flow-comet 1.6.4 → 1.7.0
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/.flow-comet/skills/flow-comet/INSTALLED_VERSION +1 -1
- package/.flow-comet/skills/flow-comet/SKILL.md +36 -6
- package/.flow-comet/skills/flow-comet/reference/artifact-template-overlay.md +154 -0
- package/.flow-comet/skills/flow-comet/reference/artifact-templates.json +387 -0
- package/.flow-comet/skills/flow-comet/reference/commit-discipline.md +1 -1
- package/.flow-comet/skills/flow-comet/reference/entry-detail.md +29 -4
- package/.flow-comet/skills/flow-comet/reference/workflow-protocol.json +22 -0
- package/.flow-comet/skills/flow-comet/scripts/context-init.mjs +26 -23
- package/.flow-comet/skills/flow-comet/scripts/evolve.mjs +174 -0
- package/.flow-comet/skills/flow-comet/scripts/guard-self-test.mjs +17 -95
- package/.flow-comet/skills/flow-comet/scripts/health.mjs +28 -4
- package/.flow-comet/skills/flow-comet/scripts/items/a.mjs +136 -19
- package/.flow-comet/skills/flow-comet/scripts/items/b.mjs +54 -18
- package/.flow-comet/skills/flow-comet/scripts/items/c.mjs +6 -6
- package/.flow-comet/skills/flow-comet/scripts/items/e.mjs +52 -10
- package/.flow-comet/skills/flow-comet/scripts/items/h.mjs +25 -10
- package/.flow-comet/skills/flow-comet/scripts/items/l.mjs +4 -2
- package/.flow-comet/skills/flow-comet/scripts/items/m.mjs +21 -13
- package/.flow-comet/skills/flow-comet/scripts/items/n.mjs +163 -0
- package/.flow-comet/skills/flow-comet/scripts/lib/ac-coverage.mjs +317 -0
- package/.flow-comet/skills/flow-comet/scripts/lib/guard-artifact-skeleton.mjs +448 -0
- package/.flow-comet/skills/flow-comet/scripts/lib/guard-evidence-gates.mjs +396 -27
- package/.flow-comet/skills/flow-comet/scripts/lib/guard-gates.mjs +1035 -172
- package/.flow-comet/skills/flow-comet/scripts/lib/guard-routing.mjs +129 -98
- package/.flow-comet/skills/flow-comet/scripts/lib/guard-state-io.mjs +8 -6
- package/.flow-comet/skills/flow-comet/scripts/lib/guard-template-fidelity.mjs +797 -232
- package/.flow-comet/skills/flow-comet/scripts/lib/guard-uidesign.mjs +91 -22
- package/.flow-comet/skills/flow-comet/scripts/lib/lessons-hits.mjs +160 -0
- package/.flow-comet/skills/flow-comet/scripts/lib/run-root.mjs +21 -0
- package/.flow-comet/skills/flow-comet/scripts/lib/section-body.mjs +92 -0
- package/.flow-comet/skills/flow-comet/scripts/lib/state-commands-admin.mjs +69 -13
- package/.flow-comet/skills/flow-comet/scripts/lib/state-commands-flow.mjs +454 -135
- package/.flow-comet/skills/flow-comet/scripts/lib/state-commands-recovery.mjs +66 -0
- package/.flow-comet/skills/flow-comet/scripts/lib/state-explain.mjs +109 -13
- package/.flow-comet/skills/flow-comet/scripts/lib/state-io.mjs +55 -2
- package/.flow-comet/skills/flow-comet/scripts/lib/state-skillmarkers.mjs +149 -34
- package/.flow-comet/skills/flow-comet/scripts/lib/system-fixtures.mjs +79 -9
- package/.flow-comet/skills/flow-comet/scripts/lib/task-graph.mjs +84 -31
- package/.flow-comet/skills/flow-comet/scripts/lib/test-fixtures.mjs +78 -8
- package/.flow-comet/skills/flow-comet/scripts/lib/test-helpers.mjs +60 -3
- package/.flow-comet/skills/flow-comet/scripts/lib/text-anchor-checks.mjs +167 -3
- package/.flow-comet/skills/flow-comet/scripts/lib/wave-projection.mjs +341 -0
- package/.flow-comet/skills/flow-comet/scripts/protocol-utils.mjs +287 -0
- package/.flow-comet/skills/flow-comet/scripts/scenarios/a.mjs +1454 -16
- package/.flow-comet/skills/flow-comet/scripts/scenarios/b.mjs +1202 -8
- package/.flow-comet/skills/flow-comet/scripts/scenarios/c.mjs +1205 -77
- package/.flow-comet/skills/flow-comet/scripts/scenarios/d.mjs +753 -5
- package/.flow-comet/skills/flow-comet/scripts/scenarios/e.mjs +52 -11
- package/.flow-comet/skills/flow-comet/scripts/scenarios/f.mjs +582 -3
- package/.flow-comet/skills/flow-comet/scripts/scenarios/g.mjs +9 -1
- package/.flow-comet/skills/flow-comet/scripts/scenarios/i.mjs +5 -0
- package/.flow-comet/skills/flow-comet/scripts/scenarios/j.mjs +22 -6
- package/.flow-comet/skills/flow-comet/scripts/scenarios/k.mjs +2 -0
- package/.flow-comet/skills/flow-comet/scripts/scenarios/l.mjs +18 -5
- package/.flow-comet/skills/flow-comet/scripts/scenarios/n.mjs +4100 -0
- package/.flow-comet/skills/flow-comet/scripts/state-schema.mjs +135 -0
- package/.flow-comet/skills/flow-comet/scripts/system-test.mjs +6 -5
- package/.flow-comet/skills/flow-comet/scripts/workflow-guard.mjs +203 -4
- package/.flow-comet/skills/flow-comet/scripts/workflow-handoff.mjs +312 -63
- package/.flow-comet/skills/flow-comet/scripts/workflow-state.mjs +5 -2
- package/.flow-comet/skills/flow-comet-archive/SKILL.md +6 -2
- package/.flow-comet/skills/flow-comet-change/SKILL.md +6 -2
- package/.flow-comet/skills/flow-comet-design/GUIDANCE.md +13 -3
- package/.flow-comet/skills/flow-comet-design/SKILL.md +15 -8
- package/.flow-comet/skills/flow-comet-dev/SKILL.md +1 -1
- package/.flow-comet/skills/flow-comet-evolve/SKILL.md +1 -0
- package/.flow-comet/skills/flow-comet-execute/SKILL.md +42 -15
- package/.flow-comet/skills/flow-comet-health/SKILL.md +1 -1
- package/.flow-comet/skills/flow-comet-integration/SKILL.md +1 -1
- package/.flow-comet/skills/flow-comet-open/GUIDANCE.md +14 -4
- package/.flow-comet/skills/flow-comet-open/SKILL.md +11 -9
- package/.flow-comet/skills/flow-comet-plan/GUIDANCE.md +13 -3
- package/.flow-comet/skills/flow-comet-plan/SKILL.md +21 -10
- package/.flow-comet/skills/flow-comet-requirement/SKILL.md +1 -1
- package/.flow-comet/skills/flow-comet-review/SKILL.md +33 -18
- package/.flow-comet/skills/flow-comet-subagent-execute/GUIDANCE.md +13 -3
- package/.flow-comet/skills/flow-comet-subagent-execute/SKILL.md +44 -16
- package/.flow-comet/skills/flow-comet-task/SKILL.md +3 -2
- package/.flow-comet/skills/flow-comet-test/SKILL.md +6 -2
