@xulthekl/team-flow 0.41.0 → 0.42.0
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/.claude/always/phase-guard.md +1 -1
- package/.claude-plugin/marketplace.json +1 -1
- package/.claude-plugin/plugin.json +1 -1
- package/.codex-plugin/plugin.json +1 -1
- package/.cursor-plugin/marketplace.json +1 -1
- package/.cursor-plugin/plugin.json +1 -1
- package/.github/plugin/marketplace.json +2 -2
- package/AGENTS.md +1 -0
- package/CHANGELOG.md +11 -1
- package/GEMINI.md +1 -1
- package/INSTALL.md +1 -1
- package/README.md +1 -1
- package/docs/README_en.md +1 -1
- package/gemini-extension.json +1 -1
- package/hooks/session-start +3 -3
- package/llms.txt +1 -1
- package/package.json +1 -1
- package/plugin.json +1 -1
- package/skills/workflow-bootstrap/SKILL.md +21 -9
- package/skills/workflow-bootstrap/references/b1-reconnaissance.md +4 -0
- package/skills/workflow-bootstrap/references/b5-path-decision.md +102 -0
|
@@ -9,7 +9,7 @@
|
|
|
9
9
|
{
|
|
10
10
|
"name": "team-flow",
|
|
11
11
|
"description": "8-state spec workflow + compound global compounding + architecture-design (4A/DDD) + local HTML prototype + product-level orchestration + bootstrap + e2e + session handoff + workflow feedback. 24 skills + 15 agents with embedded TDD, SDD, code review, debugging, delta spec sync, and design-system-driven prototyping.",
|
|
12
|
-
"version": "0.
|
|
12
|
+
"version": "0.42.0",
|
|
13
13
|
"source": "./",
|
|
14
14
|
"author": {
|
|
15
15
|
"name": "LT",
|
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "team-flow",
|
|
3
|
-
"version": "0.
|
|
3
|
+
"version": "0.42.0",
|
|
4
4
|
"description": "Unified workflow plugin: team-flow (spec-driven dev: TDD/SDD/code-review/debugging) + compound-engineering core subset (brainstorm/plan/compound/strategy/ideate/proof, global compounding) + architecture-design (4A+DDD incremental design & global compounding) + prototype (local HTML prototype, zero external deps) + e2e (Playwright E2E, AC-driven) + workflow-orchestrator (product-level workflow orchestration) + workflow-bootstrap (existing project onboarding) + session-handoff (context transfer) + workflow-feedback (issue tracking). 24 skills + 15 agents, one install.",
|
|
5
5
|
"source": "./",
|
|
6
6
|
"author": {
|
|
@@ -2,7 +2,7 @@
|
|
|
2
2
|
"name": "team-flow",
|
|
3
3
|
"displayName": "team-flow",
|
|
4
4
|
"description": "Unified workflow plugin: team-flow (spec-driven dev: TDD/SDD/code-review/debugging) + compound-engineering core subset (brainstorm/plan/compound/strategy/ideate/proof, global compounding) + architecture-design (4A+DDD incremental design & global compounding) + prototype (local HTML prototype, zero external deps) + e2e (Playwright E2E, AC-driven) + workflow-orchestrator (product-level workflow orchestration) + workflow-bootstrap (existing project onboarding) + session-handoff (context transfer) + workflow-feedback (issue tracking). 24 skills + 15 agents, one install.",
|
|
5
|
-
"version": "0.
|
|
5
|
+
"version": "0.42.0",
|
|
6
6
|
"author": {
|
|
7
7
|
"name": "LT",
|
|
8
8
|
"url": "https://github.com/LT"
|
|
@@ -6,13 +6,13 @@
|
|
|
6
6
|
},
|
|
7
7
|
"metadata": {
|
|
8
8
|
"description": "Unified workflow plugins and skills for AI coding agents (team-flow: team-flow + compound + architecture-design + prototype).",
|
|
9
|
-
"version": "0.
|
|
9
|
+
"version": "0.42.0"
|
|
10
10
|
},
|
|
11
11
|
"plugins": [
|
|
12
12
|
{
|
|
13
13
|
"name": "team-flow",
|
|
14
14
|
"description": "Unified workflow with planning artifacts, execution contracts, TDD, review gates, systematic debugging, delta spec sync, architecture-design, and local HTML prototyping.",
|
|
15
|
-
"version": "0.
