@dusky-bluehour/agent-service 0.6.2
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/LICENSE +21 -0
- package/README.md +205 -0
- package/antigravity/README.md +37 -0
- package/antigravity/agents/agent-catalog.json +72 -0
- package/antigravity/artifacts/artifact-catalog.json +184 -0
- package/antigravity/commands/command-catalog.json +942 -0
- package/antigravity/skills/code-review-and-improvement/SKILL.md +15 -0
- package/antigravity/skills/frontend-repetition-pack/SKILL.md +15 -0
- package/antigravity/skills/incident-response/SKILL.md +15 -0
- package/antigravity/skills/prd-to-production-pipeline/SKILL.md +16 -0
- package/antigravity/skills/release-and-operations/SKILL.md +15 -0
- package/antigravity/skills/security-hardening/SKILL.md +15 -0
- package/antigravity/skills/service-lifecycle-orchestration/SKILL.md +16 -0
- package/antigravity/workflows/workflow-catalog.json +362 -0
- package/catalog/tool-catalog.ko.json +296 -0
- package/claude-code/README.md +47 -0
- package/claude-code/agent-teams/team-catalog.json +69 -0
- package/claude-code/commands/command-catalog.json +942 -0
- package/claude-code/skills/code-review-and-improvement/SKILL.md +16 -0
- package/claude-code/skills/frontend-repetition-pack/SKILL.md +16 -0
- package/claude-code/skills/incident-response/SKILL.md +16 -0
- package/claude-code/skills/prd-to-production-pipeline/SKILL.md +17 -0
- package/claude-code/skills/release-and-operations/SKILL.md +16 -0
- package/claude-code/skills/security-hardening/SKILL.md +15 -0
- package/claude-code/skills/service-lifecycle-orchestration/SKILL.md +17 -0
- package/claude-code/subagents/backend-engineer.md +20 -0
- package/claude-code/subagents/code-reviewer.md +19 -0
- package/claude-code/subagents/frontend-engineer.md +20 -0
- package/claude-code/subagents/hook-refactor-engineer.md +19 -0
- package/claude-code/subagents/incident-commander.md +19 -0
- package/claude-code/subagents/lead-orchestrator.md +18 -0
- package/claude-code/subagents/operations-owner.md +20 -0
- package/claude-code/subagents/performance-engineer.md +19 -0
- package/claude-code/subagents/prd-writer.md +20 -0
- package/claude-code/subagents/product-planner.md +19 -0
- package/claude-code/subagents/qa-engineer.md +19 -0
- package/claude-code/subagents/security-engineer.md +20 -0
- package/claude-code/subagents/solution-architect.md +19 -0
- package/claude-code/subagents/sre-release-engineer.md +20 -0
- package/claude-code/subagents/ui-component-engineer.md +19 -0
- package/claude-code/workflows/workflow-catalog.json +680 -0
- package/codex/README.md +38 -0
- package/codex/automations/automation-recipes.toml +30 -0
- package/codex/commands/command-catalog.json +942 -0
- package/codex/instructions/AGENTS.override.template.md +21 -0
- package/codex/instructions/AGENTS.template.md +31 -0
- package/codex/skills/code-review-and-improvement/SKILL.md +16 -0
- package/codex/skills/code-review-and-improvement/agents/openai.yaml +4 -0
- package/codex/skills/frontend-repetition-pack/SKILL.md +15 -0
- package/codex/skills/frontend-repetition-pack/agents/openai.yaml +4 -0
- package/codex/skills/incident-response/SKILL.md +16 -0
- package/codex/skills/incident-response/agents/openai.yaml +4 -0
- package/codex/skills/prd-to-production-pipeline/SKILL.md +16 -0
- package/codex/skills/prd-to-production-pipeline/agents/openai.yaml +4 -0
- package/codex/skills/release-and-operations/SKILL.md +15 -0
- package/codex/skills/release-and-operations/agents/openai.yaml +4 -0
- package/codex/skills/security-hardening/SKILL.md +15 -0
- package/codex/skills/security-hardening/agents/openai.yaml +4 -0
- package/codex/skills/service-lifecycle-orchestration/SKILL.md +17 -0
- package/codex/skills/service-lifecycle-orchestration/agents/openai.yaml +4 -0
- package/codex/workflows/workflow-catalog.json +444 -0
- package/package.json +44 -0
- package/scripts/init.mjs +993 -0
- package/scripts/validate.mjs +591 -0
package/LICENSE
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
1
|
+
MIT License
|
|
2
|
+
|
|
3
|
+
Copyright (c) 2026
|
|
4
|
+
|
|
5
|
+
Permission is hereby granted, free of charge, to any person obtaining a copy
|
|
6
|
+
of this software and associated documentation files (the "Software"), to deal
|
|
7
|
+
in the Software without restriction, including without limitation the rights
|
|
8
|
+
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
|
9
|
+
copies of the Software, and to permit persons to whom the Software is
|
|
10
|
+
furnished to do so, subject to the following conditions:
|
|
11
|
+
|
|
12
|
+
The above copyright notice and this permission notice shall be included in all
|
|
13
|
+
copies or substantial portions of the Software.
