Mode - GPT + Gemini in parallel via Bash)
+### Tool Invocation (Race Mode - GPT + Antigravity in parallel via Bash)
 
 **🚨 Use --input file to avoid CLI argument length limits and Windows pipe issues.**
 
@@ -138,7 +138,7 @@ Race Mode에서 GPT+Gemini와 **동시에** Codex review 실행하여 3중 교
    - `{"prompt": "Review this code for [REVIEW_TYPE]. Return JSON: {issues: [{id, title, description, severity, suggestion}]}. Code: [CODE_CONTENT]"}`
    - Where `[CODE_CONTENT]` is the code text (properly JSON-escaped inside the prompt string)
 3. Script path: `[LLM_SCRIPT]` = `{{VIBE_PATH}}/hooks/scripts/llm-orchestrate.js`
-4. Run GPT + Gemini in PARALLEL (two Bash tool calls at once):
+4. Run GPT + Antigravity in PARALLEL (two Bash tool calls at once):
 
 ```bash
 # GPT review (Bash tool call 1)
@@ -146,8 +146,8 @@ node "[LLM_SCRIPT]" gpt orchestrate-json --input "[SCRATCHPAD]/review-input.json
 ```
 
 ```bash
-# Gemini review (Bash tool call 2 - run in parallel)
-node "[LLM_SCRIPT]" gemini orchestrate-json --input "[SCRATCHPAD]/review-input.json"
+# Antigravity review (Bash tool call 2 - run in parallel)
+node "[LLM_SCRIPT]" antigravity orchestrate-json --input "[SCRATCHPAD]/review-input.json"
 ```
 
 ## File Reading Policy (Mandatory)
@@ -284,6 +284,13 @@ Task(subagent_type="ui-antipattern-detector",
   prompt="Detect UI anti-patterns in: {changed_ui_files}. Check against MASTER.md if exists at .vibe/design-system/{project}/MASTER.md.")
 ```
 
+#### Visual P1 Baseline
+
+- 프로젝트 루트에 `DESIGN.md` 가 존재하면 **시각 P1 의 1 차 baseline** 으로 사용한다 (§2 Color Palette / §7 Do's & Don'ts).
+- `DESIGN.md` 부재 시 기존 폴백을 사용 (WCAG 2.1 AA + `MASTER.md` + design-audit 기본 5 차원).
+- v1 범위: hex 컬러 드리프트만 P1 후보. spacing / font 드리프트는 Phase 2+ 에서 추가.
+- 안티패턴 검출(⑧) 은 `DESIGN.md §7` 의 "DON'T" 항목을 우선 규칙으로 사용한다.
+
 **findings 통합**: ⑥⑦⑧ findings를 기존 findings[]와 병합 → P1/P2/P3 통합 정렬
 
 **⑦ Critical finding 에스컬레이션**: ui-a11y-auditor의 P1 finding은 Review Debate Team(Phase 4.5)에 자동 포함

package/skills/vibe.run/SKILL.md

@@ -69,6 +69,23 @@ Load skill `regress` with: list --feature "{feature-name}"
 
 Also load `.vibe/contracts/{feature-name}.md` if present — use it as the contract reference during implementation.
 
+### DESIGN.md Gate (UI stack only, before Phase 1)
+
+Run immediately after the contract load:
+
+```bash
+# UI stack 감지: .vibe/config.json 의 stacks[].type 중 STACK_TO_SKILLS 가 vibe.design 을 포함하는 것
+# DESIGN.md 존재: 프로젝트 루트
+test -f DESIGN.md
+```
+
+- **DESIGN.md present OR no UI stack** → silently continue
+- **DESIGN.md absent AND UI stack present**:
+  - interactive mode: 한 줄 안내 — "💡 UI 작업에 `DESIGN.md` 시각 SSOT 가 없습니다. `/vibe.design init` 으로 생성하면 시각 드리프트가 자동 검출됩니다. (생략 가능 — 1 회만 안내)"
+  - ultrawork mode: 무음 스킵 (gate 가 막지 않음)
+
+> **권유 > 강제**. DESIGN.md 부재는 절대 vibe.run 을 블록하지 않는다.
+
 ### Core Flow
 
