team-skills 1.2.0 → 1.2.2
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/README.md +63 -26
- package/package.json +1 -1
- package/skills/_team-rules/constitutional-rules.md +2 -2
- package/skills/_team-rules/first-principles.md +1 -0
- package/skills/_team-rules/four-state-protocol.md +12 -8
- package/skills/_team-rules/verification-protocol.md +6 -2
- package/skills/team-brainstorm/SKILL.md +4 -11
- package/skills/team-debug/SKILL.md +18 -13
- package/skills/team-feedback/SKILL.md +28 -44
- package/skills/team-finish/SKILL.md +22 -15
- package/skills/team-impl/SKILL.md +11 -44
- package/skills/team-orchestrator/SKILL.md +122 -111
- package/skills/team-review/SKILL.md +19 -58
- package/skills/team-score/SKILL.md +25 -44
- package/skills/team-spec/SKILL.md +38 -43
- package/skills/team-test/SKILL.md +22 -28
- package/skills/team-verify/SKILL.md +6 -10
- package/skills/using-team-skills/SKILL.md +5 -20
|
@@ -45,6 +45,8 @@ If you were dispatched as a subagent to execute a specific task, skip this skill
|
|
|
45
45
|
NO SKILL RECOMMENDATION WITHOUT SCENE ANALYSIS FIRST
|
|
46
46
|
```
|
|
47
47
|
|
|
48
|
+
> 作为 meta-skill，此 Iron Law 的核心约束是：推荐前必须分析场景，避免将用户引入错误的 Skill 流程。
|
|
49
|
+
|
|
48
50
|
## 质量职责
|
|
49
51
|
|
|
50
52
|
| 质量维度 | 产出文件 |
|
|
@@ -71,17 +73,7 @@ NO SKILL RECOMMENDATION WITHOUT SCENE ANALYSIS FIRST
|
|
|
71
73
|
|
|
72
74
|
### Step 1：分析用户场景
|
|
73
75
|
|
|
74
|
-
|
|
75
|
-
|
|
76
|
-
- 需求模糊 → 推荐 `team-brainstorm`
|
|
77
|
-
- 需求明确 → 推荐 `team-spec`
|
|
78
|
-
- 已有规格 → 推荐 `team-impl`
|
|
79
|
-
- 已有实现 → 推荐 `team-test`
|
|
80
|
-
- 已有代码 + 测试 → 推荐 `team-review`
|
|
81
|
-
- 遇到 bug → 推荐 `team-debug`
|
|
82
|
-
- 收到审查反馈 → 推荐 `team-feedback`
|
|
83
|
-
- 实现完成 → 推荐 `team-finish`
|
|
84
|
-
- 需要完整流水线 → 推荐 `team-orchestrator`
|
|
76
|
+
根据用户描述判断当前所处阶段，对照上方「Skill 选择矩阵」找到匹配的 Skill。如果场景不完全匹配单个条目，优先推荐覆盖用户核心需求的 Skill。
|
|
85
77
|
|
|
86
78
|
### Step 2：推荐并说明理由
|
|
87
79
|
|
|
@@ -120,15 +112,14 @@ NO SKILL RECOMMENDATION WITHOUT SCENE ANALYSIS FIRST
|
|
|
120
112
|
## 完成标志
|
|
121
113
|
|
|
122
114
|
```
|
|
123
|
-
状态：DONE
|
|
115
|
+
状态：DONE | DONE_WITH_CONCERNS | NEEDS_CONTEXT | BLOCKED
|
|
124
116
|
推荐 Skill：{skill-name}
|
|
125
117
|
推荐理由：{reason}
|
|
118
|
+
如有保留意见或阻塞，列出具体内容
|
|
126
119
|
```
|
|
127
120
|
|
|
128
121
|
## STOP Signals
|
|
129
122
|
|
|
130
|
-
如果你发现自己即将做以下任何一件事——立即停止，重新审视：
|
|
131
|
-
|
|
132
123
|
- 不分析场景就直接推荐 Skill
|
|
133
124
|
- 场景模糊时跳过 team-brainstorm 直接推荐实现类 Skill
|
|
134
125
|
- 凭记忆推荐 Skill 而不读取当前版本的选择矩阵
|
|
@@ -144,9 +135,3 @@ NO SKILL RECOMMENDATION WITHOUT SCENE ANALYSIS FIRST
|
|
|
144
135
|
|
|
145
136
|
- `team-brainstorm` — 需求模糊时先讨论
|
|
146
137
|
- `team-orchestrator` — 需要完整流水线时使用
|
|
147
|
-
|
|
148
|
-
## 下一步
|
|
149
|
-
|
|
150
|
-
- 根据 Skill 选择矩阵选择对应 skill 开始工作
|
|
151
|
-
- 不确定时使用 `team-brainstorm` 先讨论再决定
|
|
152
|
-
- 需要完整交付流水线时使用 `team-orchestrator`
|