@su-record/vibe 1.1.9 → 1.1.10
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/.claude/agents/explorer.md +21 -21
- package/.claude/agents/implementer.md +24 -24
- package/.claude/agents/searcher.md +26 -26
- package/.claude/agents/simplifier.md +52 -53
- package/.claude/agents/tester.md +17 -17
- package/.claude/commands/vibe.analyze.md +138 -138
- package/.claude/commands/vibe.diagram.md +48 -48
- package/.claude/commands/vibe.reason.md +174 -174
- package/.claude/commands/vibe.run.md +169 -169
- package/.claude/commands/vibe.spec.md +160 -160
- package/.claude/commands/vibe.ui.md +38 -38
- package/.claude/commands/vibe.verify.md +77 -77
- package/package.json +1 -1
- package/templates/hooks-template.json +5 -5
|
@@ -5,301 +5,301 @@ argument-hint: "feature name" or --phase N
|
|
|
5
5
|
|
|
6
6
|
# /vibe.run
|
|
7
7
|
|
|
8
|
-
|
|
8
|
+
Execute implementation based on SPEC (Implementation Agent with Multi-Model Orchestration).
|
|
9
9
|
|
|
10
10
|
## Usage
|
|
11
11
|
|
|
12
12
|
```
|
|
13
|
-
/vibe.run "
|
|
14
|
-
/vibe.run "
|
|
13
|
+
/vibe.run "feature-name" # Full implementation
|
|
14
|
+
/vibe.run "feature-name" --phase 1 # Specific Phase only
|
|
15
15
|
```
|
|
16
16
|
|
|
17
17
|
## Rules Reference
|
|
18
18
|
|
|
19
|
-
|
|
20
|
-
- `core/development-philosophy.md` -
|
|
21
|
-
- `core/quick-start.md` -
|
|
22
|
-
- `standards/complexity-metrics.md` -
|
|
23
|
-
- `quality/checklist.md` -
|
|
19
|
+
**Must follow `.vibe/rules/`:**
|
|
20
|
+
- `core/development-philosophy.md` - Surgical precision, modify only requested scope
|
|
21
|
+
- `core/quick-start.md` - Korean, DRY, SRP, YAGNI
|
|
22
|
+
- `standards/complexity-metrics.md` - Functions ≤20 lines, nesting ≤3 levels
|
|
23
|
+
- `quality/checklist.md` - Code quality checklist
|
|
24
24
|
|
|
25
25
|
## Description
|
|
26
26
|
|
|
27
|
-
PTCF
|
|
27
|
+
Read PTCF structured SPEC document and execute implementation immediately.
|
|
28
28
|
|
|
29
|
-
> **PLAN, TASKS
|
|
29
|
+
> **PLAN, TASKS documents unnecessary** - SPEC is the executable prompt
|
|
30
30
|
|
|
31
|
-
##
|
|
31
|
+
## Model Orchestration
|
|
32
32
|
|
|
33
|
-
|
|
33
|
+
Automatically select optimal model based on task type:
|
|
34
34
|
|
|
35
35
|
```
|
|
36
36
|
┌─────────────────────────────────────────────────────────────┐
|
|
37
|
-
│ Opus 4.5 (
|
|
38
|
-
│ -
|
|
39
|
-
│ -
|
|
37
|
+
│ Opus 4.5 (Orchestrator) │
|
|
38
|
+
│ - Coordinate overall flow │
|
|
39
|
+
│ - Final decisions/review │
|
|
40
40
|
└─────────────────────────┬───────────────────────────────────┘