 ```
@@ -223,9 +240,9 @@ When you include `ultrawork` (or `ulw`), ALL of these activate automatically:
 | **Boulder Loop** | Auto-continues until ALL phases complete |
 | **Context Compression** | Aggressive auto-save at 70%+ context |
 | **No Pause** | Doesn't wait for confirmation between phases |
-| **External LLMs** | Auto-consults GPT/Gemini if enabled |
+| **External LLMs** | Auto-consults GPT/Antigravity if enabled |
 | **Error Recovery** | Loops until 100% or stuck; on stuck auto-records TODO and proceeds (no user prompt) |
-| **Race Review (v2.6.9)** | Multi-LLM review (GPT+Gemini) with cross-validation |
+| **Race Review (v2.6.9)** | Multi-LLM review (GPT+Antigravity) with cross-validation |
 
 ### Boulder Loop (Inspired by Sisyphus)
 
@@ -765,12 +782,12 @@ When external LLMs are enabled in `.vibe/config.json`:
 
 | Role | Method | Condition |
 |------|--------|-----------|
-| User direct query | `gpt.question`, `gemini.question` | Hook auto-handles |
+| User direct query | `gpt.question`, `antigravity.question` | Hook auto-handles |
 | Internal orchestration | Call global script via Bash | Claude calls directly |
 
 **User questions (Hook auto-handles):**
 - `gpt.question` - GPT architecture consultation
-- `gemini.question` - Gemini Q&A/consultation
+- `antigravity.question` - Antigravity Q&A/consultation
 
 **Claude internal calls (directly via Bash):**
 ```bash
@@ -783,8 +800,8 @@ When external LLMs are enabled in `.vibe/config.json`:
 # GPT call (short prompt - CLI arg)
 node "[LLM_SCRIPT]" gpt orchestrate-json "[question content]"
 
-# Gemini call
-node "[LLM_SCRIPT]" gemini orchestrate-json "[question content]"
+# Antigravity call
+node "[LLM_SCRIPT]" antigravity orchestrate-json "[question content]"
 
 # Custom system prompt usage
 node "[LLM_SCRIPT]" gpt orchestrate-json "You are a code reviewer" "[question content]"
@@ -796,7 +813,7 @@ node "[LLM_SCRIPT]" gpt orchestrate-json --input "[SCRATCHPAD]/input.json"
 
 ### External LLM Fallback
 
-**IMPORTANT**: When GPT/Gemini hook fails, Claude MUST handle the task directly:
+**IMPORTANT**: When GPT/Antigravity hook fails, Claude MUST handle the task directly:
 
 **Fallback behavior**:
 - Do NOT retry the external LLM call
@@ -1140,7 +1157,7 @@ Then: Login success + JWT token returned
 │  Task(haiku) ─┴─→ "Find existing patterns and conventions"      │
 │                                                                 │
 │  [If GPT enabled] Bash: node "[LLM_SCRIPT]" gpt-codex orchestrate-json "[question]"
-│  [If Gemini enabled] Bash: node "[LLM_SCRIPT]" gemini orchestrate-json "[question]"
+│  [If Antigravity enabled] Bash: node "[LLM_SCRIPT]" antigravity orchestrate-json "[question]"
 └─────────────────────────────────────────────────────────────────┘
                               │
                               ↓ (wait for all to complete)
@@ -1423,10 +1440,10 @@ When starting a **new project** with brand context in SPEC, auto-generate app ic
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 
 [Check] Brand assets exist? → Skip if favicon.ico exists
-[Check] Gemini API configured? → Required for image generation
+[Check] Antigravity API configured? → Required for image generation
 [Check] SPEC has brand context? → Extract app name, colors, style
 
-[Generate] Creating app icon with Gemini Image API...
+[Generate] Creating app icon with Antigravity Image API...
   - Prompt: "App icon for [AppName], [style], [color]..."
   - Generated: 512x512 master icon
 
@@ -1459,34 +1476,34 @@ Brand:
 **Trigger Conditions:**
 - First `/vibe.run` execution (no existing icons)
 - SPEC contains brand/design context
-- Gemini API key configured (`vibe gemini key <key>`)
+- Antigravity API key configured (`vibe antigravity key <key>`)
 
 **Manual Generation:**
 ```bash
 # [LLM_SCRIPT] = {{VIBE_PATH}}/hooks/scripts/llm-orchestrate.js
-node "[LLM_SCRIPT]" gemini image "App icon for MyApp, primary color #2F6BFF, square format 1:1, simple recognizable design, works well at small sizes, no text or letters, solid or gradient background, modern minimalist" --output "./public/app-icon.png"
+node "[LLM_SCRIPT]" antigravity image "App icon for MyApp, primary color #2F6BFF, square format 1:1, simple recognizable design, works well at small sizes, no text or letters, solid or gradient background, modern minimalist" --output "./public/app-icon.png"
 ```
 
 ---
 
-### 5. Race Code Review (GPT + Gemini) + Auto-Fix (v2.6.9)
+### 5. Race Code Review (GPT + Antigravity) + Auto-Fix (v2.6.9)
 
-After all scenarios are implemented, **GPT and Gemini review in parallel with cross-validation**:
+After all scenarios are implemented, **GPT and Antigravity review in parallel with cross-validation**:
 
 > **ULTRAWORK Default**: In ULTRAWORK mode, race review is automatically enabled.
 