- package/.flow-comet/skills/flow-comet-ui-design/SKILL.md +2 -2
- package/.flow-comet/skills/flow-comet-verify/SKILL.md +12 -8
- package/README.md +3 -3
- package/package.json +1 -1
- package/scripts/dsh-bridge.mjs +2 -2
- package/scripts/lib/installer-flowkit.mjs +43 -3
|
@@ -7,7 +7,11 @@ description: "flow-kit CHANGE 阶段协议:反问澄清、change-id 自动生
|
|
|
7
7
|
|
|
8
8
|
本 Skill 为 flow-comet 的 open 节点提供 flow-kit 的 CHANGE 反问协议。
|
|
9
9
|
|
|
10
|
-
>
|
|
10
|
+
> **模板权威**:本技能产出工件的段形 = **上游模板 + 本仓登记的增量段**——唯一权威 = `flow-kit/templates/**`(段与正文),条件段 / 结构子段 / 增量段与指纹由 `../flow-comet/reference/artifact-templates.json` 登记;`.specs/archive/**` 是历史证据、**不是模板来源**——不得以「上一轮就是这么写的」对齐段形。
|
|
11
|
+
|
|
12
|
+
## 上游段表不采用(覆盖声明)
|
|
13
|
+
|
|
14
|
+
`flow-kit/prompts/0-change.md` 的自检清单把 CHANGE.md 写成 5 段(Why / What / 影响面 / 范围排除 / 验收线)——**这 5 段清单不采用**作段表:它漏了 `视觉调性` 与 `风险与未知`。段名与段序以 `flow-kit/templates/CHANGE.md` 的 7 个 H2 为准(`视觉调性` 是标题必须在场的条件段,`架构层影响声明` 是本仓登记的可选增量段)。
|
|
11
15
|
|
|
12
16
|
## 加载
|
|
13
17
|
|
|
@@ -32,7 +36,7 @@ description: "flow-kit CHANGE 阶段协议:反问澄清、change-id 自动生
|
|
|
32
36
|
|
|
33
37
|
- **判据与强制等级(唯一权威)**:判据全文与 `guarded` 绑定、`required-skill:design.flow-comet-ui-design` 声明要求、非前端 `UI-DESIGN: skipped(非前端)` 的出口口径都在 `flow-comet-ui-design` 册的「触发与跳过」段——本节不复述,只描述怎么填段。
|
|
34
38
|
- **前端项目**:按步骤 0.6 的预选结果填段(选定 / 理由 / 参考产品 / 明确排除),不要以结构形态标注「不适用」。
|
|
35
|
-
- **非前端项目**:保留该段并以**结构形态**标注「不适用」(可附一句理由,例如「CLI
|
|
39
|
+
- **非前端项目**:保留该段并以**结构形态**标注「不适用」(可附一句理由,例如「CLI 工具,无用户可见界面」)。**段缺席不再默认判非前端**——新 change 的「视觉调性」是标题必须在场的条件段,**缺席即拦**(open 出口的骨架校验与 design 出口的适用性门控都会拦);缺席只对旧 change 兼容(沿用旧口径判非前端,出口打印 `UI-DESIGN: skipped(旧 change 兼容)`)。
|
|
36
40
|
- **改判走本阶段**:change 进行中要改前端 / 非前端判定,回到这里改这一个段,不在 design / execute / verify 里就地绕开判据——改段即改变了 design 出口的判据输入。
|
|
37
41
|
|
|
38
42
|
## 状态推进
|
|
@@ -15,9 +15,19 @@ Responsibility: 技术栈选型 + ADR + 数据流。生成 DESIGN.md。
|
|
|
15
15
|
|
|
16
16
|
### 必填段清单(结构+存在级)
|
|
17
17
|
|
|
18
|
-
|
|
19
|
-
|
|
20
|
-
|
|
|
18
|
+
段形唯一权威 = `flow-kit/templates/DESIGN.md` + 登记表 `../flow-comet/reference/artifact-templates.json`(模板缺席时出口输出「未校验(模板缺失)」)。
|
|
19
|
+
|
|
20
|
+
| 文件 | 无条件段(缺席 = BLOCKED) | 条件段与增量段(在场才校验 / 结构子段随父段) |
|
|
21
|
+
|------|---------------------------|-------------------------------------------|
|
|
22
|
+
| DESIGN.md | `## 0. 技术栈选定` / `## 1. 决策清单` / `## 2. 数据流 / 架构图` / `## 5. 风险` / `## 6. 不在范围` | 条件段:`## 0.5 既有架构对齐`(标题必须在场;父段适用时其结构子段 0.5.1~0.5.3 必须同时在,整段写「不适用」才豁免)/ `## 3. 关键状态机` / `## 4. ADR 索引` / `## 9. 架构沉淀建议`(适用时子段 9.1~9.5 必须在场) |
|
|
23
|
+
|
|
24
|
+
前端 change 另生成 `UI-DESIGN.md`(标题形态 `# UI Design: <项目名>` + frontmatter 键齐备);非前端须在 CHANGE.md 的 `## 视觉调性` 段以结构形态标注「不适用」。声明本次读到的模板:
|
|
25
|
+
|
|
26
|
+
```bash
|
|
27
|
+
node .claude/skills/flow-comet/scripts/workflow-state.mjs skill-load design flow-comet-design --prompt flow-kit/prompts/2-design.md --templates DESIGN.md
|
|
28
|
+
```
|
|
29
|
+
|
|
30
|
+
加载顺序:先 `READ PROMPT:`(步骤权威),有模板的节点再 `READ TEMPLATE:`(段形权威);design 前端另有一行条件提示词 `2a-ui-design`。
|
|
21
31
|
|
|
22
32
|
guard 校验见 workflow-guard.mjs NODE_TRANSITION_GATES / templateSectionPatterns;「填得好不好」由 review 把关。
|
|
23
33
|
|
|
@@ -17,15 +17,22 @@ Responsibility: 技术栈选型 + ADR + 数据流。生成 DESIGN.md(轻量流
|
|
|
17
17
|
|
|
18
18
|
### 必填段清单(结构+存在级)
|
|
19
19
|
|
|
20
|
-
|
|
|
21
|
-
|
|
22
|
-
| DESIGN.md | `## 0
|
|
20
|
+
| 文件(模板 = `flow-kit/templates/<DOC>.md`) | 左栏 · 无条件段(模板 H2,缺席 / 空正文 = BLOCKED) | 右栏 · 条件段(**标题必须在场**,正文可写 `N/A`)/ 增量段(按登记位置) |
|
|
21
|
+
|------|------|------|
|
|
22
|
+
| `DESIGN.md` | `## 0. 技术栈选定` / `## 1. 决策清单` / `## 2. 数据流 / 架构图` / `## 5. 风险` / `## 6. 不在范围`;首部 `Change ID` 必填 | **条件段**:`## 0.5 既有架构对齐` / `## 3. 关键状态机(如有)` / `## 4. ADR 索引` / `## 9. 架构沉淀建议`——标题必须在场,不适用写 `N/A`;`## 0.5` 的结构子段 `### 0.5.1` ~ `### 0.5.3`、`## 9` 的 `### 9.1` ~ `### 9.5` 在父段适用时必在场;**增量段**:无 |
|
|
23
|
+
| `UI-DESIGN.md`(仅前端 change) | frontmatter 八个顶层键(name / description / colors / typography / spacing / rounded / motion / shadow)与 `## 1.` ~ `## 10.` 十个 H2 段 | **可选段**:`## 0. 视觉语汇对齐`(仅 brownfield 填,greenfield 可整段删除——模板允许删除的唯一段,删则其 `### 0.1` ~ `### 0.3` 一并豁免);**增量段**:无 |
|
|
24
|
+
|
|
25
|
+
模板缺席(`flow-kit/templates/` 不在场)时骨架无从派生:该工件**不校验**并输出「未校验(模板缺失)」,不拿内置骨架顶替。
|
|
26
|
+
|
|
27
|
+
**上游段表不采用(覆盖声明)**:`flow-kit/prompts/2-design.md` 的职责小节把 §3 写成 ADR、§4 写成风险,而模板的顺序是 `## 3. 关键状态机` / `## 4. ADR 索引` / `## 5. 风险`——**这两处编号互换不采用**:段名、段序与结构子段一律以 `flow-kit/templates/DESIGN.md` 为准。
|
|
23
28
|
|
|
24
29
|
> **决策清单段名须与模板精确一致**:guard 按 flow-kit 模板段名派生校验——模板段为 `## 1. 决策清单`(编号可选),**不要写成"技术决策清单"等变体**(2-design.md 指引的"技术决策"是内容要求,不是段名;实测:写变体段名会被拦截)。
|
|
25
30
|
|
|
26
|
-
guard 校验见 workflow-guard.mjs
|
|
31
|
+
guard 校验见 workflow-guard.mjs 的出口门禁与模板骨架派生(条件段 / 结构子段 / 增量段的位置与必填性来自 `../flow-comet/reference/artifact-templates.json` 登记);「填得好不好」由 review 把关。
|
|
32
|
+
|
|
33
|
+
> **模板权威**:本节点产出工件的段形 = **上游模板 + 本仓登记的增量段**——唯一权威 = `flow-kit/templates/**`(段与正文),条件段 / 结构子段 / 增量段与指纹由 `../flow-comet/reference/artifact-templates.json` 登记;`.specs/archive/**` 是历史证据、**不是模板来源**——不得以「上一轮就是这么写的」对齐段形。
|
|
27
34
|
|
|
28
|
-
|
|
35
|
+
加载顺序:先读 `READ PROMPT:`(`2-design`;前端另加 `2a-ui-design`),再读 `READ TEMPLATE:`(`DESIGN`;前端另加 `UI-DESIGN`,可选)。
|
|
29
36
|
|
|
30
37
|
# Design
|
|
31
38
|
|
|
@@ -114,10 +121,10 @@ The design node reads CHANGE.md and REQUIREMENT.md, then produces DESIGN.md thro
|
|
|
114
121
|
1. 本节点的同名条目已随路由加载——同名 required 条目(`flow-comet-design`)无需重复加载,仅需运行下方声明命令(只读取 SKILL.md 文件不叫加载,**不得跳过**真正的 Skill 工具注入步骤)。`flow-comet-ui-design` 为 **guarded 但仅前端适用**的条目——前端项目用 Skill 工具加载其已装副本后**必须**运行对应 `skill-load` 声明建标记,并在 record 载荷带 `required-skill:design.flow-comet-ui-design`;非前端项目不加载不声明(出口打印可见的 `UI-DESIGN: skipped(非前端)` 后放行);旧 change 保持渐进,缺件仅 WARN。
|
|
115
122
|
|
|
116
123
|
> **前端判据(结构级 · 唯一权威 → `flow-comet-ui-design` 册的「触发与跳过」段)**:判据全文与 `guarded` 绑定的判级口径都在该册——本节不复述,判定与出口行为一律以该册为准。
|
|
117
|
-
2.