|
|
14
|
+
|
|
15
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
|
16
|
+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
17
|
+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
|
18
|
+
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
|
19
|
+
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
|
20
|
+
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
|
|
21
|
+
SOFTWARE.
|
package/README.md
ADDED
|
@@ -0,0 +1,205 @@
|
|
|
1
|
+
# tri-agent-manager
|
|
2
|
+
|
|
3
|
+
Claude Code, Antigravity, Codex에서 공통으로 사용하실 수 있는 서비스 운영형 스킬/워크플로우 패키지입니다.
|
|
4
|
+
|
|
5
|
+
## 목적
|
|
6
|
+
|
|
7
|
+
- 기획(PRD) → 설계 → 개발 → 코드리뷰/보안 → 배포/운영까지 동일한 품질 기준으로 실행하실 수 있습니다.
|
|
8
|
+
- 반복 작업(UI 컴포넌트화, Hook 분리, 최적화)을 역할/명령 단위로 고정해 일관성을 높입니다.
|
|
9
|
+
- 툴이 달라도 동일한 계약(명령 ID, 품질 게이트, 워크플로우)으로 결과를 맞출 수 있습니다.
|
|
10
|
+
|
|
11
|
+
## 용어 설명
|
|
12
|
+
|
|
13
|
+
| 용어 | 의미 |
|
|
14
|
+
| --- | --- |
|
|
15
|
+
| Skill | 작업 방식/규칙을 담은 실행 단위입니다. 툴마다 포맷은 다르지만 역할은 같습니다. |
|
|
16
|
+
| Workflow | 단계형 실행 시나리오입니다. 각 단계는 진입/종료 조건을 가집니다. |
|
|
17
|
+
| Command Contract | 입력/실행/품질게이트/산출물을 고정한 명령 계약입니다. |
|
|
18
|
+
| Agent / Subagent | 역할별 실행 주체입니다. 리드가 전문 역할에 위임할 수 있습니다. |
|
|
19
|
+
| Artifact (Antigravity) | 단계 입출력 계약 문서입니다. 스키마 검증 통과 후 다음 단계로 진행합니다. |
|
|
20
|
+
| PRD→구현 파이프라인 | PRD는 대화형으로 작성하고, 구현은 최종 수동 세팅 전까지 순차 자동 실행합니다. |
|
|
21
|
+
|
|
22
|
+
## 핵심 기능
|
|
23
|
+
|
|
24
|
+
| 기능 | 설명 | 관련 워크플로우 |
|
|
25
|
+
| --- | --- | --- |
|
|
26
|
+
| 표준 E2E 딜리버리 | 기획부터 배포 준비까지 표준 단계를 실행합니다. | `WF-SVC-001` |
|
|
27
|
+
| 코드리뷰/개선 루프 | 리뷰 지적 → 수정 → 재검증 루프를 수행합니다. | `WF-SVC-003` / `WF-SVC-002` |
|
|
28
|
+
| 보안 하드닝 | 위협 모델링 + 취약점 감사 + 재검증을 수행합니다. | `WF-SVC-004` / `WF-SVC-003` |
|
|
29
|
+
| 반복 프론트 개선 | 컴포넌트화, Hook 분리, 성능 최적화를 진행합니다. | `WF-SVC-006` / `WF-SVC-002` |
|
|
30
|
+
| 장애 대응 | 복구, RCA, 재발 방지를 수행합니다. | `WF-SVC-007` / `WF-SVC-005` |
|
|
31
|
+
| PRD→구현 자동 파이프라인 | PRD 대화형 작성 후 구현을 자동 순차 실행합니다. | `WF-PRD-IMPLEMENT-E2E` |
|
|
32
|
+
|
|
33
|
+
## 목록 (토글)
|
|
34
|
+
|
|
35
|
+
<details>
|
|
36
|
+
<summary><strong>Claude Code - Skills</strong></summary>
|
|
37
|
+
|
|
38
|
+
| Skill | 목적 |
|
|
39
|
+
| --- | --- |
|
|
40
|
+
| `service-lifecycle-orchestration` | 기획~배포 전체 수명주기를 실행합니다. |
|
|
41
|
+
| `code-review-and-improvement` | 리뷰/개선 루프를 실행합니다. |
|
|
42
|
+
| `security-hardening` | 위협/취약점 기반 보안 강화를 수행합니다. |