|
|
41
41
|
│
|
|
42
42
|
┌─────────────────────┼─────────────────────┐
|
|
43
43
|
↓ ↓ ↓
|
|
44
44
|
┌─────────┐ ┌─────────┐ ┌─────────┐
|
|
45
45
|
│ Haiku │ │ Sonnet │ │ Haiku │
|
|
46
|
-
│
|
|
46
|
+
│(Explore)│ │ (Impl) │ │ (Test) │
|
|
47
47
|
└─────────┘ └─────────┘ └─────────┘
|
|
48
48
|
```
|
|
49
49
|
|
|
50
|
-
###
|
|
50
|
+
### Task Calls by Role
|
|
51
51
|
|
|
52
|
-
|
|
|
53
|
-
|
|
54
|
-
|
|
|
55
|
-
|
|
|
56
|
-
|
|
|
57
|
-
|
|
|
58
|
-
|
|
|
52
|
+
| Task Type | Model | Task Parameter |
|
|
53
|
+
|-----------|-------|----------------|
|
|
54
|
+
| Codebase exploration | Haiku 4.5 | `model: "haiku"` |
|
|
55
|
+
| Core implementation | Sonnet 4 | `model: "sonnet"` |
|
|
56
|
+
| Test writing | Haiku 4.5 | `model: "haiku"` |
|
|
57
|
+
| Architecture decisions | Opus 4.5 | Main session |
|
|
58
|
+
| Final review | Opus 4.5 | Main session |
|
|
59
59
|
|
|
60
|
-
###
|
|
60
|
+
### External LLM Usage (When Enabled)
|
|
61
61
|
|
|
62
|
-
`.vibe/config.json
|
|
62
|
+
When external LLMs are enabled in `.vibe/config.json`:
|
|
63
63
|
|
|
64
|
-
|
|
|
65
|
-
|
|
66
|
-
|
|
|
67
|
-
| UI/UX
|
|
64
|
+
| Role | Model | Condition |
|
|
65
|
+
|------|-------|-----------|
|
|
66
|
+
| Architecture/Debugging | GPT 5.2 | When `vibe gpt <key>` executed |
|
|
67
|
+
| UI/UX Design | Gemini 3 | When `vibe gemini <key>` executed |
|
|
68
68
|
|
|
69
|
-
|
|
70
|
-
- `mcp__vibe-gpt__chat` - GPT 5.2
|
|
71
|
-
- `mcp__vibe-gemini__chat` - Gemini 3 UI/UX
|
|
69
|
+
When external LLM enabled, automatically called via MCP:
|
|
70
|
+
- `mcp__vibe-gpt__chat` - GPT 5.2 architecture consultation
|
|
71
|
+
- `mcp__vibe-gemini__chat` - Gemini 3 UI/UX consultation
|
|
72
72
|
|
|
73
|
-
##
|
|
73
|
+
## Semantic Code Analysis (hi-ai MCP)
|
|
74
74
|
|
|
75
|
-
|
|
75
|
+
Use hi-ai MCP semantic tools to accurately understand codebase before implementation:
|
|
76
76
|
|
|
77
|
-
| MCP
|
|
78
|
-
|
|
79
|
-
| `mcp__vibe__find_symbol` |
|
|
80
|
-
| `mcp__vibe__find_references` |
|
|
81
|
-
| `mcp__vibe__analyze_complexity` |
|
|
82
|
-
| `mcp__vibe__validate_code_quality` |
|
|
77
|
+
| MCP Tool | Purpose | When Used |
|
|
78
|
+
|----------|---------|-----------|
|
|
79
|
+
| `mcp__vibe__find_symbol` | Find symbol definitions | Locate class/function |
|
|
80
|
+
| `mcp__vibe__find_references` | Find references | Analyze impact scope |
|
|
81
|
+
| `mcp__vibe__analyze_complexity` | Analyze complexity | Determine refactoring need |
|
|
82
|
+