|
|
124
|
+
2. 加载完成后**立即**运行声明命令(节点退出与证据记录会核对声明标记;声明如实记录加载动作,不等于产出证明)。`--templates` 声明本节点该读的产出物模板:`design` 的必填模板是 `DESIGN.md`;**前端 change 另加 `UI-DESIGN.md` 时写成 `--templates DESIGN.md,UI-DESIGN.md`**(`UI-DESIGN` 属可选声明,非前端不加):
|
|
118
125
|
|
|
119
126
|
```bash
|
|
120
|
-
node .claude/skills/flow-comet/scripts/workflow-state.mjs skill-load design flow-comet-design --prompt flow-kit/prompts/2-design.md
|
|
127
|
+
node .claude/skills/flow-comet/scripts/workflow-state.mjs skill-load design flow-comet-design --prompt flow-kit/prompts/2-design.md --templates DESIGN.md
|
|
121
128
|
```
|
|
122
129
|
|
|
123
130
|
> **跑 skill-load 声明命令 ≠ 加载**:声明只把“哪次会话加载了哪个 skill、按哪份协议工作”写进状态供 exit/record 核对;真正加载只有第 1 步的 Skill 工具能做到。
|
|
@@ -146,7 +153,7 @@ node .claude/skills/flow-comet/scripts/workflow-state.mjs record design '{"summa
|
|
|
146
153
|
|--------------|-------|-----------------|
|
|
147
154
|
| `design-artifacts` | DESIGN.md (or DESIGN-lite.md) exists with section 0 | artifact-exists |
|
|
148
155
|
| `architecture-aligned` | Section 0.5 populated for brownfield | 执行纪律(review 把关),guard 不校验 |
|
|
149
|
-
| `ui-design-artifacts` | UI-DESIGN.md exists for frontend projects |
|
|
156
|
+
| `ui-design-artifacts` | UI-DESIGN.md exists for frontend projects | 结构级(guard 强制:前端 change 缺 `.specs/<change-id>/UI-DESIGN.md` → 新 change BLOCKED、旧 change WARN;判定依据 = `CHANGE.md` 的「视觉调性」段,判据见 `flow-comet-ui-design` 册) |
|
|
150
157
|
|
|
151
158
|
## Exit Check
|
|
152
159
|
|
|
@@ -7,7 +7,7 @@ description: "flow-kit DEV 阶段协议:TDD + 6 维自查 + diff 边界 verify
|
|
|
7
7
|
|
|
8
8
|
本 Skill 为 flow-comet 的 execute/subagent-execute 节点提供 flow-kit 的完整开发规则集。
|
|
9
9
|
|
|
10
|
-
>
|
|
10
|
+
> **模板权威**:本技能产出工件的段形 = **上游模板 + 本仓登记的增量段**——唯一权威 = `flow-kit/templates/**`(段与正文),条件段 / 结构子段 / 增量段与指纹由 `../flow-comet/reference/artifact-templates.json` 登记;`.specs/archive/**` 是历史证据、**不是模板来源**——不得以「上一轮就是这么写的」对齐段形。
|
|
11
11
|
|
|
12
12
|
## 加载
|
|
13
13
|
|
|
@@ -32,6 +32,7 @@ node .claude/skills/flow-comet/scripts/evolve.mjs scan --root .
|
|
|
32
32
|
- **只读**每个 `.specs/archive/<日期>-<change-id>/DESIGN.md` 的架构沉淀段(第 9 节)——**不读该文档的其它段**,防 change 级冻结决策被错误升级为项目级
|
|
33
33
|
- 输出候选清单:`<归档目录名>#<序号>` + 条目 + 目标文档;并列出本次已扫的 change(含无沉淀段、无 `DESIGN.md` 的目录),扫描范围对用户完全可见
|
|
34
34
|
- 候选落点按沉淀子段归类:可复用抽象 / 禁动清单 → `CONTEXT.md`「既有抽象索引」;项目级技术决策 → `CONTEXT.md`「已锁决策」;依赖 → `CONTEXT.md`「技术栈」;跨模块契约 → `ARCHITECTURE.md`「跨模块契约」(该文档不在场时需先建立,否则应用会被拒绝且零改动)
|
|
35
|
+
- **「结构问题」段(只读上报,与沉淀候选互不依赖)**:同一条命令末尾输出 `EVOLVE: 结构问题(.specs/ARCHITECTURE.md):…`——`.specs/ARCHITECTURE.md` 在场时按上游模板的 8 个 H2 与「ADR 列表」段的 `### ADR-<数字>` 条目字段形态比对;文件缺席输出「跳过」、骨架不可派生输出「未校验」、可比对时输出 0 项或逐条问题。该段**不进退出码、不写 state**,也不改变候选清单
|
|
35
36
|
|
|
36
37
|
### 2. 逐项 review(用户参与,不可跳过)
|
|
37
38
|
|
|
@@ -17,15 +17,17 @@ In subagent mode (default) this node is a coordinator: it does not write impleme
|
|
|
17
17
|
|
|
18
18
|
### 必填段清单(结构+存在级)
|
|
19
19
|
|
|
20
|
-
|
|
|
21
|
-
|
|
22
|
-
| SUMMARY.md | `## verify
|
|
20
|
+
| 文件(模板 = `flow-kit/templates/<DOC>.md`) | 左栏 · 无条件段(模板 H2,缺席 / 空正文 = BLOCKED) | 右栏 · 条件段(**标题必须在场**,正文可写 `N/A`)/ 增量段(按登记位置) |
|
|
21
|
+
|------|------|------|
|
|
22
|
+
| `SUMMARY.md` | `## 做了什么(一段话)` / `## 改动文件` / `## verify 输出(必填)` / `## 6 维自查`(须含实质内容)/ `## 越界检查(必填)` / `## 是否触发新工作` / `## 完成判定`;首部四字段 `Change ID` / `Task ID` / `完成时间` / `AI 角色` 必填 | **条件段**:`## 数据库迁移(涉及 schema 变更必填 · 来自 4-dev 步骤 1.7 / R4.5)` / `## 破坏性变更(涉及破坏性改动必填 · 来自 4-dev 步骤 1.8 / R4.6 / B4 老项目护栏)` / `## 决策与偏离(如有)`——标题必须在场,不适用写 `N/A`;**增量段**:`## 自检方法`(整段必填,紧随 `## 越界检查` 或文末)、`## LESSONS 扫描`(整段必填,紧随 `## 自检方法`) |
|
|
23
23
|
|
|
24
|
-
> **注意**:flow-kit 的 SUMMARY 模板**不含 `## 自检方法`
|
|
24
|
+
> **注意**:flow-kit 的 SUMMARY 模板**不含 `## 自检方法` 与 `## LESSONS 扫描` 两段**——它们是 flow-comet 登记的增量段(前者校验自检方法声明,后者校验开工前扫描可核对)。使用 `flow-kit/templates/SUMMARY.md` 填写后,**必须按上方右栏补写这两段**:`## 自检方法` 声明 brooks-review / cache-brooks / builtin-quickcheck 三值之一及降级证据,`## LESSONS 扫描` 声明命中条目与逐条回应;否则新 change 的 exit 会被 BLOCKED。
|
|
25
25
|
|
|
26
|
-
guard 校验见 workflow-guard.mjs
|
|
26
|
+
guard 校验见 workflow-guard.mjs 的 execute 出口门禁(必填段 + 6 维自查实质内容 + 自检方法声明三值 + LESSONS 扫描段与命中核对)与模板骨架派生(段标题 / 段序 / 首部字段 / 增量段位置);「填得好不好」由 review 把关。
|
|
27
27
|
|
|
28
|
-
>
|
|
28
|
+
> **模板权威**:本节点产出工件的段形 = **上游模板 + 本仓登记的增量段**——唯一权威 = `flow-kit/templates/**`(段与正文),条件段 / 结构子段 / 增量段与指纹由 `../flow-comet/reference/artifact-templates.json` 登记;`.specs/archive/**` 是历史证据、**不是模板来源**——不得以「上一轮就是这么写的」对齐段形。
|
|
29
|
+
|
|
30
|
+
加载顺序:先读 `READ PROMPT:`(`4-dev`),再读 `READ TEMPLATE:`(`SUMMARY`)。
|
|
29
31
|
|
|
30
32
|
### 执行模型(按 executionMode,用户显式选择)
|
|
31
33
|
|
|
@@ -36,6 +38,24 @@ guard 校验见 workflow-guard.mjs NODE_TRANSITION_GATES / W1-B;「填得好
|
|
|
36
38
|
|
|
37
39
|
执行模式由 `node .claude/skills/flow-comet/scripts/workflow-state.mjs status` 输出(`executionMode` / `directOverride`)决定,切换用 `node .claude/skills/flow-comet/scripts/workflow-state.mjs execution-mode <subagent|direct>`。direct 是逃生口,必须用户显式调用才生效。
|
|
38
40
|
|
|
41
|
+
### 节点视图三字段(currentNode / stateCurrentNode / routeNode)
|
|
42
|
+
|
|
43
|
+
`status` 与 `next` 读的是同一份节点视图,但三个字段各说一件事,不能互相替代:
|
|
44
|
+
|
|
45
|
+
- **currentNode** = **产物推导**的节点(按 `.specs/<change-id>/` 的工件推导出「按文件看现在在哪」;语义逐字未变);
|
|
46
|
+
- **stateCurrentNode** = **state 文件原值**(`.flow-comet/flow-comet-state.json` 里 `currentNode` 的原样读数,由 entry / exit / advance / reenter 写入;语义逐字未变);
|
|
47
|
+
- **routeNode** = **路由视图**(经「进行中保护」/ Fix 受控归位后的节点;常规路径下等于 `next` 的 `NODE:` 行,决定你现在该推进哪个节点)。
|
|
48
|
+
|
|
49
|
+
三者不一致不是 bug,而是同一事实的三个视角:`ROUTE:` 行给出 `derived=… state=… route=…` 的对账(derived = 产物推导、state = state 原值、route = 路由视图)与下一步动作。**Fix 批次以 `FIX-BATCH` 行与 `next` 的 `NODE` 为准**——归位目标就是你要先跑出口的节点;预检与出口都跑在那个节点上,不要按 `stateCurrentNode` 或 `currentNode` 自行改道。
|
|
50
|
+
|
|
51
|
+
**Node view: the three fields (currentNode / stateCurrentNode / routeNode)** — `status` and `next` read the same node view, but the three fields say three different things and cannot substitute for each other:
|
|
52
|
+
|
|
53
|
+
- **currentNode** = artifact-derived (the node derived from the artifacts under `.specs/<change-id>/`, i.e. "where the files say we are"; semantics unchanged);
|
|
54
|
+
- **stateCurrentNode** = the raw state-file value (the `currentNode` read verbatim from `.flow-comet/flow-comet-state.json`, written by entry / exit / advance / reenter; semantics unchanged);
|
|
55
|
+
- **routeNode** = the routing view (after in-progress protection / controlled Fix-batch rollback; on the regular path it equals the `NODE:` line printed by `next` and decides which node you should advance now).