|
|
43
|
+
| `frontend-repetition-pack` | UI 컴포넌트화/Hook/최적화를 수행합니다. |
|
|
44
|
+
| `release-and-operations` | 배포/모니터링/인수인계를 수행합니다. |
|
|
45
|
+
| `incident-response` | 장애 대응~재발 방지를 수행합니다. |
|
|
46
|
+
| `prd-to-production-pipeline` | PRD 대화형 작성 후 순차 자동 구현을 수행합니다. |
|
|
47
|
+
|
|
48
|
+
</details>
|
|
49
|
+
|
|
50
|
+
<details>
|
|
51
|
+
<summary><strong>Claude Code - Subagents</strong></summary>
|
|
52
|
+
|
|
53
|
+
| Subagent | 역할 |
|
|
54
|
+
| --- | --- |
|
|
55
|
+
| `lead-orchestrator` | 전체 단계를 오케스트레이션합니다. |
|
|
56
|
+
| `prd-writer` | 마스터/세부 PRD를 작성합니다. |
|
|
57
|
+
| `product-planner` | 요구사항 잠금/범위 관리를 수행합니다. |
|
|
58
|
+
| `solution-architect` | ADR/기술 경계를 정의합니다. |
|
|
59
|
+
| `backend-engineer` | 계약 우선 API 구현을 수행합니다. |
|
|
60
|
+
| `frontend-engineer` | 프론트 기능 구현을 수행합니다. |
|
|
61
|
+
| `ui-component-engineer` | 반복 UI를 컴포넌트화합니다. |
|
|
62
|
+
| `hook-refactor-engineer` | Hook 분리/정규화를 수행합니다. |
|
|
63
|
+
| `performance-engineer` | 성능 병목을 개선합니다. |
|
|
64
|
+
| `qa-engineer` | 테스트/회귀 차단을 수행합니다. |
|
|
65
|
+
| `code-reviewer` | 코드리뷰/개선안을 도출합니다. |
|
|
66
|
+
| `security-engineer` | 위협/취약점 대응을 수행합니다. |
|
|
67
|
+
| `sre-release-engineer` | CI/CD 게이트/배포를 수행합니다. |
|
|
68
|
+
| `operations-owner` | 운영 대시보드/런북을 관리합니다. |
|
|
69
|
+
| `incident-commander` | 장애 워룸/복구 우선순위를 관리합니다. |
|
|
70
|
+
|
|
71
|
+
</details>
|
|
72
|
+
|
|
73
|
+
<details>
|
|
74
|
+
<summary><strong>Antigravity - Skills / Agents</strong></summary>
|
|
75
|
+
|
|
76
|
+
| 타입 | ID | 설명 |
|
|
77
|
+
| --- | --- | --- |
|
|
78
|
+
| Skill | `service-lifecycle-orchestration` | Plan-Implement-Test-Review-Release를 실행합니다. |
|
|
79
|
+
| Skill | `code-review-and-improvement` | 리뷰/개선 루프를 실행합니다. |
|
|
80
|
+
| Skill | `security-hardening` | 보안 하드닝을 수행합니다. |
|
|
81
|
+
| Skill | `frontend-repetition-pack` | 프론트 반복 개선을 수행합니다. |
|
|
82
|
+
| Skill | `release-and-operations` | 배포/운영 전환을 수행합니다. |
|
|
83
|
+
| Skill | `incident-response` | 장애 대응을 수행합니다. |
|
|
84
|
+
| Skill | `prd-to-production-pipeline` | PRD 대화형→자동 구현을 수행합니다. |
|
|
85
|
+
| Agent | `manager-agent` | 기본 오케스트레이터 역할을 수행합니다. |
|
|
86
|
+
| Agent | `prd-writer-agent` | PRD 대화형 작성을 수행합니다. |
|
|
87
|
+
| Agent | `implementation-agent` | 구현 자동 실행을 수행합니다. |
|
|
88
|
+
| Agent | `review-security-agent` | 리뷰/보안 검증을 수행합니다. |
|
|
89
|
+
| Agent | `release-ops-agent` | 최종 세팅/배포를 수행합니다. |