 ```
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
-🏁 RACE CODE REVIEW (GPT + Gemini)
+🏁 RACE CODE REVIEW (GPT + Antigravity)
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 
 [Step 1] Parallel review execution...
   ├─ GPT Codex: Reviewing...
-  └─ Gemini: Reviewing...
+  └─ Antigravity: Reviewing...
 
 [Step 2] Cross-validation results:
   ┌──────────────────────────────────────────────────────────────────┐
-  │ Issue                          │ GPT │ Gemini │ Codex │ Confidence│
+  │ Issue                          │ GPT │ Antigravity │ Codex │ Confidence│
   │────────────────────────────────│─────│────────│───────│───────────│
   │ Timing attack in password      │ ✅  │ ✅     │ ✅    │ 100% → P1 │
   │ Rate limiting missing          │ ✅  │ ✅     │ ✅    │ 100% → P1 │
@@ -1509,7 +1526,7 @@ After all scenarios are implemented, **GPT and Gemini review in parallel with cr
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 ```
 
-**Race Review Invocation (GPT + Gemini in parallel via Bash):**
+**Race Review Invocation (GPT + Antigravity in parallel via Bash):**
 
 **🚨 Use --input file to avoid CLI argument length limits and Windows pipe issues.**
 
@@ -1518,7 +1535,7 @@ After all scenarios are implemented, **GPT and Gemini review in parallel with cr
    - `{"prompt": "Review this code for security, performance, and best practices. Return JSON: {issues: [{id, title, description, severity, suggestion}]}. Code: [CODE_CONTENT]"}`
    - Where `[CODE_CONTENT]` is the code text (properly JSON-escaped inside the prompt string)
 3. Script path: `[LLM_SCRIPT]` = `{{VIBE_PATH}}/hooks/scripts/llm-orchestrate.js`
-4. Run GPT + Gemini in PARALLEL (two Bash tool calls at once):
+4. Run GPT + Antigravity in PARALLEL (two Bash tool calls at once):
 
 ```bash
 # GPT review (Bash tool call 1)
@@ -1526,8 +1543,8 @@ node "[LLM_SCRIPT]" gpt orchestrate-json --input "[SCRATCHPAD]/review-input.json
 ```
 
 ```bash
-# Gemini review (Bash tool call 2 - run in parallel)
-node "[LLM_SCRIPT]" gemini orchestrate-json --input "[SCRATCHPAD]/review-input.json"
+# Antigravity review (Bash tool call 2 - run in parallel)
+node "[LLM_SCRIPT]" antigravity orchestrate-json --input "[SCRATCHPAD]/review-input.json"
 ```
 