|
|
15
|
+
"version": "0.42.0",
|
|
16
16
|
"source": ".",
|
|
17
17
|
"author": {
|
|
18
18
|
"name": "LT",
|
package/AGENTS.md
CHANGED
|
@@ -70,6 +70,7 @@ spec 驱动开发过程。8 态变更机:`exploring → specifying → bridgin
|
|
|
70
70
|
### 7. workflow-bootstrap(1 skill,既有项目接入,v0.15.0 脚本+子代理侦察)
|
|
71
71
|
- 一次性"侦察 + 基线建立":代码库侦察 → 架构基线文档化 → 领域词汇提取 → 目录初始化 → CLAUDE.md 初始化(B4.5,v0.41.0)→ 进入 workflow-orchestrator
|
|
72
72
|
- **B1 侦察(v0.15.0)**:先跑确定性脚本 `scripts/recon-probe.sh`(可复现固定化采集,含 Git 子仓库结构 v0.41.0)→ 并行派发 `codebase-recon-analyst` 子代理语义增强(7 维度含 git-structure)→ 主代理汇总写 baseline.md(维度闭合校验 v0.41.0)
|
|
73
|
+
- **B5 路径决策(v0.42.0 降级路径设计)**:两组选项——第一组确保产物可信(A 逆向重建按复杂度分级推荐 / B 交叉验证主代理比对产出 cross-check-report.md / C 真实性核对用户确认门),第二组直接进入流程(D-F 原 A-C 平移)
|
|
73
74
|
- 仅首次使用时执行;`docs/architecture/baseline.md` 已存在则跳过
|
|
74
75
|
|
|
75
76
|
### 8. session-handoff(1 skill,会话交接,v0.16.0 新增)
|
package/CHANGELOG.md
CHANGED
|
@@ -6,7 +6,17 @@ The format loosely follows Keep a Changelog.
|
|
|
6
6
|
|
|
7
7
|
## [Unreleased]
|
|
8
8
|
|
|
9
|
-
### Added(workflow-bootstrap
|
|
9
|
+
### Added(workflow-bootstrap B5 降级路径设计 + session-start 语言要求,v0.42.0)
|
|
10
|
+
|
|
11
|
+
- **B5 路径决策重构(降级路径设计)**:4 平权选项 → 两组降级选项——第一组确保产物可信(A 逆向重建按复杂度分级推荐(≥5 模块或 ≥10 源文件强推)/ B 交叉验证主代理机械比对产出 cross-check-report.md / C 真实性核对用户确认门),第二组直接进入流程(D-F 原 A-C 平移)
|
|
12
|
+
- **references/b5-path-decision.md(新建)**:B5 详细流程(分组展示模板、A 复杂度判据、B 比对清单 + 维度覆盖说明(architecture 显式排除理由)、C 两种模式核对协议(跟随 B / 独立抽查))
|
|
13
|
+
- **b1-reconnaissance.md Success Output 同步**:产出清单增加 cross-check-report.md + 重建建议
|
|
14
|
+
- **hooks/session-start 语言要求**:Inject workflow context 的 msg 追加 `## Language` 指令——要求中文交流,技术术语/代码/命令保持英文
|
|
15
|
+
- **AGENTS.md 同步**:workflow-bootstrap 描述补 B5 降级路径
|
|
16
|
+
|
|
17
|
+
## [0.41.0] - 2026-08-07
|
|
18
|
+
|
|
19
|
+
### Added(workflow-bootstrap CLAUDE.md 初始化 + Git 多子仓库识别)
|
|
10
20
|
|
|
11
21
|
- **B4.5 CLAUDE.md 初始化**(参考 Claude Code /init):将项目基本信息(项目概况/开发命令/工作区结构/Git 结构)+ team-flow 产物结构索引写入工作区 CLAUDE.md;维护原则=索引级+结构级;已有 CLAUDE.md 追加模式(产物结构 + Git 管理规则章节),无则从 `templates/claude-md-team-flow.md` 新建
|
|
12
22
|
- **CLAUDE.md 标准模板**:`skills/workflow-bootstrap/templates/claude-md-team-flow.md`(新建),覆盖 /init 内容 + team-flow 产物结构表(代码强制 + 设计层面)
|
package/GEMINI.md
CHANGED
|
@@ -8,7 +8,7 @@ The workflow is self-contained and does not require OpenSpec or Superpowers at r