|
|
90
|
+
|
|
91
|
+
</details>
|
|
92
|
+
|
|
93
|
+
<details>
|
|
94
|
+
<summary><strong>Codex - Skills</strong></summary>
|
|
95
|
+
|
|
96
|
+
| Skill | 목적 |
|
|
97
|
+
| --- | --- |
|
|
98
|
+
| `service-lifecycle-orchestration` | 기획~배포 표준 실행을 수행합니다. |
|
|
99
|
+
| `code-review-and-improvement` | 리뷰/개선 루프를 수행합니다. |
|
|
100
|
+
| `security-hardening` | 보안 강화를 수행합니다. |
|
|
101
|
+
| `frontend-repetition-pack` | UI/Hook/성능 개선을 수행합니다. |
|
|
102
|
+
| `release-and-operations` | 릴리즈/운영 전환을 수행합니다. |
|
|
103
|
+
| `incident-response` | 장애 대응을 수행합니다. |
|
|
104
|
+
| `prd-to-production-pipeline` | PRD 대화형→순차 자동 구현을 수행합니다. |
|
|
105
|
+
|
|
106
|
+
</details>
|
|
107
|
+
|
|
108
|
+
## PRD→구현 자동 파이프라인 (신규)
|
|
109
|
+
|
|
110
|
+
- 워크플로우: `WF-PRD-IMPLEMENT-E2E`
|
|
111
|
+
- 단계:
|
|
112
|
+
1. 마스터 PRD 작성 (`CMD-PLAN-PRD-MASTER`) - 대화형
|
|
113
|
+
2. 세부 PRD 분해 (`CMD-PLAN-PRD-DETAILS`) - 대화형
|
|
114
|
+
3. 구현 부트스트랩 (`CMD-PLAN-IMPLEMENTATION-BOOTSTRAP`) - 대화형
|
|
115
|
+
4. 순차 자동 구현 (`CMD-DEV-SEQUENTIAL-AUTORUN`) - 자동
|
|
116
|
+
5. 검증/보안 - 자동
|
|
117
|
+
6. 최종 세팅/배포 준비 - 수동
|
|
118
|
+
|
|
119
|
+
PRD 단계에서는 UI 컴포넌트화, 톤앤매너, 성능, 보안, 가독성 기준을 명시하시고, 자동 단계에서는 배치별 검증을 통과해야 다음 배치로 진행됩니다.
|
|
120
|
+
|
|
121
|
+
## 설치/업데이트 사용법
|
|
122
|
+
|
|
123
|
+
### 가장 쉬운 시작 (pnpm dlx, 전부 대화형)
|
|
124
|
+
|
|
125
|
+
```bash
|
|
126
|
+
pnpm dlx tri-agent-manager --interactive
|
|
127
|
+
```
|
|
128
|
+
|
|
129
|
+
위 한 줄로 아래를 터미널에서 순서대로 선택하실 수 있습니다.
|
|
130
|
+
|
|
131
|
+
1. 설치 경로
|
|
132
|
+
2. 프리셋
|
|
133
|
+
3. 도구(Claude Code/Antigravity/Codex)
|
|
134
|
+
4. 구성요소(스킬, 워크플로우 등)
|
|
135
|
+
5. 실행 확인
|
|
136
|
+
|
|
137
|
+
조작 키:
|
|
138
|
+
|
|
139
|
+
- `↑/↓`: 이동
|
|
140
|
+
- `←/→`: 해제/선택 또는 항목 이동
|
|
141
|
+
- `Space`: 다중 선택 토글
|
|
142
|
+
- `Enter`: 확정
|
|
143
|
+
|
|
144
|
+
업데이트도 동일하게 대화형으로 진행하실 수 있습니다.
|
|
145
|
+
|
|
146
|
+
```bash
|
|
147
|
+
pnpm dlx tri-agent-manager update --interactive
|
|
148
|
+
```
|
|
149
|
+
|
|
150
|
+
`setup` 별칭도 동일하게 동작합니다.
|
|
151
|
+
|
|
152
|
+
```bash
|
|
153
|
+
pnpm dlx tri-agent-manager setup
|
|
154
|
+
```
|
|
155
|
+
|
|
156
|
+
### npm 사용 시
|
|
157
|
+
|
|
158
|
+
```bash
|
|
159
|
+
npx tri-agent-manager --interactive
|
|
160
|
+
npx tri-agent-manager update --interactive
|
|
161
|
+
```
|
|
162
|
+
|
|
163
|
+
이전 CLI 별칭 `tri-agent-os`도 호환용으로 함께 제공됩니다.