| `mcp__vibe__validate_code_quality` | Validate quality | Verify post-implementation quality |
|
|
83
83
|
|
|
84
|
-
###
|
|
84
|
+
### Semantic Analysis Flow
|
|
85
85
|
|
|
86
86
|
```
|
|
87
|
-
|
|
87
|
+
Start Implementation
|
|
88
88
|
│
|
|
89
|
-
├─→ find_symbol:
|
|
89
|
+
├─→ find_symbol: Locate exact position of function/class to modify
|
|
90
90
|
│
|
|
91
|
-
├─→ find_references:
|
|
91
|
+
├─→ find_references: Check all places using that symbol
|
|
92
92
|
│
|
|
93
|
-
├─→ analyze_complexity:
|
|
93
|
+
├─→ analyze_complexity: Check existing code complexity
|
|
94
94
|
│
|
|
95
95
|
↓
|
|
96
|
-
|
|
96
|
+
Implementation (with accurate understanding of impact scope)
|
|
97
97
|
│
|
|
98
98
|
↓
|
|
99
|
-
validate_code_quality:
|
|
99
|
+
validate_code_quality: Verify quality after implementation
|
|
100
100
|
```
|
|
101
101
|
|
|
102
|
-
###
|
|
102
|
+
### Context Management (Session Continuity)
|
|
103
103
|
|
|
104
|
-
| MCP
|
|
105
|
-
|
|
106
|
-
| `mcp__vibe__start_session` |
|
|
107
|
-
| `mcp__vibe__auto_save_context` |
|
|
108
|
-
| `mcp__vibe__restore_session_context` |
|
|
109
|
-
| `mcp__vibe__save_memory` |
|
|
104
|
+
| MCP Tool | Purpose |
|
|
105
|
+
|----------|---------|
|
|
106
|
+
| `mcp__vibe__start_session` | Start session, restore previous context |
|
|
107
|
+
| `mcp__vibe__auto_save_context` | Auto-save current state |
|
|
108
|
+
| `mcp__vibe__restore_session_context` | Restore previous session context |
|
|
109
|
+
| `mcp__vibe__save_memory` | Save important decisions/patterns |
|
|
110
110
|
|
|
111
|
-
|
|
112
|
-
|
|
111
|
+
**On session start**: `mcp__vibe__start_session` auto-restores previous context
|
|
112
|
+
**On session end**: Hook auto-executes `mcp__vibe__auto_save_context`
|
|
113
113
|
|
|
114
114
|
## Process
|
|
115
115
|
|
|
116
|
-
### 1. SPEC
|
|
116
|
+
### 1. Read SPEC and Config
|
|
117
117
|
|
|
118
|
-
`.vibe/specs/{
|
|
118
|
+
Parse `.vibe/specs/{feature-name}.md`:
|
|
119
119
|
|
|
120
|
-
|
|
|
121
|
-
|
|
122
|
-
| `<role>` | AI
|
|
123
|
-
| `<context>` |
|
|
124
|
-
| `<task>` | Phase
|
|
125
|
-
| `<constraints>` |
|
|
126
|
-
| `<output_format>` |
|
|
127
|
-
| `<acceptance>` |
|
|
120
|
+
| Section | Purpose |
|
|
121
|
+
|---------|---------|
|
|
122
|
+
| `<role>` | AI role definition |
|
|
123
|
+
| `<context>` | Background, tech stack, related code |
|
|
124
|
+
| `<task>` | Phase-by-phase task list |
|
|
125
|
+
| `<constraints>` | Constraints |
|
|
126
|
+
| `<output_format>` | Files to create/modify |
|
|
127
|
+