|
|
|
8
8
|
|
|
9
9
|
|
|
10
10
|
<!-- team-flow-phase-guard-start -->
|
|
11
|
-
# team-flow v0.
|
|
11
|
+
# team-flow v0.42.0 | 阶段: {{state}} | 工作流: {{workflow}}
|
|
12
12
|
当前阶段允许的操作由 workflow-start 路由规则定义。
|
|
13
13
|
禁止跨越 DP gate 进入下一阶段。变更范围以 execution-contract.md 的 Intent Lock 为准。
|
|
14
14
|
<!-- team-flow-phase-guard-end -->
|
package/INSTALL.md
CHANGED
package/README.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
# team-flow
|
|
2
2
|
|
|
3
|
-
> 当前版本:`v0.
|
|
3
|
+
> 当前版本:`v0.42.0`
|
|
4
4
|
|
|
5
5
|
> 统一插件:**team-flow**(spec 驱动开发)+ **compound-engineering 核心子集**(全局复利)+ **architecture-design**(4A+DDD 增量设计)+ **prototype**(本地 HTML 原型)+ **e2e**(AC 驱动 E2E)+ **workflow-orchestrator**(产品级编排)+ **workflow-bootstrap**(既有项目接入)。一次安装,七套能力协同。
|
|
6
6
|
|
package/docs/README_en.md
CHANGED
|
@@ -126,7 +126,7 @@ npm install -g team-flow
|
|
|
126
126
|
|
|
127
127
|
### Version
|
|
128
128
|
|
|
129
|
-
- Current: `v0.
|
|
129
|
+
- Current: `v0.42.0`
|
|
130
130
|
- v0.9.1 highlights: DP-4 execution-mode recommendations, a portable runtime across 17 platforms, and a raw-package smoke with no plugin-root variable.
|
|
131
131
|
- Self-contained — no OpenSpec or Superpowers runtime required
|
|
132
132
|
- Upstream: [Fission-AI/OpenSpec](https://github.com/Fission-AI/OpenSpec), [obra/superpowers](https://github.com/obra/superpowers)
|
package/gemini-extension.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "team-flow",
|
|
3
3
|
"description": "Unified workflow plugin: team-flow (spec-driven dev) + compound-engineering core subset + architecture-design (4A/DDD) + prototype (local HTML). 24 skills, one install.",
|
|
4
|
-
"version": "0.
|
|
4
|
+
"version": "0.42.0",
|
|
5
5
|
"contextFileName": "GEMINI.md"
|
|
6
6
|
}
|
package/hooks/session-start
CHANGED
|
@@ -1,11 +1,11 @@
|
|
|
1
1
|
#!/usr/bin/env bash
|
|
2
|
-
# v0.
|
|
2
|
+
# v0.42.0: auto-sync CLI version with plugin version
|
|
3
3
|
set -e
|
|
4
4
|
|
|
5
5
|
# ═══════════════════════════════════════════════════════════════
|
|
6
6
|
# Plugin version (update this when releasing new versions)
|
|
7
7
|
# ═══════════════════════════════════════════════════════════════
|
|
8
|
-
PLUGIN_VERSION="0.