|
|
164
|
+
|
|
165
|
+
### 고급 옵션(필요한 경우)
|
|
166
|
+
|
|
167
|
+
- `--preset`: 미리 정의된 구성을 선택합니다.
|
|
168
|
+
- `--tool`: 툴을 직접 선택합니다. (`codex,claude-code` 등)
|
|
169
|
+
- `--components`: 구성요소를 직접 선택합니다. (`skills,workflows,commands` 등)
|
|
170
|
+
- `--interactive`: 한글 대화형 선택 UI를 사용합니다.
|
|
171
|
+
- `--dry-run`: 실제 반영 없이 계획만 확인합니다.
|
|
172
|
+
- `--yes`: 확인 프롬프트를 생략합니다.
|
|
173
|
+
|
|
174
|
+
중복 선택 입력(예: `codex,codex,claude-code`)은 자동으로 1회로 정리됩니다.
|
|
175
|
+
|
|
176
|
+
## 프리셋
|
|
177
|
+
|
|
178
|
+
- `balanced-core`: 기본 운영팩(추천)
|
|
179
|
+
- `full-suite`: 전체 설치
|
|
180
|
+
- `security-first`: 보안 우선
|
|
181
|
+
- `frontend-refactor`: 프론트 반복개선
|
|
182
|
+
- `prd-to-production`: PRD→구현 자동 파이프라인
|
|
183
|
+
|
|
184
|
+
## 체크리스트 자동화
|
|
185
|
+
|
|
186
|
+
배포 전 자동 점검:
|
|
187
|
+
|
|
188
|
+
```bash
|
|
189
|
+
npm run prepublish:check
|
|
190
|
+
```
|
|
191
|
+
|
|
192
|
+
`prepublishOnly`가 설정되어 있어 `npm publish` 시에도 자동 실행됩니다.
|
|
193
|
+
|
|
194
|
+
## 다음 단계 추천
|
|
195
|
+
|
|
196
|
+
1. `pnpm dlx tri-agent-manager --interactive`로 `prd-to-production` 프리셋을 먼저 설치해 보세요.
|
|
197
|
+
2. 첫 실행은 `--dry-run`으로 계획만 확인하신 뒤 실제 반영을 진행해 주세요.
|
|
198
|
+
3. 반영 후 `npm run prepublish:check`로 검증을 고정하신 뒤 배포해 주세요.
|
|
199
|
+
|
|
200
|
+
## 상세 구성 문서
|
|
201
|
+
|
|
202
|
+
- 구조/설계/계약 상세: `docs/COMPOSITION.ko.md`
|
|
203
|
+
- 대화형 UX 흐름 그림: `docs/UX-FLOW.ko.md`
|
|
204
|
+
- 배포 절차 가이드: `docs/DEPLOYMENT-GUIDE.ko.md`
|
|
205
|
+
- 업데이트 절차 가이드: `docs/UPDATE-GUIDE.ko.md`
|
|
@@ -0,0 +1,37 @@
|
|
|
1
|
+
# Antigravity 운영팩
|
|
2
|
+
|
|
3
|
+
이 폴더는 Antigravity의 Manager-에이전트-Artifact 흐름을 기준으로, 서비스 운영 전체를 실행 가능한 형태로 정리한 패키지입니다.
|
|
4
|
+
|
|
5
|
+
## 폴더 구성
|
|
6
|
+
|
|
7
|
+
- `skills/`: 역할별 실행 스킬
|
|
8
|
+
- `agents/agent-catalog.json`: 에이전트 역할 카탈로그
|
|
9
|
+
- `artifacts/artifact-catalog.json`: 단계별 Artifact 계약
|
|
10
|
+
- `workflows/workflow-catalog.json`: 오케스트레이션 워크플로우
|
|
11
|
+
- `commands/command-catalog.json`: 공통 명령 계약
|
|
12
|
+
|
|
13
|
+
## 빠른 시작
|
|
14
|
+
|
|
15
|
+
1. `WF-SVC-001`(기본 E2E), `WF-SVC-004`(보안), `WF-PRD-IMPLEMENT-E2E`(PRD→구현) 등 목적별 워크플로우를 선택해 주세요.
|
|
16
|
+
2. 각 단계의 `input_artifact` 필드를 먼저 검증해 주세요.
|
|
17
|
+
3. 지정된 `agent`가 `commands`를 수행합니다.
|
|
18
|
+
4. `output_artifact` 검증 후 다음 단계로 진행해 주세요.