|
|
56
|
+
|
|
57
|
+
A mismatch is not a bug but the same fact seen from three angles: the `ROUTE:` line gives the `derived=… state=… route=…` reconciliation (derived = artifact derivation, state = raw state value, route = routing view) plus the next action. **For a Fix batch, trust the `FIX-BATCH` line and the `NODE` printed by `next`** — the rollback target is the node whose exit you must run first; run preflight and exit on that node instead of rerouting by `stateCurrentNode` or `currentNode`.
|
|
58
|
+
|
|
39
59
|
### 任务范围
|
|
40
60
|
|
|
41
61
|
execute 节点**只处理 `parallel="false"`(或未标注 parallel)的 pending 任务**。
|
|
@@ -78,7 +98,7 @@ execute 节点**只处理 `parallel="false"`(或未标注 parallel)的 pendi
|
|
|
78
98
|
|
|
79
99
|
要点(五要素 + 并行波次附加两条):
|
|
80
100
|
|
|
81
|
-
1. **`git add -- <自己字面路径>`** + **`git commit -- <同一路径>`**:逐条字面路径(取自本任务 `write_files
|
|
101
|
+
1. **`git add -- <自己字面路径>`** + **`git commit -m "<消息>" -- <同一路径>`**:逐条字面路径(取自本任务 `write_files`),两步路径集逐字一致——提交只含自己的文件;消息(含尾注)写在 `--` **之前**,写成 `-m` 在 `--` 之后会被当成 pathspec;禁用 `.` / `-A` / `-u` 等宽泛形态。
|
|
82
102
|
2. **锁失败重试**:并发提交撞 `index.lock` / `cannot lock ref` ⇒ **失败即重试**,不得因锁失败改用宽泛命令。
|
|
83
103
|
3. **禁裸 `add` / `commit -a` / `stash` / `clean` / `reset`**:并发写者共享**同一个索引与工作树**——宽泛命令有**并发索引竞态**(把同伴半成品纳入自己的提交),`stash` / `clean` / `reset` 更会破坏同伴的未提交工作。
|
|
84
104
|
4. **并行波次禁用 `git commit --amend`**:共享工作区下 HEAD 会被并行写者推进;确需修正必须**先断言 `git rev-parse HEAD` 等于自己刚提交的哈希**,否则一律改用**新提交追加**。
|
|
@@ -110,13 +130,19 @@ execute 节点**只处理 `parallel="false"`(或未标注 parallel)的 pendi
|
|
|
110
130
|
|
|
111
131
|
对 TASK.md 每个 pending 串行任务,协调者执行:
|
|
112
132
|
|
|
113
|
-
1. **读 task 块,构造 handoff request**:读 `<task>` XML 块(`action` / `read_files` / `write_files` / `verify` / `done`)。若内容有歧义,停下询问——不要猜。构造 handoff request 内容:task 全文 + DESIGN §0/§0.5 + AC + read/write_files 边界。运行 `node .claude/skills/flow-comet/scripts/workflow-handoff.mjs request <task-id>` 记录。**委托即记 request
|
|
114
|
-
2.
|
|
133
|
+
1. **读 task 块,构造 handoff request**:读 `<task>` XML 块(`action` / `read_files` / `write_files` / `verify` / `done`)。若内容有歧义,停下询问——不要猜。构造 handoff request 内容:task 全文 + DESIGN §0/§0.5 + AC + read/write_files 边界。运行 `node .claude/skills/flow-comet/scripts/workflow-handoff.mjs request <task-id>` 记录。**委托即记 request(机制保证,不是经验之谈)**——`result` 在写盘前校验该任务存在 request 记录:缺失时新 change **BLOCKED**(state 字节零改写),并给出恢复指引「先照抄下面的命令记录委托(`description` 可省略),再重录 result」——命令形如 `node ".claude/skills/flow-comet/scripts/workflow-handoff.mjs" request T02`(设计形态,安装时脚本根按平台改写;引擎打印的就是带真实路径与任务号的那一行——脚本路径整条加引号,整行可直接执行);旧 change 只 WARN。零提交的 result 与只交自身 `<task-id>-SUMMARY.md` 的 result 两种形态同样被拦——该校验先于零提交语义与提交子集校验,那两者都从 request 取范围,因此不存在「先记 result 再补 request」的可行顺序。
|
|
134
|
+
2. **委托子代理**:
|
|
135
|
+
**委托顺序(硬顺序)**:先让 request 成功,再启动子代理——同一任务的 `workflow-handoff.mjs request`(步骤 1)与子代理启动是串行两步,禁止把 request 与子代理启动放进同一个并发调用(例如 `Promise.all` 里同时发 request 与 Agent / spawn_agent)。这是执行纪律,不是机制拦截:平台工具层没有可挂的判定点,引擎不会替你排队。**两件事别混**:本条管的是「request 与子代理启动的先后」,它没有挂点;「`result` 需要一条 request」是机制保证——缺 request 的 result 在写盘前被拒(新 change BLOCKED,恢复句见步骤 1)。
|
|
136
|
+
**Delegation order (mandatory)**: a successful request first, then launch the subagent — the `workflow-handoff.mjs request` call and the subagent launch are two sequential steps; never inside one concurrent call (e.g. request plus Agent / spawn_agent in the same `Promise.all`). This is execution discipline, not a mechanism interception: the platform tool layer offers no hook point, so the engine will not queue it for you. **Two different things**: this rule is about the order of the request and the subagent launch, and nothing enforces it; "a result needs a request" is a mechanism guarantee — a result without one is refused before anything is written (a new change blocks, recovery line in step 1).
|
|
137
|
+
按平台通道委托 fresh-context 子代理(三平台同一身份判据,差异只在谁建树;`isolation: "worktree"` 是属性①的可选实现,非强制——见「并行安全四属性契约」)。handoff prompt 强制协议:子代理用 **Skill 工具**加载 flow-comet-dev 并按 `flow-kit/prompts/4-dev.md` 协议执行(TDD RED/GREEN/REFACTOR、LESSONS 扫描、既有抽象 grep、verify、6 维自查、越界检查、原子 commit `<type>(<scope>): <subject>`(scope 推荐可选:子系统短词;禁 change-id/日期作 scope;任务号放提交正文尾注 Task: <task-id>,change-id 不入标题)),按 `flow-kit/templates/SUMMARY.md` 模板写 `.specs/<change-id>/<task-id>-SUMMARY.md`(标题/首部/段序保真,另补 flow-comet 增量 `## 自检方法` 段),回传 Return Contract(含 commitHash + greenEvidence + completedChecks + selfReview)。**SUMMARY 自引用 hash 注**:提交内容无法包含自身 hash——若子代理 amend 提交,SUMMARY 内记录的 commitHash 为 amend 前值属预期(文件集一致即可,以实际提交对象为准)。
|
|
115
138
|
**提交从属规则**:任务专属的 `<task-id>-SUMMARY.md`(位于 `.specs/<change-id>/`,是 flow-comet 强制产物)允许随任务提交属流程默认豁免——目标仓库的既有规定优先,若目标仓库忽略清单等既有规定拒绝其入库,被拒即为正确行为,严禁 force-add 强加越库提交;委托校验对任务摘要的豁免属于校验宽容度,不是入库指令。
|
|
116
|
-
**提交面 pathspec 纪律(五要素,见 Guidance 同名小节)**:handoff prompt 必须要求子代理用 `git add -- <自己字面路径>` + `git commit --
|
|
139
|
+
**提交面 pathspec 纪律(五要素,见 Guidance 同名小节)**:handoff prompt 必须要求子代理用 `git add -- <自己字面路径>` + `git commit -m "<消息>" -- <同一路径>`(消息在 `--` 之前)提交、锁失败重试、禁裸 `add` / `commit -a` / `stash` / `clean` / `reset`(适用条件:**多个写者共享同一工作区**;各自独立工作区内的局部 `reset` / `clean` 不在禁令内)。**集成纪律(四要素,见 Guidance 同名小节)**:协调者集成时优先 `git merge --no-ff`、降级 `cherry-pick` 记因、顺序按 `depends_on` 拓扑序 + 同层 task id 升序、冲突按机械 / 语义两类分别处置。
|
|
140
|
+
**写前读模板(handoff prompt 固定句式)**:prompt 必须写明「开工前先读 <本任务产出物的模板路径>,按其标题 / 首部 / 段序填写,并按本仓登记补写增量段」——模板路径逐任务给出(本节点默认 `flow-kit/templates/SUMMARY.md`);只写「按模板填写」视为委托前检查未完成,不得发出委托。
|
|
117
141
|
3. **记录 handoff result**:子代理回传后,运行 `node .claude/skills/flow-comet/scripts/workflow-handoff.mjs result <task-id> '<JSON>'` 记录(Return Contract 含 commitHash + greenEvidence + completedChecks + selfReview)。
|
|
118
|
-
4.