| `<acceptance>` | Verification criteria |
|
|
128
128
|
|
|
129
|
-
`.vibe/config.json
|
|
130
|
-
-
|
|
129
|
+
Check `.vibe/config.json`:
|
|
130
|
+
- External LLM enablement (`models.gpt.enabled`, `models.gemini.enabled`)
|
|
131
131
|
|
|
132
|
-
### 2. Feature
|
|
132
|
+
### 2. Check Feature File
|
|
133
133
|
|
|
134
|
-
`.vibe/features/{
|
|
135
|
-
- BDD Scenarios
|
|
136
|
-
-
|
|
134
|
+
`.vibe/features/{feature-name}.feature`:
|
|
135
|
+
- Check BDD Scenarios
|
|
136
|
+
- Use as test cases
|
|
137
137
|
|
|
138
|
-
### 3. Phase
|
|
138
|
+
### 3. Phase-by-Phase Implementation (Parallel Task Calls)
|
|
139
139
|
|
|
140
|
-
`<task>`
|
|
140
|
+
Follow `<task>` section Phase order:
|
|
141
141
|
|
|
142
142
|
```
|
|
143
|
-
Phase
|
|
143
|
+
Phase Start
|
|
144
144
|
│
|
|
145
|
-
├─→ Task(haiku):
|
|
146
|
-
│ "
|
|
145
|
+
├─→ Task(haiku): Codebase analysis
|
|
146
|
+
│ "Analyze related files and patterns"
|
|
147
147
|
│
|
|
148
|
-
├─→ [GPT
|
|
149
|
-
│ "
|
|
148
|
+
├─→ [GPT enabled] MCP(vibe-gpt): Architecture review
|
|
149
|
+
│ "Review if this design is appropriate"
|
|
150
150
|
│
|
|
151
|
-
├─→ [Gemini
|
|
152
|
-
│ "UI
|
|
151
|
+
├─→ [Gemini enabled] MCP(vibe-gemini): UI/UX consultation
|
|
152
|
+
│ "Suggest UI implementation direction"
|
|
153
153
|
│
|
|
154
154
|
↓
|
|
155
|
-
Opus:
|
|
155
|
+
Opus: Synthesize analysis results, decide implementation direction
|
|
156
156
|
│
|
|
157
157
|
↓
|
|
158
|
-
Task(sonnet):
|
|
159
|
-
"
|
|
158
|
+
Task(sonnet): Core implementation
|
|
159
|
+
"Implement code according to SPEC"
|
|
160
160
|
│
|
|
161
161
|
↓
|
|
162
|
-
Task(haiku):
|
|
163
|
-
"
|
|
162
|
+
Task(haiku): Write tests
|
|
163
|
+
"Write tests for implemented code"
|
|
164
164
|
│
|
|
165
165
|
↓
|
|
166
|
-
Opus:
|
|
166
|
+
Opus: Final review and next Phase
|
|
167
167
|
```
|
|
168
168
|
|
|
169
|
-
|
|
169
|
+
**Parallel execution example:**
|
|
170
170
|
```javascript
|
|
171
|
-
//
|
|
172
|
-
Task(haiku) -
|
|
173
|
-
Task(haiku) -
|
|
174
|
-
// →
|
|
175
|
-
|
|
176
|
-
//
|
|
177
|
-
Task(sonnet) -
|
|
178
|
-
Task(haiku) -
|
|
171
|
+
// Independent tasks run in parallel
|
|
172
|
+
Task(haiku) - Code analysis
|
|
173
|
+
Task(haiku) - Dependency check
|
|
174
|
+
// → Run simultaneously
|
|
175
|
+
|
|
176
|
+
// Sequential tasks
|
|
177
|
+
Task(sonnet) - Implementation (after analysis complete)
|
|
178
|
+
Task(haiku) - Tests (after implementation complete)
|
|
179
179
|
```
|
|
180
180
|
|
|
181
|
-
1.
|
|
182
|
-
2.
|
|
183
|
-
3.
|
|
184
|
-
4.