|
|
19
|
+
|
|
20
|
+
## 반복 작업 역할 분리
|
|
21
|
+
|
|
22
|
+
- UI 컴포넌트화: `frontend-repetition-agent` + `CMD-DEV-FE-UI-COMPONENTIZE`
|
|
23
|
+
- Hook 분리: `frontend-repetition-agent` + `CMD-DEV-FE-HOOK-SEPARATE`
|
|
24
|
+
- 성능 최적화: `performance-agent` + `CMD-DEV-PERF-OPTIMIZE`
|
|
25
|
+
- PRD 작성: `prd-writer-agent`
|
|
26
|
+
|
|
27
|
+
## 운영 규칙
|
|
28
|
+
|
|
29
|
+
- Manager 에이전트가 단계 전환 권한을 가집니다.
|
|
30
|
+
- Artifact 검증 실패 시 단계 중단 후 원인 수정을 우선해 주세요.
|
|
31
|
+
- 보안/배포 단계는 우회 없이 게이트를 통과해야 합니다.
|
|
32
|
+
|
|
33
|
+
## 참고 문서 (검증일: 2026-02-17)
|
|
34
|
+
|
|
35
|
+
- [Google Antigravity 발표](https://developers.googleblog.com/build-with-google-antigravity-our-new-agentic-development-platform/)
|
|
36
|
+
- [Antigravity 공식 문서](https://antigravity.dev/docs)
|
|
37
|
+
- [Antigravity 워크플로우 문서](https://antigravity.im/documentation)
|
|
@@ -0,0 +1,72 @@
|
|
|
1
|
+
{
|
|
2
|
+
"schema_version": "1.0.0",
|
|
3
|
+
"agents": [
|
|
4
|
+
{
|
|
5
|
+
"id": "prd-writer-agent",
|
|
6
|
+
"role": "PRD 작성",
|
|
7
|
+
"responsibility": "마스터 PRD와 세부 PRD를 대화형으로 작성하고 구현 수용 기준을 명시",
|
|
8
|
+
"default_workflows": [
|
|
9
|
+
"WF-PRD-IMPLEMENT-E2E"
|
|
10
|
+
]
|
|
11
|
+
},
|
|
12
|
+
{
|
|
13
|
+
"id": "manager-agent",
|
|
14
|
+
"role": "오케스트레이터",
|
|
15
|
+
"responsibility": "워크플로우 단계 제어, 에이전트 체인 구성, Artifact 검증 게이트 실행",
|
|
16
|
+
"default_workflows": [
|
|
17
|
+
"WF-SVC-001",
|
|
18
|
+
"WF-SVC-002",
|
|
19
|
+
"WF-SVC-003"
|
|
20
|
+
]
|
|
21
|
+
},
|
|
22
|
+
{
|
|
23
|
+
"id": "incident-manager-agent",
|
|
24
|
+
"role": "인시던트 오케스트레이터",
|
|
25
|
+
"responsibility": "장애 대응 워룸 흐름 제어",
|
|
26
|
+
"default_workflows": [
|
|
27
|
+
"WF-SVC-004"
|
|
28
|
+
]
|
|
29
|
+
},
|
|
30
|
+
{
|
|
31
|
+
"id": "product-planner-agent",
|
|
32
|
+
"role": "기획",
|
|
33
|
+
"responsibility": "요구사항 잠금, 범위 제어, 승인 기준 명시"
|
|
34
|
+
},
|
|
35
|
+
{
|
|
36
|
+
"id": "implementation-agent",
|
|
37
|
+
"role": "개발",
|
|
38
|
+
"responsibility": "API 구현, UI 컴포넌트화, Hook 분리, 성능 최적화"
|
|
39
|
+
},
|
|
40
|
+
{
|
|
41
|
+
"id": "frontend-repetition-agent",
|
|
42
|
+
"role": "프론트 반복개선",
|
|
43
|
+
"responsibility": "UI 컴포넌트화와 Hook 분리 전담"
|
|
44
|
+
},
|
|
45
|
+
{
|
|
46
|
+
"id": "performance-agent",
|
|
47
|
+
"role": "성능",
|
|
48
|
+
"responsibility": "측정 기반 병목 제거와 전후 성능 검증"
|
|
49
|
+
},
|
|
50
|
+
{
|
|
51
|
+
"id": "qa-agent",