|
|
119
|
-
|
|
142
|
+
4. **每任务预检(`--preflight` · 只报结构问题)**:子代理回传并记下 `workflow-handoff.mjs result` 之后(direct 模式则在执行者交出 SUMMARY 之后),运行 `node .claude/skills/flow-comet/scripts/workflow-guard.mjs exit execute --preflight` —— 它只跑**结构性门禁**(本节点声明工件的存在性 / SUMMARY 关键段 / 模板保真骨架 / 入口工件骨架〔在场才查〕/ TASK 7 字段〔仅 `plan` 节点生效〕),**不跑**进展类门禁(证据、任务集签名、串行 pending、可委托并行、在飞委托、verify 命令执行),不推进状态、不写 state、也不写模板提示账本。**本家族的预检作用域** = 本节点声明工件的存在性(协议声明的 file 类产物,缺件即 BLOCK,逐项给出路径与模板指针)+ SUMMARY 关键段 + 模板保真骨架 + 入口三件(`CHANGE` / `REQUIREMENT` / `DESIGN`,存在才查);TASK 7 字段是 plan 出口的判据,对 execute 家族是 no-op,不在本步骤的范围内。**预检通过 = 结构面通过,不代表进展门禁会通过**(推进仍由完整 exit / `--apply` 决定)。价值:在「result 已记、尚未标 done」这个时点就能当场看到该任务 SUMMARY 的缺段 / 结构子段 / 条件段 / 首部字段问题——这正是本步骤存在的通道;不跑预检,这些问题会被更早的进展门禁挡下、攒到节点收尾一起爆。边界:它只报结构问题;**完整 `exit`(不带 `--apply`)仍是节点收尾的全链预演**——会先报进展类问题(串行 pending / 可委托并行 / 证据),预检不替代它。它只读、不写提示账本,因此**缺 H2 段只报类别计数**;结构子段与父段写法建议照常点名。要看被计数的那些段名,请等节点收尾的完整 `exit`(它维护提示账本),或直接按模板逐段对照。
|
|
143
|
+
**Per-task preflight (`--preflight`, structural only)**: after the subagent returns and its result is recorded (in direct mode, after the executor hands over the SUMMARY), run `node .claude/skills/flow-comet/scripts/workflow-guard.mjs exit execute --preflight` — it runs only the structural gates (existence of the node's declared artifacts / SUMMARY key sections / template-fidelity skeleton / intake-artifact skeleton, checked only when present / TASK seven fields, which apply at `plan` alone) and skips the progress gates (evidence, task-set signature, serial pending, delegable parallel tasks, in-flight handoffs, real verify execution); it does not advance, does not write workflow state, and does not write the template hint ledger. **The preflight scope of this family** = the existence of the node's declared file artifacts (a missing one blocks and names each path with its template) + the SUMMARY key sections + the template-fidelity skeleton + the three intake documents (`CHANGE` / `REQUIREMENT` / `DESIGN`, checked only when present); the TASK seven fields belong to the `plan` exit and are a no-op for the execute family, so they are not part of this step. **Passing the preflight means the structural face passes; it does not mean the progress gates will pass** (advancing is still decided by the full exit / `--apply`). Value: the task's SUMMARY defects (missing section / structure subsection / conditional section / header field) surface right away at the point where the result is recorded and the task is not yet marked done — that is exactly the channel this step exists for; without it those defects are blocked by earlier progress gates and pile up at the node exit. Boundary: it reports structural problems only; the **full `exit` without `--apply` remains the node's whole-chain rehearsal** and reports progress-class problems first (serial pending / delegable parallel tasks / evidence) — preflight does not replace it. Because it is read-only and writes no hint ledger, **a missing H2 section is reported as a category count only**; a missing structural subsection, and the advice naming the parent section that should hold it, are still named. For the counted headings, wait for the node's closing full `exit` (which keeps that hint ledger) or compare the document against the template section by section.
|
|
144
|
+
5. **验收 SUMMARY,TASK.md 标 done**:确认 SUMMARY 按 `flow-kit/templates/SUMMARY.md` 填写且**模板保真**(标题 `# SUMMARY:` / 首部 4 字段 / 段序一致)、含 `## 自检方法` 与 `## LESSONS 扫描` 两个增量段(前者声明 brooks-review / cache-brooks / builtin-quickcheck 三值之一)、verify 输出真实、6 维自查与越界检查有实质内容;通过后在 TASK.md 将任务标 `status="done"` 并加时间戳。
|
|
145
|
+
6. **下一个 pending 任务**:重复步骤 1-5。
|
|
120
146
|
|
|
121
147
|
> 原 Steps 的 TDD / LESSONS / verify / 6 维自查 / 越界检查 / commit 协议**移入 handoff prompt 作为子代理的强制协议**,协调者不亲自执行。
|
|
122
148
|
> **Return Contract 非代码任务**:纯文档 / 纯配置任务子代理回传时,`greenEvidence` 与 `redEvidence` 均允许 `{"command":"N/A (non-code task)","output":"..."}` 形态(command 字段存在即可通过 W1-D 校验;redEvidence 缺失在新 change 下仍会被 W1-D 拦截,因此必须显式提供此形态而非省略字段)。
|
|
@@ -148,6 +174,7 @@ This node is truly done when:
|
|
|
148
174
|
- **Agent thought**: "I see a bug in an adjacent file, let me fix it along the way." **Actual risk**: "Fixing along the way" without a new task or CHANGE violates R7.1 (scope control). Must stop and create a new task or CHANGE.
|
|
149
175
|
- **Agent thought**: "This task is taking long, let me skip the self-review." **Actual risk**: Skipping self-review defers quality issues to the review node, where they become Critical items requiring fix tasks. Better to catch early.
|
|
150
176
|
- **Agent thought**: "这个 parallel 任务小,我顺手在主会话做了。" **Actual risk**: 违反并行委托设计——subagent-execute 节点被静默跳过,丢失并行隔离与 write_files 冲突防护;且 execute 出口的 all-tasks-done 校验会 BLOCKED(parallel 无 handoffResult)。parallel="true" 任务只能由 subagent-execute 委托。
|
|
177
|
+
- **Agent thought**: "同时发 request 和启动子代理,省一个来回。" **Actual risk**: 先让 request 成功,再启动子代理——禁止把 request 与子代理启动放进同一个并发调用(例如 `Promise.all`);这是执行纪律,不是机制拦截(平台层没有挂点,引擎不替你排队)。与它相邻的另一件事已是机制保证:`result` 在写盘前校验该任务有 request 记录,缺失即新 change BLOCKED 并给出补 request 的恢复句。The order is mandatory: **a successful request first, then launch the subagent**, **never inside one concurrent call**; this is **execution discipline, not a mechanism interception** — the platform tool layer has no hook point to enforce it.
|
|
151
178
|
|
|
152
179
|
## 修复回路状态机路径
|
|
153
180
|
|
|
@@ -185,10 +212,10 @@ Load `flow-comet-dev` during this Node and record completed check `required-skil
|
|
|
185
212
|
**加载声明(阶段层 · 双步硬规则)**:本节点技能已由入口路由经 Skill 工具加载(你正在阅读的就是它);本节点 Required Skill Calls 的加载与声明同样不可跳过:
|
|
186
213
|
|
|
187
214
|
1. **用 Skill 工具加载** `flow-comet-dev`(本节点 Required Skill Call)。**不得跳过**——只读取 SKILL.md 文件不叫加载;真正让 flow-comet-dev 指令生效的是 Skill 工具把它注入本次会话。委托子代理时,子代理同样必须用 Skill 工具加载 flow-comet-dev(按 `flow-kit/prompts/4-dev.md` 协议执行)。
|
|
188
|
-
2.