|
|
181
|
+
1. **Related code analysis**: Task(haiku) explores `<context>` related code
|
|
182
|
+
2. **File creation/modification**: Task(sonnet) implements per `<output_format>`
|
|
183
|
+
3. **Constraint compliance**: Check `<constraints>`
|
|
184
|
+
4. **Run verification**: Execute verification commands
|
|
185
185
|
|
|
186
|
-
### 4. Acceptance Criteria
|
|
186
|
+
### 4. Acceptance Criteria Verification
|
|
187
187
|
|
|
188
|
-
`<acceptance>`
|
|
189
|
-
- [ ]
|
|
190
|
-
- [ ]
|
|
191
|
-
- [ ]
|
|
188
|
+
Check `<acceptance>` checklist:
|
|
189
|
+
- [ ] Each criterion pass status
|
|
190
|
+
- [ ] Tests pass
|
|
191
|
+
- [ ] Build succeeds
|
|
192
192
|
|
|
193
|
-
### 5. SPEC
|
|
193
|
+
### 5. Update SPEC
|
|
194
194
|
|
|
195
|
-
|
|
195
|
+
Check completed Tasks:
|
|
196
196
|
```markdown
|
|
197
197
|
## Task
|
|
198
198
|
<task>
|
|
199
199
|
### Phase 1: Backend
|
|
200
|
-
1. [x] DB
|
|
201
|
-
2. [x] API
|
|
200
|
+
1. [x] Write DB schema ✅
|
|
201
|
+
2. [x] API endpoint ✅
|
|
202
202
|
...
|
|
203
203
|
</task>
|
|
204
204
|
```
|
|
205
205
|
|
|
206
|
-
## TRUST 5
|
|
206
|
+
## TRUST 5 Principles
|
|
207
207
|
|
|
208
|
-
|
|
208
|
+
Follow during implementation:
|
|
209
209
|
|
|
210
|
-
|
|
|
211
|
-
|
|
212
|
-
| **T**est-first |
|
|
213
|
-
| **R**eadable |
|
|
214
|
-
| **U**nified |
|
|
215
|
-
| **S**ecured |
|
|
216
|
-
| **T**rackable |
|
|
210
|
+
| Principle | Description |
|
|
211
|
+
|-----------|-------------|
|
|
212
|
+
| **T**est-first | Write tests first |
|
|
213
|
+
| **R**eadable | Clear code |
|
|
214
|
+
| **U**nified | Consistent style |
|
|
215
|
+
| **S**ecured | Consider security |
|
|
216
|
+
| **T**rackable | Logging, monitoring |
|
|
217
217
|
|
|
218
218
|
## Input
|
|
219
219
|
|
|
220
|
-
- `.vibe/specs/{
|
|
221
|
-
- `.vibe/features/{
|
|
222
|
-
- `CLAUDE.md` (
|
|
220
|
+
- `.vibe/specs/{feature-name}.md` (PTCF SPEC)
|
|
221
|
+
- `.vibe/features/{feature-name}.feature` (BDD)
|
|
222
|
+
- `CLAUDE.md` (project context)
|
|
223
223
|
|
|
224
224
|
## Output
|
|
225
225
|
|
|
226
|
-
-
|
|
227
|
-
-
|
|
228
|
-
- SPEC
|
|
226
|
+
- Implemented code files
|
|
227
|
+
- Test files
|
|
228
|
+
- Updated SPEC document (checkmarks)
|
|
229
229
|
|
|
230
230
|
## Example
|
|
231
231
|
|
|
232
232
|
```
|
|
233
|
-
User: /vibe.run "
|
|
233
|
+
User: /vibe.run "brick-game"
|
|
234
234
|
|
|
235
235
|
Claude:
|
|
236
|
-
📄 SPEC
|
|
236
|
+
📄 Reading SPEC: .vibe/specs/brick-game.md
|
|
237
237
|
|
|
238
|
-
<role>
|
|
239
|
-
-
|
|
240
|
-
- Phaser.js
|
|
238
|
+
<role> analysis:
|
|
239
|
+
- Web game senior developer
|
|
240
|
+
- Phaser.js expert
|
|
241
241
|
|
|
242
|
-
<context>
|
|
243
|
-
-
|
|
244
|
-
-
|
|
242
|
+
<context> analysis:
|
|
243
|
+
- Tech stack: Phaser.js, TypeScript
|
|
244
|
+
- New project
|
|
245
245
|
|
|
246
|
-
<task>
|
|
247
|
-
- Phase 1:
|
|
248
|
-
- Phase 2:
|
|
249
|
-
- Phase 3: UI/UX (3
|
|
250
|
-
- Phase 4:
|
|
246
|
+
<task> analysis:
|
|
247
|
+
- Phase 1: Project setup (3 tasks)
|
|
248
|
+
- Phase 2: Game logic (5 tasks)
|
|
249
|
+
- Phase 3: UI/UX (3 tasks)
|
|
250
|
+
- Phase 4: Testing (2 tasks)
|
|
251
251
|
|
|
252
|
-
🚀 Phase 1
|
|
252
|
+
🚀 Starting Phase 1...