|
|
52
|
+
"role": "검증",
|
|
53
|
+
"responsibility": "단위/통합 테스트 고정과 회귀 차단"
|
|
54
|
+
},
|
|
55
|
+
{
|
|
56
|
+
"id": "review-security-agent",
|
|
57
|
+
"role": "리뷰/보안",
|
|
58
|
+
"responsibility": "코드 리뷰, 위협 모델링, 의존성 감사"
|
|
59
|
+
},
|
|
60
|
+
{
|
|
61
|
+
"id": "release-ops-agent",
|
|
62
|
+
"role": "배포/운영",
|
|
63
|
+
"responsibility": "CI/CD 게이트, 점진 배포, 모니터링, 인수인계"
|
|
64
|
+
}
|
|
65
|
+
],
|
|
66
|
+
"routing_rules": [
|
|
67
|
+
"PRD 기반 구현 요청은 prd-writer-agent로 시작",
|
|
68
|
+
"반복 프론트 개선은 frontend-repetition-agent 우선",
|
|
69
|
+
"보안 이슈는 review-security-agent를 반드시 포함",
|
|
70
|
+
"배포 단계는 release-ops-agent 단독 승인"
|
|
71
|
+
]
|
|
72
|
+
}
|
|
@@ -0,0 +1,184 @@
|
|
|
1
|
+
{
|
|
2
|
+
"schema_version": "1.0.0",
|
|
3
|
+
"artifacts": [
|
|
4
|
+
{
|
|
5
|
+
"id": "ART-REQUEST-BRIEF",
|
|
6
|
+
"required_fields": [
|
|
7
|
+
"문제정의",
|
|
8
|
+
"목표지표",
|
|
9
|
+
"제약사항"
|
|
10
|
+
]
|
|
11
|
+
},
|
|
12
|
+
{
|
|
13
|
+
"id": "ART-SCOPE-ADR",
|
|
14
|
+
"required_fields": [
|
|
15
|
+
"범위",
|
|
16
|
+
"비범위",
|
|
17
|
+
"ADR",
|
|
18
|
+
"롤백전략"
|
|
19
|
+
]
|
|
20
|
+
},
|
|
21
|
+
{
|
|
22
|
+
"id": "ART-IMPLEMENTATION-PACK",
|
|
23
|
+
"required_fields": [
|
|
24
|
+
"변경목록",
|
|
25
|
+
"테스트초안",
|
|
26
|
+
"리스크메모"
|
|
27
|
+
]
|
|
28
|
+
},
|
|
29
|
+
{
|
|
30
|
+
"id": "ART-TEST-REPORT",
|
|
31
|
+
"required_fields": [
|
|
32
|
+
"실행결과",
|
|
33
|
+
"실패원인",
|
|
34
|
+
"재현절차"
|
|
35
|
+
]
|
|
36
|
+
},
|
|
37
|
+
{
|
|
38
|
+
"id": "ART-QUALITY-SECURITY-REPORT",
|
|
39
|
+
"required_fields": [
|
|
40
|
+
"차단이슈",
|
|
41
|
+
"권고이슈",
|
|
42
|
+
"취약점결과"
|
|
43
|
+
]
|
|
44
|
+
},
|
|
45
|
+
{
|
|
46
|
+
"id": "ART-RELEASE-READY",
|
|
47
|
+
"required_fields": [
|
|
48
|
+
"게이트결과",
|
|
49
|
+
"배포계획",
|
|
50
|
+
"모니터링계획"
|
|
51
|
+
]
|
|
52
|
+
},
|
|
53
|
+
{
|
|
54
|
+
"id": "ART-OPERATIONS-HANDBOOK",
|
|
55
|
+
"required_fields": [
|
|
56
|
+
"런북",
|
|
57
|
+
"알림정책",
|
|
58
|
+
"인수인계체크리스트"
|
|
59
|
+
]
|
|
60
|
+
},
|
|
61
|
+
{
|
|
62
|
+
"id": "ART-FRONTEND-BASELINE",
|
|
63
|
+
"required_fields": [
|
|
64
|
+
"중복UI목록",
|
|
65
|
+
"렌더링지표",
|
|
66
|
+
"리팩터대상"
|
|
67
|
+
]
|
|
68
|
+
},
|
|
69
|
+
{
|
|
70
|
+
"id": "ART-COMPONENTIZED-UI",
|
|
71
|
+
"required_fields": [
|
|
72
|
+
"컴포넌트목록",
|
|
73
|
+
"Props계약",
|
|
74
|
+
"스토리"
|
|
75
|
+
]
|
|
76
|
+
},
|
|
77
|
+
{
|
|
78
|