|
|
8
|
+
PLUGIN_VERSION="0.42.0"
|
|
9
9
|
|
|
10
10
|
# ═══════════════════════════════════════════════════════════════
|
|
11
11
|
# Step 1: Auto-sync CLI version with plugin version
|
|
@@ -45,7 +45,7 @@ sync_tf_version
|
|
|
45
45
|
# Step 2: Inject workflow context
|
|
46
46
|
# ═══════════════════════════════════════════════════════════════
|
|
47
47
|
|
|
48
|
-
msg="<EXTREMELY_IMPORTANT>\nYou have team-flow installed. Invoke /team-flow:workflow-start or /team-flow:workflow-orchestrator ONLY when active team-flow artifacts exist (\`.team-flow.yaml\`, \`proposal.md\`, \`execution-contract.md\`, \`specs/\`) or the user explicitly requests it. For ordinary coding tasks, do NOT invoke them.\n</EXTREMELY_IMPORTANT>\n<EXTREMELY_IMPORTANT>\n## Workspace Directory Rule\n\"Base directory for this skill\" = skill source location (read SKILL.md/references/ ONLY). User project = \`pwd\`. ALL project artifacts (changes/, prd/, docs/, .team-flow/, etc.) MUST operate under \`pwd\`, NEVER the skill base directory. Run \`pwd\` first in every skill execution.\n</EXTREMELY_IMPORTANT>"
|
|
48
|
+
msg="<EXTREMELY_IMPORTANT>\nYou have team-flow installed. Invoke /team-flow:workflow-start or /team-flow:workflow-orchestrator ONLY when active team-flow artifacts exist (\`.team-flow.yaml\`, \`proposal.md\`, \`execution-contract.md\`, \`specs/\`) or the user explicitly requests it. For ordinary coding tasks, do NOT invoke them.\n</EXTREMELY_IMPORTANT>\n<EXTREMELY_IMPORTANT>\n## Workspace Directory Rule\n\"Base directory for this skill\" = skill source location (read SKILL.md/references/ ONLY). User project = \`pwd\`. ALL project artifacts (changes/, prd/, docs/, .team-flow/, etc.) MUST operate under \`pwd\`, NEVER the skill base directory. Run \`pwd\` first in every skill execution.\n</EXTREMELY_IMPORTANT>\n<EXTREMELY_IMPORTANT>\n## Language\nCommunicate with the user in Chinese (中文). Technical terms, code, and commands remain in English.\n</EXTREMELY_IMPORTANT>"
|
|
49
49
|
|
|
50
50
|
# Three platforms share the same message, only output format differs.
|
|
51
51
|
if [ -n "${CURSOR_PLUGIN_ROOT:-}" ]; then
|
package/llms.txt
CHANGED
|
@@ -3,7 +3,7 @@
|
|
|
3
3
|
## Overview
|
|
4
4
|
spec-superflow is a self-contained workflow integration plugin for Claude Code, Cursor, OpenAI Codex CLI/App, GitHub Copilot CLI, Gemini CLI, OpenCode, WorkBuddy, and Trae. It merges spec-driven planning artifacts (proposal, specs, design, tasks) with disciplined execution guardrails (TDD, review gates, controlled handoff) into one unified workflow.
|
|
5
5
|
|
|
6
|
-
Current version: v0.
|
|
6
|
+
Current version: v0.42.0.
|
|
7
7
|
|
|
8
8
|
## Key Documents
|
|
9
9
|
- README.md: Chinese homepage with full usage guide and FAQ
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@xulthekl/team-flow",
|
|
3
|
-
"version": "0.
|
|
3
|
+
"version": "0.42.0",
|
|
4
4
|
"description": "Unified plugin (24 skills + 15 agents) integrating team-flow, compound-engineering, architecture-design, prototype, design-system, workflow-orchestrator, workflow-bootstrap, e2e, session-handoff, workflow-feedback for multi-agent coding tools.",
|
|
5
5
|
"main": "dist/index.js",
|
|
6
6
|
"types": "dist/index.d.ts",
|
package/plugin.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "team-flow",
|
|
3
|
-
"version": "0.