 **Confidence-based Priority:**
@@ -1558,7 +1575,7 @@ node "[LLM_SCRIPT]" gemini orchestrate-json --input "[SCRATCHPAD]/review-input.j
 
 ### Codex Code Review (Codex 플러그인 활성화 시)
 
-GPT+Gemini race와 **동시에** Codex review 실행:
+GPT+Antigravity race와 **동시에** Codex review 실행:
 
 ```
 /codex:review
@@ -1567,14 +1584,14 @@ GPT+Gemini race와 **동시에** Codex review 실행:
 결과를 race review 교차 검증에 포함 — 3중 리뷰:
 
 ```markdown
-| Issue | GPT | Gemini | Codex | Confidence |
+| Issue | GPT | Antigravity | Codex | Confidence |
 |-------|-----|--------|-------|------------|
 | {이슈} | ✅/❌ | ✅/❌ | ✅/❌ | {%} |
 ```
 
 ### 6. Quality Report (Auto-generated)
 
-After all scenarios complete + Gemini review, **quality report is auto-generated**:
+After all scenarios complete + Antigravity review, **quality report is auto-generated**:
 
 ```
 ┌─────────────────────────────────────────────────────────────────┐
@@ -1742,11 +1759,11 @@ Then: Login success + JWT token returned
 ✅ Scenario 4 passed!
 
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
-🔍 GEMINI CODE REVIEW
+🔍 ANTIGRAVITY CODE REVIEW
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 
-📤 Sending code to Gemini...
-📝 Gemini feedback:
+📤 Sending code to Antigravity...
+📝 Antigravity feedback:
   1. [Security] Need timing attack prevention → Fixing...
   2. [Performance] Unnecessary DB call → Fixing...
 
@@ -1767,13 +1784,13 @@ Then: Login success + JWT token returned
 │  | 4 | Password reset link   | ✅    | 0       |               │
 │                                                                 │
 │  📈 Quality score: 94/100                                       │
-│  Build: ✅ | Tests: ✅ | Types: ✅ | Gemini: ✅ (2 applied)     │
+│  Build: ✅ | Tests: ✅ | Types: ✅ | Antigravity: ✅ (2 applied)     │
 │                                                                 │
 │  ⏱️ Started: {start_time}                                        │
 │  ⏱️ Completed: {getCurrentTime 결과}                             │
 └─────────────────────────────────────────────────────────────────┘
 
-🎉 Implementation complete! All scenarios passed + Gemini review applied.
+🎉 Implementation complete! All scenarios passed + Antigravity review applied.
 
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 🔍 AUTO REVIEW (13+ Agents)
@@ -1867,7 +1884,7 @@ Grades:
 |------|---------------|-----------|
 | **Scenario Complete** | 95 | Each scenario must score ≥95 |
 | **Phase Complete** | 95 | Average of all scenarios ≥95 |
-| **Feature Complete** | 95 | All phases complete + Gemini review |
+| **Feature Complete** | 95 | All phases complete + Antigravity review |
 
 ### Auto-Fix Triggers
 

package/skills/vibe.spec/SKILL.md

@@ -62,13 +62,13 @@ Phase 2: Plan (skill: plan)
     ↓
 Phase 3: SPEC 작성 (skill: spec)
     - PTCF 구조 SPEC 문서 + Feature(BDD) 파일
-    - Parallel research (GPT/Gemini/Claude agents)
+    - Parallel research (GPT/Antigravity/Claude agents)
     - Large scope 자동 분할
     - Ambiguity scan + 품질 게이트(100점, 수렴까지 루프)
     - 출력: .vibe/specs/{feature}.md + .vibe/features/{feature}.feature
     ↓
 Phase 4: SPEC Review (skill: spec-review)
-    - Race Review (GPT + Gemini, 라운드 수 캡 없음, 수렴까지 루프)
+    - Race Review (GPT + Antigravity, 라운드 수 캡 없음, 수렴까지 루프)
     - (옵션) Codex adversarial review
     - 사용자 최종 체크포인트
     ↓
@@ -345,7 +345,7 @@ Load skill `spec` with input: .vibe/plans/{feature}.md
 