|
|
215
|
+
2. 加载完成后**立即**运行声明命令(节点退出与证据记录会核对声明标记;声明如实记录加载动作,不等于产出证明)。`--templates` 声明本节点该读的产出物模板(execute → `SUMMARY.md`):
|
|
189
216
|
|
|
190
217
|
```bash
|
|
191
|
-
node .claude/skills/flow-comet/scripts/workflow-state.mjs skill-load execute flow-comet-dev --prompt flow-kit/prompts/4-dev.md
|
|
218
|
+
node .claude/skills/flow-comet/scripts/workflow-state.mjs skill-load execute flow-comet-dev --prompt flow-kit/prompts/4-dev.md --templates SUMMARY.md
|
|
192
219
|
```
|
|
193
220
|
|
|
194
221
|
> **跑 skill-load 声明命令 ≠ 加载**:声明只把“哪次会话加载了哪个 skill、按哪份协议工作”写进状态供 exit/record 核对;真正加载只有第 1 步的 Skill 工具能做到。
|
|
@@ -220,7 +247,7 @@ node .claude/skills/flow-comet/scripts/workflow-state.mjs record execute '{"summ
|
|
|
220
247
|
|--------------|-------|-----------------|
|
|
221
248
|
| `build-evidence` | At least one SUMMARY.md produced | artifact-exists |
|
|
222
249
|
| `all-tasks-done` | All tasks in TASK.md have status="done"(pending 串行任务 → BLOCKED);done 任务须有对应 SUMMARY(新 change 强制 BLOCKED,旧 change WARN 渐进) | content-check |
|
|
223
|
-
| `verify-output-real` | Every SUMMARY.md has verify output (not fabricated) |
|
|
250
|
+
| `verify-output-real` | Every SUMMARY.md has verify output (not fabricated) | 结构级(guard 强制:`## verify 输出` 段在场且正文非空,骨架校验同批拦截);输出是否真实由 review 把关(伪造不可机检) |
|
|
224
251
|
| `no-design-changes` | REQUIREMENT.md and DESIGN.md not modified | 执行纪律(review 把关),guard 不校验 |
|
|
225
252
|
| `self-check-method` | Every SUMMARY.md declares `## 自检方法`(brooks-review / cache-brooks / builtin-quickcheck) | W1-B 段级校验 |
|
|
226
253
|
|
|
@@ -25,7 +25,7 @@ description: "Use only when explicitly invoked as /flow-comet-health; periodic c
|
|
|
25
25
|
|
|
26
26
|
| 层 | 内容 | 产出者 |
|
|
27
27
|
|---|---|---|
|
|
28
|
-
| 确定性层 | 项目上下文一致性 · 经验条目编号连续性 · 技术债表项数 · 冗余工具在场判定 · 版本历史统计 | `health.mjs
|
|
28
|
+
| 确定性层 | 项目上下文一致性 · 经验条目编号连续性 · 技术债表项数 · 冗余工具在场判定 · 版本历史统计 · **架构文档结构**(`.specs/ARCHITECTURE.md` 在场时按上游模板的 8 个 H2 与「ADR 列表」段的 `### ADR-<数字>` 条目形态报告结构问题;文件缺席输出「跳过」、骨架不可派生输出「未校验」——只读报告,不阻断、不写状态) | `health.mjs`(逐项带计数、判据与复现命令) |
|
|
29
29
|
| 增补层 | 代码体检工具的 4 维结果(代码质量 / 架构 / 技术债 / 测试质量) | agent:工具在场时执行体检技能,结果**并入同一份报告**并标注来源 |
|
|
30
30
|
| 降级声明 | 可选工具不在场的原因 + 判据说明 | `health.mjs`(每份报告都有) |
|
|
31
31
|
|
|
@@ -7,7 +7,7 @@ description: "flow-kit INTEGRATION 阶段协议:集成验证 + UAT + 失败诊
|
|
|
7
7
|
|
|
8
8
|
本 Skill 为 flow-comet 的 verify 和 archive 节点提供 flow-kit 的集成验证和归档流程。
|
|
9
9
|
|
|
10
|
-
>
|
|
10
|
+
> **模板权威**:本技能产出工件的段形 = **上游模板 + 本仓登记的增量段**——唯一权威 = `flow-kit/templates/**`(段与正文),条件段 / 结构子段 / 增量段与指纹由 `../flow-comet/reference/artifact-templates.json` 登记;`.specs/archive/**` 是历史证据、**不是模板来源**——不得以「上一轮就是这么写的」对齐段形。
|
|
11
11
|
|
|
12
12
|
## 加载
|
|
13
13
|
|
|
@@ -15,10 +15,20 @@ Responsibility: CHANGE 反问 + REQUIREMENT 需求分析。生成 CHANGE.md 和
|
|
|
15
15
|
|
|
16
16
|
### 必填段清单(exit guard 校验,结构+存在级)
|
|
17
17
|
|
|
18
|
-
|
|
19
|
-
|
|
20
|
-
|
|
|
21
|
-
|
|
18
|
+
段形唯一权威 = `flow-kit/templates/<DOC>.md` + 登记表 `../flow-comet/reference/artifact-templates.json`(模板缺席时出口输出「未校验(模板缺失)」,不回退任何内置段名副本)。
|
|
19
|
+
|
|
20
|
+
| 文件 | 无条件段(缺席 = BLOCKED) | 条件段与增量段(在场才校验 / 位置按登记表) |
|
|
21
|
+
|------|---------------------------|-------------------------------------------|
|
|
22
|
+
| CHANGE.md | `## Why(为什么做)` / `## What(做什么)` / `## 影响面` / `## 范围排除(这次不做)` / `## 验收线(粗粒度,不是 AC)` / `## 风险与未知` | 条件段 `## 视觉调性`:标题必须在场(前端项目填调性;非前端显式写「不适用」——缺席即 BLOCKED);增量段 `## 架构层影响声明`(可选) |
|
|
23
|
+
| REQUIREMENT.md | `## 用户故事` / `## 验收准则(AC)` / `## 范围切分` / `## 非功能性需求` / `## 依赖与假设` | 无 |
|
|
24
|
+
|
|
25
|
+
段名宽容:省括号 / 换括号 / 大小写差异均按归一后的段名匹配,核心词不同即视为缺段。声明本次读到的模板(缺声明在 record 时被拦):
|
|
26
|
+
|
|
27
|
+
```bash
|
|
28
|
+
node .claude/skills/flow-comet/scripts/workflow-state.mjs skill-load open flow-comet-change --prompt flow-kit/prompts/0-change.md --templates CHANGE.md,REQUIREMENT.md
|
|
29
|
+
```
|
|
30
|
+
|
|
31
|
+
加载顺序:先 `READ PROMPT:`(步骤权威),有模板的节点再 `READ TEMPLATE:`(段形权威);design 前端另有一行条件提示词 `2a-ui-design`。
|
|
22
32
|
|
|
23
33
|
guard 校验见 workflow-guard.mjs NODE_TRANSITION_GATES / templateSectionPatterns;「填得好不好」由 review 把关。
|
|
24
34
|
|
|
@@ -17,14 +17,16 @@ Responsibility: CHANGE 反问 + REQUIREMENT 需求分析。生成 CHANGE.md 和
|
|
|
17
17
|
|
|
18
18
|
### 必填段清单(结构+存在级)
|
|
19
19
|
|
|
20
|
-
|
|
|
21
|
-
|
|
22
|
-
| CHANGE.md | `## Why(为什么做)`
|
|
23
|
-
| REQUIREMENT.md | `## 用户故事` / `## 验收准则(AC)`
|
|
20
|
+
| 文件(模板 = `flow-kit/templates/<DOC>.md`) | 左栏 · 无条件段(模板 H2,缺席 / 空正文 = BLOCKED) | 右栏 · 条件段(**标题必须在场**,正文可写 `N/A`)/ 增量段(按登记位置) |
|
|
21
|
+
|------|------|------|
|
|
22
|
+
| `CHANGE.md` | `## Why(为什么做)` / `## What(做什么)` / `## 影响面` / `## 范围排除(这次不做)` / `## 验收线(粗粒度,不是 AC)` / `## 风险与未知`;首部 `Change ID` 必填 | **条件段**:`## 视觉调性(前端项目必填,由 0-change 步骤 0.6 预选填入)`——标题必须在场,非前端以结构形态标注「不适用」;**增量段**:`## 架构层影响声明`(可选,位置 = 文末 / `## 风险与未知` 之后) |
|
|
23
|
+
| `REQUIREMENT.md` | `## 用户故事` / `## 验收准则(AC)` / `## 范围切分` / `## 非功能性需求` / `## 依赖与假设`;`## 范围切分` 的结构子段 `### v1(本次必做)` / `### v2(下一轮考虑,不本次)` / `### out(永远不做)` 在父段适用时必在场;首部 `Change ID` 必填 | 无模板条件段、无增量段——`## 范围切分` 整段写 `N/A` 时其结构子段随之豁免 |
|
|
24
24
|
|
|
25
|
-
guard 校验见 workflow-guard.mjs
|
|
25
|
+
模板缺席(`flow-kit/templates/` 不在场)时骨架无从派生:该工件**不校验**并输出「未校验(模板缺失)」,不拿内置骨架顶替。guard 校验见 workflow-guard.mjs 的出口门禁与模板骨架派生(条件段 / 结构子段 / 增量段的位置与必填性来自 `../flow-comet/reference/artifact-templates.json` 登记);「填得好不好」由 review 把关。
|
|
26
26
|
|
|
27
|
-
>
|
|
27
|
+
> **模板权威**:本节点产出工件的段形 = **上游模板 + 本仓登记的增量段**——唯一权威 = `flow-kit/templates/**`(段与正文),条件段 / 结构子段 / 增量段与指纹由 `../flow-comet/reference/artifact-templates.json` 登记;`.specs/archive/**` 是历史证据、**不是模板来源**——不得以「上一轮就是这么写的」对齐段形。
|
|
28
|
+
|
|
29
|
+
加载顺序:先读 `READ PROMPT:`(`0-change`、`1-requirement`),再读 `READ TEMPLATE:`(`CHANGE`、`REQUIREMENT`)。
|
|
28
30
|
|
|
29
31
|
# Open
|
|
30
32
|
|
|
@@ -111,11 +113,11 @@ The open node loads `flow-comet-change` for the CHANGE phase (structured questio
|
|
|
111
113
|
**加载声明(阶段层 · 双步硬规则)**:本节点技能已由入口路由经 Skill 工具加载(你正在阅读的就是它);本节点 Required Skill Calls 的加载与声明同样不可跳过:
|
|
112
114
|
|
|
113
115
|
1. **用 Skill 工具加载** `flow-comet-change` 与 `flow-comet-requirement`(本节点 Required Skill Calls)。**不得跳过**——只读取 SKILL.md 文件不叫加载;真正让这两个 skill 的指令生效的是 Skill 工具把它们注入本次会话。
|
|
114
|
-
2.