|
|
3
|
+
"version": "0.42.0",
|
|
4
4
|
"description": "Unified workflow plugin: team-flow (spec-driven dev: TDD/SDD/code-review/debugging) + compound-engineering core subset (brainstorm/plan/compound/strategy/ideate/proof, global compounding) + architecture-design (4A+DDD incremental design & global compounding) + prototype (local HTML prototype, zero external deps) + e2e (Playwright E2E, AC-driven) + workflow-orchestrator (product-level workflow orchestration) + workflow-bootstrap (existing project onboarding) + session-handoff (context transfer) + workflow-feedback (issue tracking). 24 skills + 15 agents, one install.",
|
|
5
5
|
"author": {
|
|
6
6
|
"name": "LT"
|
|
@@ -209,15 +209,25 @@ Step 2: 确定性提取(兜底)
|
|
|
209
209
|
|
|
210
210
|
**不做的事**:不写 SOP 流程(由 skill 负责)、不写变更级状态机(由 workflow skill 负责)、不列可自动发现的文件清单。
|
|
211
211
|
|
|
212
|
-
### B5: Path Decision
|
|
212
|
+
### B5: Path Decision(路径判断,v0.42.0 重构:降级路径设计)
|
|
213
213
|
|
|
214
|
-
|
|
214
|
+
询问用户。**分组展示**——用户先决定"要不要把地基打牢"(第一组),再决定"打完地基做什么"(第二组):
|
|
215
215
|
|
|
216
|
-
>
|
|
217
|
-
> - A)
|
|
218
|
-
> - B)
|
|
219
|
-
> - C)
|
|
220
|
-
>
|
|
216
|
+
> **第一组:确保 bootstrap 产物可信**
|
|
217
|
+
> - A) 逆向重建完整架构(**复杂项目强推 / 简单项目弱推**;→ 进入 workflow-orchestrator ARCH 阶段 reconstruction 模式,arch-reverse-analyst,重建 PASS 后 `tf arch init --mode reconstruction --baseline-ref prd/vN/` 打戳,全量命令见 b5-path-decision.md)
|
|
218
|
+
> - B) 交叉验证与对齐 bootstrap 产物(→ 主代理机械比对,产出 `docs/architecture/cross-check-report.md` 疑点清单)
|
|
219
|
+
> - C) 真实性核对(→ 针对 B 疑点清单(跟随 B)或 baseline.md 语义结论抽查(独立),用户逐条确认,修正 baseline.md)
|
|
220
|
+
>
|
|
221
|
+
> **第二组:直接进入 team-flow 流程**
|
|
222
|
+
> - D) 我有一个具体的功能需求(→ 进入 workflow-orchestrator S1,S3 后进 ARCH 产品级架构设计)
|
|
223
|
+
> - E) 我有一个模糊的产品方向(→ ce-brainstorm)
|
|
224
|
+
> - F) 我只是想建立架构基线,暂不开发新需求(→ 结束)
|
|
225
|
+
|
|
226
|
+
**A 分级推荐判据**(基于 B1 侦察结果):
|
|
227
|
+
- 复杂项目(≥5 模块 或 ≥10 源文件,复用 B2 触发判据)→ 提示"建议优先重建完整架构"
|
|
228
|
+
- 简单项目(<5 模块 且 <10 源文件)→ 提示"项目较简单,可直接进入流程,重建为可选项"
|
|
229
|
+
|
|
230
|
+
**B/C 串联闭环**:选 B → 主代理产出疑点清单 → 提示是否进入 C(是 → 逐条 AskUserQuestion 核对并修正 baseline.md;否 → 疑点留档标记 deferred,不阻断)。
|
|
221
231
|
|
|
222
232
|
**逆向重建衔接(v0.36.3)**:ARCH 阶段 reconstruction 模式复用本 bootstrap 的 recon-probe.sh(--ddl-out)+ codebase-recon-analyst,并派发 `arch-reverse-analyst` 补 BC/状态/指令事件维度(见 `references/agents/arch-reverse-analyst.md`)。
|
|
223
233
|
|
|
@@ -225,6 +235,8 @@ Step 2: 确定性提取(兜底)
|
|
|
225
235
|
- 将 `docs/architecture/baseline.md` 和 `docs/architecture/CONCEPTS.md` 的路径传递给 workflow-orchestrator S1
|
|
226
236
|
- 作为后续流程的架构上下文
|
|
227
237
|
|
|
238
|
+
详细流程(分组展示模板、A 复杂度判据来源、B 7 类比对清单 + report 格式、C 核对协议)见 `references/b5-path-decision.md`。
|
|
239
|
+
|
|
228
240
|
## Guardrails
|
|
229
241
|
|
|
230
242
|
- **一次性**:bootstrap 只在首次接入时执行。如果 `docs/architecture/baseline.md` 已存在,提示用户"基线已建立,是否要重新侦察?"