 1. Project analysis (explorer agent)
 2. config.json references 로드
-3. Parallel research (GPT + Gemini + Claude agents, 8개 병렬)
+3. Parallel research (GPT + Antigravity + Claude agents, 8개 병렬)
 4. UI/UX Design Intelligence (UI 키워드 시 자동)
 5. PTCF SPEC 작성 (Large scope 자동 분할)
 6. Feature file (BDD) 생성
@@ -369,7 +369,7 @@ Load skill `spec-review` with feature: {feature-name}
 
 1. SPEC/Feature 파일 로드 (single/split 자동 감지)
 2. Quality Validation (100점 게이트, 수렴까지 auto-fix 루프)
-3. Race Review (GPT + Gemini parallel, 라운드 수 캡 없음, P1=0 + 수렴 시 종료)
+3. Race Review (GPT + Antigravity parallel, 라운드 수 캡 없음, P1=0 + 수렴 시 종료)
 4. (옵션) Codex adversarial review
 5. Review Debate Team (2+ P1/P2 이슈 시)
 6. 사용자 최종 체크포인트

package/skills/vibe.utils/SKILL.md

@@ -19,7 +19,7 @@ Collection of utility tools. Use with options.
 /vibe.utils --e2e "scenario"         # E2E browser test (Playwright)
 /vibe.utils --e2e --visual           # Visual regression test
 /vibe.utils --e2e --record           # Video recording
-/vibe.utils --image "description"    # Generate image with Gemini (icon, banner, etc.)
+/vibe.utils --image "description"    # Generate image with Antigravity (icon, banner, etc.)
 /vibe.utils --image --icon "AppName" # Generate app icon/favicon
 /vibe.utils --build-fix              # Fix build errors (minimal diff)
 /vibe.utils --clean                  # Remove dead code + DELETION_LOG
@@ -35,8 +35,8 @@ Read and follow `agents/ui-previewer.md` for UI preview generation.
 
 Generate UI preview from description or design folder.
 
-- **Gemini enabled**: UI mockup image + component code generation (via `llm-orchestrate.js` → `gemini-api`)
-- **Gemini disabled**: ASCII art fallback
+- **Antigravity enabled**: UI mockup image + component code generation
+- **Antigravity disabled**: ASCII art fallback
 
 **Input types:**
 
@@ -86,10 +86,10 @@ Generate Mermaid diagrams for architecture visualization.
 
 ## --image (Image Generation)
 
-Generate images using Gemini Image API.
+Generate images using the Antigravity image backend.
 
-- **Default**: Gemini Flash Image (`gemini-2.5-flash-image`) - 빠르고 가벼운 이미지 생성
-- **--pro**: Gemini Pro Image (`gemini-3-pro-image-preview`) - 고품질 이미지 생성
+- **Default**: Antigravity fast image - 빠르고 가벼운 이미지 생성
+- **--pro**: Antigravity pro image - 고품질 이미지 생성
 
 ### MANDATORY Tool Invocation
 
@@ -98,19 +98,19 @@ Generate images using Gemini Image API.
 **Step 0: Script path:**
 - `[LLM_SCRIPT]` = `{{VIBE_PATH}}/hooks/scripts/llm-orchestrate.js`
 
-**General image generation (Gemini Flash Image):**
+**General image generation (Antigravity fast image):**
 ```bash
-node "[LLM_SCRIPT]" gemini image "IMAGE_DESCRIPTION" --output "OUTPUT_PATH"
+node "[LLM_SCRIPT]" antigravity image "IMAGE_DESCRIPTION" --output "OUTPUT_PATH"
 ```
 