|
|
253
253
|
|
|
254
|
-
✅ Phase 1
|
|
255
|
-
- package.json
|
|
256
|
-
- TypeScript
|
|
257
|
-
- Phaser.js
|
|
254
|
+
✅ Phase 1 complete
|
|
255
|
+
- package.json created
|
|
256
|
+
- TypeScript configured
|
|
257
|
+
- Phaser.js installed
|
|
258
258
|
|
|
259
|
-
🚀 Phase 2
|
|
260
|
-
[
|
|
259
|
+
🚀 Starting Phase 2...
|
|
260
|
+
[Implementation continues...]
|
|
261
261
|
|
|
262
|
-
✅
|
|
263
|
-
📊 Acceptance Criteria
|
|
264
|
-
✅
|
|
265
|
-
✅
|
|
266
|
-
✅
|
|
267
|
-
✅ npm run build
|
|
262
|
+
✅ All Phases complete!
|
|
263
|
+
📊 Verifying Acceptance Criteria...
|
|
264
|
+
✅ Game start/end works
|
|
265
|
+
✅ Ball-paddle collision handling
|
|
266
|
+
✅ Score display
|
|
267
|
+
✅ npm run build succeeds
|
|
268
268
|
|
|
269
|
-
🎉
|
|
269
|
+
🎉 Implementation complete!
|
|
270
270
|
```
|
|
271
271
|
|
|
272
|
-
### Phase
|
|
272
|
+
### Phase-specific Execution
|
|
273
273
|
|
|
274
274
|
```
|
|
275
|
-
User: /vibe.run "
|
|
275
|
+
User: /vibe.run "brick-game" --phase 2
|
|
276
276
|
|
|
277
277
|
Claude:
|
|
278
|
-
📄 SPEC
|
|
279
|
-
🎯 Phase 2
|
|
278
|
+
📄 Reading SPEC: .vibe/specs/brick-game.md
|
|
279
|
+
🎯 Executing Phase 2 only.
|
|
280
280
|
|
|
281
|
-
Phase 2:
|
|
282
|
-
1. [ ]
|
|
283
|
-
2. [ ]
|
|
284
|
-
3. [ ]
|
|
285
|
-
4. [ ]
|
|
286
|
-
5. [ ]
|
|
281
|
+
Phase 2: Game Logic
|
|
282
|
+
1. [ ] Paddle movement implementation
|
|
283
|
+
2. [ ] Ball physics engine
|
|
284
|
+
3. [ ] Brick collision handling
|
|
285
|
+
4. [ ] Score system
|
|
286
|
+
5. [ ] Game over conditions
|
|
287
287
|
|
|
288
|
-
🚀
|
|
288
|
+
🚀 Starting implementation...
|
|
289
289
|
```
|
|
290
290
|
|
|
291
291
|
## Error Handling
|
|
292
292
|
|
|
293
|
-
|
|
294
|
-
1.
|
|
295
|
-
2. `<constraints>`
|
|
296
|
-
3.
|
|
297
|
-
4.
|
|
293
|
+
On failure:
|
|
294
|
+
1. Check error message
|
|
295
|
+
2. Review `<constraints>`
|
|
296
|
+
3. Fix code and retry
|
|
297
|
+
4. If continues to fail, report to user
|
|
298
298
|
|
|
299
299
|
## Next Step
|
|
300
300
|
|
|
301
301
|
```
|
|
302
|
-
/vibe.verify "
|
|
302
|
+
/vibe.verify "brick-game"
|
|
303
303
|
```
|
|
304
304
|
|
|
305
305
|
---
|