|
|
116
|
+
2. 加载完成后**立即**运行声明命令,每个协议文件对应一条(节点退出与证据记录会核对声明标记;声明如实记录加载动作,不等于产出证明)。`--templates` 声明本节点该读的产出物模板,**每条命令都写全清单**(两个标记各一份,节点级取并集):
|
|
115
117
|
|
|
116
118
|
```bash
|
|
117
|
-
node .claude/skills/flow-comet/scripts/workflow-state.mjs skill-load open flow-comet-change --prompt flow-kit/prompts/0-change.md
|
|
118
|
-
node .claude/skills/flow-comet/scripts/workflow-state.mjs skill-load open flow-comet-requirement --prompt flow-kit/prompts/1-requirement.md
|
|
119
|
+
node .claude/skills/flow-comet/scripts/workflow-state.mjs skill-load open flow-comet-change --prompt flow-kit/prompts/0-change.md --templates CHANGE.md,REQUIREMENT.md
|
|
120
|
+
node .claude/skills/flow-comet/scripts/workflow-state.mjs skill-load open flow-comet-requirement --prompt flow-kit/prompts/1-requirement.md --templates CHANGE.md,REQUIREMENT.md
|
|
119
121
|
```
|
|
120
122
|
|
|
121
123
|
> **跑 skill-load 声明命令 ≠ 加载**:声明只把“哪次会话加载了哪个 skill、按哪份协议工作”写进状态供 exit/record 核对;真正加载只有第 1 步的 Skill 工具能做到。
|
|
@@ -15,9 +15,19 @@ Responsibility: 拆原子任务(XML 格式)+ 波次划分。生成 TASK.md
|
|
|
15
15
|
|
|
16
16
|
### 必填段清单(结构+存在级)
|
|
17
17
|
|
|
18
|
-
|
|
19
|
-
|
|
20
|
-
|
|
|
18
|
+
段形唯一权威 = `flow-kit/templates/TASK.md` + 登记表 `../flow-comet/reference/artifact-templates.json`(模板缺席时出口输出「未校验(模板缺失)」)。
|
|
19
|
+
|
|
20
|
+
| 文件 | 无条件段(缺席 = BLOCKED) | 条件段与增量段(在场才校验 / 位置按登记表) |
|
|
21
|
+
|------|---------------------------|-------------------------------------------|
|
|
22
|
+
| TASK.md | `## 波次划分` / `## 任务清单` / `## 状态字段说明` / `## 阻塞日志` / `## Fix 任务(来自 REVIEW / INTEGRATION)`(以上段形按模板)+ 至少一个 `<task>` 块 + 每个任务含 `<verify>` 字段 + 新格式任务的 7 字段在场(集合见本册 `SKILL.md`「必填段清单」表行;单一权威 = 技能脚本的 `lib/task-graph.mjs` 导出的 `TASK7_REQUIRED` 常量,本文不复述集合) | 无(任务规模 / 粒度 / 字段内容质量属模板要求,guard 不拦,由 review 把关);`id` 是 `<task>` 元素的属性(不是字段:`<task id="T01" parallel="true" status="pending">`) |
|
|
23
|
+
|
|
24
|
+
声明本次读到的模板:
|
|
25
|
+
|
|
26
|
+
```bash
|
|
27
|
+
node .claude/skills/flow-comet/scripts/workflow-state.mjs skill-load plan flow-comet-plan --prompt flow-kit/prompts/3-task.md --templates TASK.md
|
|
28
|
+
```
|
|
29
|
+
|
|
30
|
+
加载顺序:先 `READ PROMPT:`(步骤权威),有模板的节点再 `READ TEMPLATE:`(段形权威);design 前端另有一行条件提示词 `2a-ui-design`。
|
|
21
31
|
|
|
22
32
|
guard 校验见 workflow-guard.mjs NODE_TRANSITION_GATES / W1-B;「填得好不好」由 review 把关。
|
|
23
33
|
|
|
@@ -17,21 +17,27 @@ Responsibility: 拆原子任务(XML 格式)+ 波次划分。生成 TASK.md
|
|
|
17
17
|
|
|
18
18
|
### 必填段清单(结构+存在级)
|
|
19
19
|
|
|
20
|
-
|
|
|
21
|
-
|
|
22
|
-
| TASK.md | 至少一个 `<task>` 块 + 每个任务含 `<verify>` 字段 + 新格式任务的 7 字段在场(`name` / `read_files` / `write_files` / `action` / `verify` / `done` / `depends_on
|
|
20
|
+
| 文件(模板 = `flow-kit/templates/<DOC>.md`) | 左栏 · 无条件段(模板 H2,缺席 / 空正文 = BLOCKED) | 右栏 · 条件段(**标题必须在场**,正文可写 `N/A`)/ 增量段(按登记位置) |
|
|
21
|
+
|------|------|------|
|
|
22
|
+
| TASK.md | 至少一个 `<task>` 块 + 每个任务含 `<verify>` 字段 + 新格式任务的 7 字段在场(`name` / `read_files` / `write_files` / `action` / `verify` / `done` / `depends_on`);模板 5 个 H2 段 `## 波次划分` / `## 任务清单` / `## 状态字段说明` / `## 阻塞日志` / `## Fix 任务(来自 REVIEW / INTEGRATION)` | 无模板条件段、无增量段(5 个 H2 全部无条件)——任务块的 7 字段与任务图(依赖环 / 缺失依赖 / 并行写写冲突)属元素级判据,见下方「硬性拦截」段 |
|
|
23
23
|
|
|
24
|
-
|
|
24
|
+
首部字段 `Change ID` / `关联` 已在场时值不得为空(键集不强制)。
|
|
25
25
|
|
|
26
|
-
|
|
26
|
+
guard 校验见 workflow-guard.mjs NODE_TRANSITION_GATES / W1-B:任务块结构、每任务 `<verify>`、**新格式任务的 7 字段在场**、以及 TASK.md 的模板骨架(段标题 / 段序 / 首部字段值非空)都由 guard 拦(新 change 缺任一 = 出口 BLOCKED;旧格式模板 / 旧 change 渐进 WARN);「填得好不好」——字段内容质量与任务粒度——由 review 把关。
|
|
27
27
|
|
|
28
|
-
|
|
28
|
+
> **模板权威**:本节点产出工件的段形 = **上游模板 + 本仓登记的增量段**——唯一权威 = `flow-kit/templates/**`(段与正文),条件段 / 结构子段 / 增量段与指纹由 `../flow-comet/reference/artifact-templates.json` 登记;`.specs/archive/**` 是历史证据、**不是模板来源**——不得以「上一轮就是这么写的」对齐段形。
|
|
29
|
+
|
|
30
|
+
加载顺序:先读 `READ PROMPT:`(`3-task`),再读 `READ TEMPLATE:`(`TASK`)。
|
|
31
|
+
|
|
32
|
+
波次散文一致性:`## 波次划分` 中标记为并行([P])的任务必须与 XML 任务 `parallel="true"` 一致——新 change 不一致 BLOCKED(旧 change WARN 渐进)。波次投影一致性:`## 波次划分` 声明的波次必须等于任务的依赖层级(判据与恢复指引见下方「波次形态约束」)——新 change 不一致 BLOCKED(旧 change WARN 渐进);判不了按成因二分——**载体缺席**(`## 波次划分` 不在 / 段名无从派生)→ 打印「未校验(无波次段)」并跳过该判据(未验证 ≠ 通过),**内容不可判**(段在场而没有一行 Wave 行能落到真实任务 id)→ 出口打印 `TASK.md 波次划分段内容不可判`,新 change BLOCKED / 旧 change WARN 渐进(该行进入 WARN 计数);任务集里的任务还必须被某条 Wave 行覆盖(修复族除外,漏写的以「未声明」点名)。
|
|
29
33
|
|
|
30
34
|
### 波次形态约束(依赖驱动多趟 · 规划期必守)
|
|
31
35
|
|
|
32
|
-
**约束本体**:并行任务按依赖关系组织——每个并行任务显式声明 `depends_on
|
|
36
|
+
**约束本体**:并行任务按依赖关系组织——每个并行任务显式声明 `depends_on`,依赖面的合法性只取决于依赖图本身:`depends_on` 无环且引用的任务都存在,依赖面即合法;波次编号另按「波次 = 依赖层级的投影」(下一条)单独校验——依赖面合法不等于波次声明合法,两条判据各自独立拦截。全并行、全串行、串→并、并→串四种基本形态与任意混排序列(如 `S→P→S`、`P→S→P`)一律放行——引擎按依赖拓扑自动分趟(每趟委托全部依赖已满足的并行任务,趟间回串行消化),混排序列由多趟路由按依赖拓扑自动分趟消化,并行任务的先后位置不受任何限制。
|
|
33
37
|
|
|
34
|
-
|
|
38
|
+
**波次 = 依赖层级的投影(波次编号的判据)**:`## 波次划分` 里的波次编号不是自由排版,而是任务依赖层级的投影——① 每个任务的波次 = 其依赖中最晚波次的下一波(无依赖 = 第 1 波);② 同波任务必须彼此没有依赖;③ 想让某任务晚一点跑,就写 `depends_on` 把顺序声明出来,而不是把它挪到后面的波却不声明。声明波次与依赖层级不一致(同波依赖 / 排在依赖之前 / 明明只需更早波次却排得更靠后)→ 新 change 出口 **BLOCKED**:明细逐条点名「任务 / 依赖 / 当前波 / 应到波」,并给两条恢复出路——把该任务移到第 N 波,或补 `depends_on` 把顺序声明出来(要保留当前波次,就把「它为何排在后面」的依赖写成显式 `depends_on`);旧 change 只 WARN 渐进。**判不了按成因二分**:**载体缺席**(无 `## 波次划分` 段 / 段名无从派生)→ 打印「未校验(无波次段)」并跳过该判据(未验证 ≠ 通过);**内容不可判**(段在场而没有一行 Wave 行能落到真实任务 id——正文只写 `N/A`、Wave 行点的是不存在的 id、行首强调记号让 Wave 行不成立)→ 出口打印 `TASK.md 波次划分段内容不可判`,新 change BLOCKED / 旧 change WARN 渐进(该行进入 WARN 计数)。**覆盖义务**:TASK 里的每个非修复族任务都必须被某个 Wave 行覆盖(修复族不要求排进波次),漏写的任务在明细里以「未声明」的当前波逐条点名——没有「当前波」可挪,恢复出路是把它写进应到波的 Wave 行。`replan` 重校走同一判据、同一报文,不豁免。**与混排合法化并存**:本条只要求「波次编号与依赖层级一致」,不恢复旧的「并行块必须连续成组」形态约束——上面的串→并→串混排照旧合法,并行任务的先后位置照旧不受限制。
|
|
39
|
+
|
|
40
|
+
**硬性拦截(规划出口 BLOCKED,附恢复指引)**:任务块结构(至少一个 `<task>` 块)/ 每任务 `<verify>` 字段 / 新格式任务的 7 字段在场(集合见上表;新 change 缺任一即 BLOCKED,旧格式模板 / 旧 change 渐进 WARN)/ 任务图依赖环 / 缺失依赖(`depends_on` 引用不存在的任务 id)/ 并行写写冲突(并行任务 `write_files` 重叠)/ 波次声明与依赖层级不一致(明细与两条恢复出路见上一条)——任一项不成立即在规划出口被 BLOCKED(读写重叠仅 WARN),按提示修正后重新 exit 即可通过。任务间依赖不得成环(如 A→B→A),也不得构成不可满足的依赖链。
|
|
35
41
|
|
|
36
42
|
**警示**:落笔时先画依赖方向再标 [P],避免出口返工。
|
|
37
43
|
|
|
@@ -83,11 +89,14 @@ This node decomposes the technical design into atomic, executable tasks with cle
|
|
|
83
89
|
- `done`: One sentence completion criteria, corresponding to an AC sub-item.