-**Pro quality (Gemini Pro Image):**
+**Pro quality (Antigravity pro image):**
 ```bash
-node "[LLM_SCRIPT]" gemini image "IMAGE_DESCRIPTION" --pro --output "OUTPUT_PATH"
+node "[LLM_SCRIPT]" antigravity image "IMAGE_DESCRIPTION" --pro --output "OUTPUT_PATH"
 ```
 
 **With size option:**
 ```bash
-node "[LLM_SCRIPT]" gemini image "IMAGE_DESCRIPTION" --size "1920x1080" --output "OUTPUT_PATH"
+node "[LLM_SCRIPT]" antigravity image "IMAGE_DESCRIPTION" --size "1920x1080" --output "OUTPUT_PATH"
 ```
 
 ### How to Parse User Request
@@ -125,11 +125,11 @@ node "[LLM_SCRIPT]" gemini image "IMAGE_DESCRIPTION" --size "1920x1080" --output
 ### Examples
 
 ```bash
-node "[LLM_SCRIPT]" gemini image "A cute Gemini AI character mascot, colorful, friendly" --output "./gemini-character.png"
+node "[LLM_SCRIPT]" antigravity image "A friendly AI character mascot, colorful, approachable" --output "./ai-character.png"
 
-node "[LLM_SCRIPT]" gemini image "Professional website banner, modern design" --pro --size "1920x400" --output "./banner.png"
+node "[LLM_SCRIPT]" antigravity image "Professional website banner, modern design" --pro --size "1920x400" --output "./banner.png"
 
-node "[LLM_SCRIPT]" gemini image "Modern minimal logo design" --output "./public/logo.png"
+node "[LLM_SCRIPT]" antigravity image "Modern minimal logo design" --output "./public/logo.png"
 ```
 
 ### Output Format
@@ -152,15 +152,15 @@ Use the same command with a pre-built icon prompt template:
 
 ```bash
 # --icon "MyApp"
-node "[LLM_SCRIPT]" gemini image "App icon for MyApp, square format 1:1, simple recognizable design, works well at small sizes, no text or letters, solid or gradient background, modern minimalist" --output "./public/app-icon.png"
+node "[LLM_SCRIPT]" antigravity image "App icon for MyApp, square format 1:1, simple recognizable design, works well at small sizes, no text or letters, solid or gradient background, modern minimalist" --output "./public/app-icon.png"
 
 # --icon "MyApp" --color "#2F6BFF"
-node "[LLM_SCRIPT]" gemini image "App icon for MyApp, primary color #2F6BFF, square format 1:1, simple recognizable design, works well at small sizes, no text or letters, solid or gradient background, modern minimalist" --output "./public/app-icon.png"
+node "[LLM_SCRIPT]" antigravity image "App icon for MyApp, primary color #2F6BFF, square format 1:1, simple recognizable design, works well at small sizes, no text or letters, solid or gradient background, modern minimalist" --output "./public/app-icon.png"
 ```
 
 ### Prerequisites
 
-- Gemini API key configured (`vibe gemini key <key>`)
+- Antigravity API key configured (`vibe antigravity key <key>`)
 
 ---
 

package/skills/vibe.verify/SKILL.md

@@ -184,6 +184,26 @@ For each UI scenario in Feature file:
 - DOM 기반: `div class="nav-wrapper mx-4 flex..."` = 200+ chars
 - 전자를 사용해야 시나리오 50개도 한 세션에서 검증 가능
 