+
"id": "ART-HOOK-REFACTORED",
|
|
79
|
+
"required_fields": [
|
|
80
|
+
"Hook목록",
|
|
81
|
+
"입출력계약",
|
|
82
|
+
"테스트"
|
|
83
|
+
]
|
|
84
|
+
},
|
|
85
|
+
{
|
|
86
|
+
"id": "ART-PERF-IMPROVED",
|
|
87
|
+
"required_fields": [
|
|
88
|
+
"전후지표",
|
|
89
|
+
"병목원인",
|
|
90
|
+
"개선내역"
|
|
91
|
+
]
|
|
92
|
+
},
|
|
93
|
+
{
|
|
94
|
+
"id": "ART-SYSTEM-BOUNDARY",
|
|
95
|
+
"required_fields": [
|
|
96
|
+
"신뢰경계",
|
|
97
|
+
"자산목록",
|
|
98
|
+
"인증인가흐름"
|
|
99
|
+
]
|
|
100
|
+
},
|
|
101
|
+
{
|
|
102
|
+
"id": "ART-SECURITY-GAPS",
|
|
103
|
+
"required_fields": [
|
|
104
|
+
"위협목록",
|
|
105
|
+
"취약점목록",
|
|
106
|
+
"우선순위"
|
|
107
|
+
]
|
|
108
|
+
},
|
|
109
|
+
{
|
|
110
|
+
"id": "ART-HARDENED-BUILD",
|
|
111
|
+
"required_fields": [
|
|
112
|
+
"보완코드",
|
|
113
|
+
"영향분석",
|
|
114
|
+
"재검증계획"
|
|
115
|
+
]
|
|
116
|
+
},
|
|
117
|
+
{
|
|
118
|
+
"id": "ART-SECURITY-VERIFIED",
|
|
119
|
+
"required_fields": [
|
|
120
|
+
"재검증결과",
|
|
121
|
+
"잔여위험",
|
|
122
|
+
"예외승인"
|
|
123
|
+
]
|
|
124
|
+
},
|
|
125
|
+
{
|
|
126
|
+
"id": "ART-INCIDENT-ALERT",
|
|
127
|
+
"required_fields": [
|
|
128
|
+
"탐지시각",
|
|
129
|
+
"증상",
|
|
130
|
+
"영향범위"
|
|
131
|
+
]
|
|
132
|
+
},
|
|
133
|
+
{
|
|
134
|
+
"id": "ART-INCIDENT-TIMELINE",
|
|
135
|
+
"required_fields": [
|
|
136
|
+
"타임라인",
|
|
137
|
+
"조치내역",
|
|
138
|
+
"복구시각"
|
|
139
|
+
]
|
|
140
|
+
},
|
|
141
|
+
{
|
|
142
|
+
"id": "ART-ROOT-CAUSE-FIX",
|
|
143
|
+
"required_fields": [
|
|
144
|
+
"원인",
|
|
145
|
+
"수정내역",
|
|
146
|
+
"회귀검증"
|
|
147
|
+
]
|
|
148
|
+
},
|
|
149
|
+
{
|
|
150
|
+
"id": "ART-POSTMORTEM-PACK",
|
|
151
|
+
"required_fields": [
|
|
152
|
+
"RCA",
|
|
153
|
+
"재발방지액션",
|
|
154
|
+
"담당자"
|
|
155
|
+
]
|
|
156
|
+
},
|
|
157
|
+
{
|
|
158
|
+
"id": "ART-PRD-MASTER",
|
|
159
|
+
"required_fields": [
|
|
160
|
+
"문제정의",
|
|
161
|
+
"목표지표",
|
|
162
|
+
"기능범위",
|
|
163
|
+
"비기능요구사항"
|
|
164
|
+
]
|
|
165
|
+
},
|
|
166
|
+
{
|
|
167
|
+
"id": "ART-PRD-DETAILS",
|
|
168
|
+
"required_fields": [
|
|
169
|
+
"에픽목록",
|
|
170
|
+
"세부기능PRD",
|
|
171
|
+
"수용기준",
|
|
172
|
+
"의존성순서"
|
|
173
|
+
]
|
|
174
|
+
},
|
|
175
|
+
{
|
|
176
|
+
"id": "ART-IMPLEMENTATION-BOOTSTRAP",
|
|
177
|
+
"required_fields": [
|
|
178
|
+
"배치실행순서",
|
|
179
|
+
"자동실행범위",
|
|
180
|
+
"수동설정체크리스트"
|
|
181
|
+
]
|
|
182
|
+
}
|
|
183
|
+
]
|
|
184
|
+
}
|