|
|
84
90
|
|
|
85
91
|
7. **Wave division**: Group tasks by dependency graph:
|
|
92
|
+
- **Waves are a projection of dependency levels**: each task's wave is one past the latest wave among its dependencies (no dependency = wave 1), and tasks in the same wave must not depend on one another; to make a task run later, declare the order with `depends_on` instead of moving it to a later wave without declaring it.
|
|
86
93
|
- Cross layer = sequential execution.
|
|
87
94
|
- Same-layer tasks may carry `parallel="true"` only when the wave genuinely contains a concurrent companion and their `write_files` do not overlap; an isolated task stays serial.
|
|
88
95
|
- Flow-comet override: the permissive flow-kit upstream semantics — "no conflict ⇒ [P]" and "same layer = same wave, parallel" — are superseded by this node's rule; the vendored upstream files stay read-only and are not modified.
|
|
89
96
|
- Output wave diagram: `Wave 1: T01[P], T02[P]` etc.; isolated tasks appear as serial steps.
|
|
90
97
|
|
|
98
|
+
A declared wave that disagrees with the dependency level — a same-wave dependency, a task placed before its dependency, or a conservative placement later than needed — blocks a new change at the plan exit: every mismatch is named as task / dependencies / declared wave / expected wave, together with the two ways out (move it to wave N, or add the `depends_on` that explains the order), while an older change only warns. When the section is absent (no `## 波次划分` section, or its name cannot be derived), the exit prints "not checked (no wave section)" and skips the rule — unverified is not a pass. When the section is present but no Wave line resolves to a real task (a body that only says N/A, a line naming an unknown id, a leading emphasis mark that breaks the line), the content cannot be judged: a new change blocks there, an older change warns, and that warning is counted with the other warnings. Every task in the set must be covered by some Wave line (the fix-task family excepted); a task the section never mentions is named with "undeclared" as its current wave, and the way out is to write it into the wave it should reach. `replan` runs the same judgement with the same wording. This rule constrains the agreement between wave numbers and dependency levels only: it does not bring back the old "parallel blocks must be contiguous" shape constraint, so the mixed-sequence freedom above stands (serial → parallel → serial stays legal).
|
|
99
|
+
|
|
91
100
|
8. **LESSONS scan**: Grep `.specs/LESSONS.md` for keywords related to planned file paths or actions. If active lessons hit, declare difference or confirm still applies.
|
|
92
101
|
|
|
93
102
|
The full task decomposition protocol, XML template, and constraints are in:
|
|
@@ -102,6 +111,7 @@ This node is truly done when:
|
|
|
102
111
|
- Every `write_files` is strictly within DESIGN.md touched + new modules range (not in forbidden list).
|
|
103
112
|
- Every `[P]` mark has a genuine concurrent companion in its wave with non-overlapping `write_files`; isolated tasks are serial (a fully serial plan is valid when no genuine companion exists).
|
|
104
113
|
- Wave division diagram is clear and has no circular dependencies.
|
|
114
|
+
- Wave numbers agree with dependency levels: each wave is one past the latest dependency wave, and tasks sharing a wave have no dependencies between them (an absent wave section is reported as "not checked", never as a pass; a section whose content cannot be judged blocks a new change and only warns on an older change).
|
|
105
115
|
- Task numbering is continuous.
|
|
106
116
|
|
|
107
117
|
### Red flags
|
|
@@ -110,6 +120,7 @@ This node is truly done when:
|
|
|
110
120
|
- **Agent thought**: "verify: tests should pass" is good enough. **Actual risk**: Vague verify commands cannot be executed. Must be specific: `pytest tests/test_x.py -v` not "tests should pass".
|
|
111
121
|
- **Agent thought**: "write_files can include any file the task might touch." **Actual risk**: Including DESIGN forbidden list modules in write_files bypasses R7.3 + R6.5 boundary enforcement. Strict control is mandatory.
|
|
112
122
|
- **Agent thought**: "Same layer means these tasks can all be marked [P]." **Actual risk**: A same-layer task with no concurrent companion, or with overlapping `write_files`, would be delegated as parallel in violation of this node's rule. Mark `[P]` by companion and write boundaries, never by layer alone.
|
|
123
|
+
- **Agent thought**: "This task just feels like it should run later, so I'll put it in a later wave." **Actual risk**: A wave number is the dependency level projected onto the plan — moving a task to a later wave without declaring the dependency is exactly the mismatch the exit blocks. Declare `depends_on` instead of moving the wave.
|
|
113
124
|
- **Agent thought**: "read_files and write_files are the same thing." **Actual risk**: read_files includes reuse targets and reference modules; write_files is strictly the modification boundary. They serve different purposes (B3 old-project guardrail).
|
|
114
125
|
|
|
115
126
|
## Entry Check
|
|
@@ -131,10 +142,10 @@ The plan node loads `flow-comet-task` to perform task decomposition. It reads DE
|
|
|
131
142
|
**加载声明(阶段层 · 双步硬规则)**:本节点技能已由入口路由经 Skill 工具加载(你正在阅读的就是它);本节点 Required Skill Calls 的加载与声明同样不可跳过:
|
|
132
143
|
|
|
133
144
|
1. **用 Skill 工具加载** `flow-comet-task`(本节点 Required Skill Call)。**不得跳过**——只读取 SKILL.md 文件不叫加载;真正让 flow-comet-task 指令生效的是 Skill 工具把它注入本次会话。
|
|
134
|
-
2.
|
|
145
|
+
2. 加载完成后**立即**运行声明命令(节点退出与证据记录会核对声明标记;声明如实记录加载动作,不等于产出证明)。`--templates` 声明本节点该读的产出物模板(plan → `TASK.md`):
|
|
135
146
|
|
|
136
147
|
```bash
|
|
137
|
-
node .claude/skills/flow-comet/scripts/workflow-state.mjs skill-load plan flow-comet-task --prompt flow-kit/prompts/3-task.md
|
|
148
|
+
node .claude/skills/flow-comet/scripts/workflow-state.mjs skill-load plan flow-comet-task --prompt flow-kit/prompts/3-task.md --templates TASK.md
|
|
138
149
|
```
|
|
139
150
|
|
|
140
151
|
> **跑 skill-load 声明命令 ≠ 加载**:声明只把“哪次会话加载了哪个 skill、按哪份协议工作”写进状态供 exit/record 核对;真正加载只有第 1 步的 Skill 工具能做到。
|
|
@@ -7,7 +7,7 @@ description: "flow-kit REQUIREMENT 阶段协议:AC(Given/When/Then)+ v1·v
|
|
|
7
7
|
|
|
8
8
|
本 Skill 为 flow-comet 的 open 节点提供 flow-kit 的 REQUIREMENT 需求分析。
|
|
9
9
|
|
|
10
|
-
>
|
|
10
|
+
> **模板权威**:本技能产出工件的段形 = **上游模板 + 本仓登记的增量段**——唯一权威 = `flow-kit/templates/**`(段与正文),条件段 / 结构子段 / 增量段与指纹由 `../flow-comet/reference/artifact-templates.json` 登记;`.specs/archive/**` 是历史证据、**不是模板来源**——不得以「上一轮就是这么写的」对齐段形。
|
|
11
11
|
|
|
12
12
|
## 加载
|
|
13
13
|
|