+### 3.2 Visual Drift Detection (--from-design-md)
+
+**활성 조건**: 프로젝트 루트에 `DESIGN.md` 존재 + UI 스택. 또는 `--from-design-md` 플래그 명시.
+
+**동작**:
+
+```
+Load skill `vibe.design` with: verify --files=<changed-ui-files>
+```
+
+`vibe.design verify` 는 `DESIGN.md §2 Color Palette` 의 hex 토큰 셋을 추출(`extractHexTokens`)하고,
+변경 파일에서 토큰 밖 hex 하드코딩을 검출(`findHardcodedColors`) 한다.
+
+**판정**:
+- P1 drift (토큰 밖 hex) → `/vibe.verify` 실패
+- DESIGN.md 부재 → 안내 후 silent skip (블록하지 않음)
+- v1 범위: hex only. spacing / font 드리프트는 Phase 2+
+
+**성능 목표**: <1s / 100 파일 (변경 파일만 스캔 — `/vibe.verify` 기존 정책 일치).
+
 ### 3.5 Step Count (MANDATORY before Quality Report)
 
 Read `.vibe/metrics/current-run.json` to obtain the tool-call count from the preceding `/vibe.run` (and any follow-up work in this session). Then append a record to `.vibe/metrics/history.jsonl` and include the count in the Quality Report below.

package/hooks/gemini-hooks.json

@@ -1,73 +0,0 @@
-{
-  "hooks": {
-    "SessionStart": [
-      {
-        "command": "VIBE_CLI=gemini node {{VIBE_PATH}}/hooks/scripts/session-start.js"
-      }
-    ],
-    "BeforeTool": [
-      {
-        "tool_name": "shell",
-        "command": "VIBE_CLI=gemini node {{VIBE_PATH}}/hooks/scripts/sentinel-guard.js Bash"
-      },
-      {
-        "tool_name": "shell",
-        "command": "VIBE_CLI=gemini node {{VIBE_PATH}}/hooks/scripts/pre-tool-guard.js Bash"
-      },
-      {
-        "tool_name": "edit",
-        "command": "VIBE_CLI=gemini node {{VIBE_PATH}}/hooks/scripts/sentinel-guard.js Edit"
-      },
-      {
-        "tool_name": "edit",
-        "command": "VIBE_CLI=gemini node {{VIBE_PATH}}/hooks/scripts/pre-tool-guard.js Edit"
-      },
-      {
-        "tool_name": "write_file",
-        "command": "VIBE_CLI=gemini node {{VIBE_PATH}}/hooks/scripts/sentinel-guard.js Write"
-      },
-      {
-        "tool_name": "write_file",
-        "command": "VIBE_CLI=gemini node {{VIBE_PATH}}/hooks/scripts/pre-tool-guard.js Write"
-      }
-    ],
-    "AfterTool": [
-      {
-        "tool_name": "edit",
-        "command": "VIBE_CLI=gemini node {{VIBE_PATH}}/hooks/scripts/code-check.js"
-      },
-      {
-        "tool_name": "write_file",
-        "command": "VIBE_CLI=gemini node {{VIBE_PATH}}/hooks/scripts/code-check.js"
-      },
-      {
-        "tool_name": "edit",
-        "command": "VIBE_CLI=gemini node {{VIBE_PATH}}/hooks/scripts/post-edit.js"
-      }
-    ],
-    "BeforeAgent": [
-      {
-        "command": "VIBE_CLI=gemini node {{VIBE_PATH}}/hooks/scripts/prompt-dispatcher.js"
-      }
-    ],
-    "Notification": [
-      {
-        "matcher": "context_window_80",
-        "command": "VIBE_CLI=gemini node {{VIBE_PATH}}/hooks/scripts/context-save.js medium"
-      },
-      {
-        "matcher": "context_window_90",
-        "command": "VIBE_CLI=gemini node {{VIBE_PATH}}/hooks/scripts/context-save.js high"
-      },
-      {
-        "matcher": "context_window_95",
-        "command": "VIBE_CLI=gemini node {{VIBE_PATH}}/hooks/scripts/context-save.js critical"
-      }
-    ],
-    "SessionEnd": [
-      {
-        "command": "VIBE_CLI=gemini node {{VIBE_PATH}}/hooks/scripts/stop-notify.js"
-      }
-    ]
-  }
-}