|
|
@@ -238,9 +250,9 @@ Step 2: 确定性提取(兜底)
|
|
|
238
250
|
|
|
239
251
|
每次交互结束时说明:
|
|
240
252
|
1. 已完成的阶段(B1-B1.5-B2-B3-B4-B4.5-B5)
|
|
241
|
-
2. 已产出的制品(baseline.md / conventions / ARCHITECTURE.md / PHYSICAL-MODEL.md / schema-baseline.sql / API-INDEX.md / INDEX.md / CONCEPTS.md / CLAUDE.md / 目录结构)
|
|
253
|
+
2. 已产出的制品(baseline.md / conventions / ARCHITECTURE.md / PHYSICAL-MODEL.md / schema-baseline.sql / API-INDEX.md / INDEX.md / CONCEPTS.md / CLAUDE.md / cross-check-report.md(如执行 B)/ 目录结构)
|
|
242
254
|
3. 下一步建议(调用哪个 skill)
|
|
243
255
|
|
|
244
256
|
## Success Output
|
|
245
257
|
|
|
246
|
-
接入结束时输出:侦察模式、产出清单(baseline.md / conventions / ARCHITECTURE.md / DATABASE.md / PHYSICAL-MODEL.md / schema-baseline.sql / API-INDEX.md / INDEX.md / docs/architecture/CONCEPTS.md / CLAUDE.md / 目录结构)、关键发现(技术栈/模块数/架构模式/测试覆盖)、下一步建议。完整模板见 `references/b1-reconnaissance.md`「Success Output 模板」。
|
|
258
|
+
接入结束时输出:侦察模式、产出清单(baseline.md / conventions / ARCHITECTURE.md / DATABASE.md / PHYSICAL-MODEL.md / schema-baseline.sql / API-INDEX.md / INDEX.md / docs/architecture/CONCEPTS.md / CLAUDE.md / cross-check-report.md(如执行 B)/ 目录结构)、关键发现(技术栈/模块数/架构模式/测试覆盖)、下一步建议。完整模板见 `references/b1-reconnaissance.md`「Success Output 模板」。
|
|
@@ -111,6 +111,10 @@ bash ${CLAUDE_PLUGIN_ROOT}/skills/workflow-bootstrap/scripts/recon-probe.sh --ro
|
|
|
111
111
|
- 测试覆盖:约 60%(JUnit 5 + Mockito)
|
|
112
112
|
- ⚠ INCOMPLETE:<缺失维度名>(如 api-surface——Deep 模式未派发)
|
|
113
113
|
|
|
114
|
+
建议:
|
|
115
|
+
→ 复杂项目(≥5 模块 或 ≥10 源文件)优先逆向重建完整架构
|
|
116
|
+
→ 交叉验证与真实性核对:docs/architecture/cross-check-report.md(如执行 B)
|
|
117
|
+
|
|
114
118
|
下一步:
|
|
115
119
|
→ 运行 /workflow-orchestrator 开始产品级流程
|
|
116
120
|
→ 或运行 /architecture-design 进行完整架构设计
|
|
@@ -0,0 +1,102 @@
|
|
|
1
|
+
# B5 路径决策详细流程(v0.42.0:降级路径设计)
|
|
2
|
+
|
|
3
|
+
> 主代理在 B5 询问用户时展示分组选项,并按用户选择执行 A/B/C 或进入第二组流程。
|
|
4
|
+
|
|
5
|
+
## 分组展示模板
|
|
6
|
+
|
|
7
|
+
询问用户时,按以下模板展示(**两组分开展示**):
|
|
8
|
+
|
|
9
|
+
```
|
|
10
|
+
第一组:确保 bootstrap 产物可信
|
|
11
|
+
A 逆向重建完整架构
|
|
12
|
+
B 交叉验证与对齐 bootstrap 产物
|
|
13
|
+
C 真实性核对
|
|
14
|
+
第二组:直接进入 team-flow 流程
|
|
15
|
+
D 我有一个具体的功能需求
|
|
16
|
+
E 我有一个模糊的产品方向
|
|
17
|
+
F 仅建立基线,暂不开发
|
|
18
|
+
```
|
|
19
|
+
|
|
20
|
+
**多选**:第一组可多选(如 A + B),第二组单选。第一组优先——用户选第二组前,主代理应提示"bootstrap 产物未经交叉验证,直接进入流程可能有偏差"(advisory,不阻断)。
|
|
21
|
+
|
|
22
|
+
## A 逆向重建:复杂度分级推荐
|
|
23
|
+
|
|
24
|
+
**判据来源**(复用 B1 侦察结果,无需新采集):
|
|
25
|
+
- 模块数:codebase-recon-analyst modules 维度的模块计数
|
|
26
|
+
- 源文件数:recon-probe JSON `loc.total_files`
|
|
27
|
+
|
|
28
|
+
**分级推荐**:
|
|
29
|
+
|
|
30
|
+
| 条件 | 推荐措辞 |
|
|
31
|
+
|------|---------|
|
|
32
|
+
| ≥5 模块 或 ≥10 源文件(复杂) | "建议优先重建完整架构——项目有实质架构,基线快照不足以支撑后续设计" |
|
|
33
|
+
| <5 模块 且 <10 源文件(简单) | "项目较简单,可直接进入流程,逆向重建为可选项" |
|
|
34
|
+
|
|
35
|
+
**去向**:A → workflow-orchestrator ARCH 阶段 reconstruction 模式(arch-reverse-analyst,`references/agents/arch-reverse-analyst.md`)→ architecture-reviewer product 视角评审(≤3 轮循环)→ PASS 后 `tf arch init --mode reconstruction --baseline-ref prd/vN/` 打戳。
|
|
36
|
+
|
|
37
|
+
## B 交叉验证与对齐:主代理机械比对
|
|
38
|
+
|
|
39
|
+
> B1 刚结束主代理手里材料最全(recon-probe JSON + 各维度摘要 + baseline.md),交叉验证是机械比对,**主代理直接执行,不派发子代理**(LT 决策 v0.42.0)。
|
|
40
|
+
|
|
41
|
+
**比对原则**:确定性数据(脚本机械采集)为锚,语义结论(子代理 LLM 推断)为被检对象。只标记冲突,不重跑侦察。
|
|
42
|
+
|
|
43
|
+
**比对清单**:
|
|
44
|
+
|
|
45
|
+
| # | 比对项 | 确定性证据(脚本字段) | 语义结论(子代理维度) |
|
|
46
|
+
|---|--------|----------------------|----------------------|
|
|
47
|
+
| 1 | 技术栈 | `dependencies.manifests` | technology |
|
|
48
|
+
| 2 | 模块边界 | `directory_tree.dirs` | modules |
|
|
49
|
+
| 3 | 数据模型 | `ddl_extraction.tables` | data-model |
|
|
50
|
+
| 4 | API 表面 | `directory_tree.dirs` 按 controller/router/route/api 目录名过滤(best-effort) | api-surface |
|
|
51
|
+
| 5 | 测试现状 | `tests.test_file_count` | test-docs |
|
|
52
|
+
| 6 | Git 结构 | `git_structure` | git-structure |
|
|
53
|
+
| 7 | 文档声明 | `docs.doc_files` | baseline.md 中引用的文档 |
|
|
54
|
+
|
|
55
|
+
**维度覆盖说明**:
|
|
56
|
+
- **architecture 维度不纳入机械比对**(显式排除):架构模式(DDD/MVC/微服务/单体)是纯语义判定,无确定性锚点。其验证交由 A 重建时的 arch-reverse-analyst 深化 + architecture-reviewer 审查,B 阶段不机械比对。
|
|
57
|
+
- **API 表面无独立确定性扫描字段**(recon-probe.sh 无 controller/router 采集),第 4 项降级为 best-effort 语义比对——从 `directory_tree.dirs` 过滤 controller/router/route/api 目录名作锚,若目录树无明显 API 目录则标注"无锚点,跳过该项"。
|
|
58
|
+
|
|
59
|
+
**疑点判定**:确定性证据存在但语义结论未反映 → 疑点(如依赖清单有 spring-boot 但技术栈未提);语义结论声称但确定性证据缺失 → 疑点(如测试覆盖声称 60% 但 test_file_count=0)。
|
|
60
|
+
|
|
61
|
+
**输出**:`docs/architecture/cross-check-report.md`
|
|
62
|
+
|
|
63
|
+
```markdown
|
|
64
|
+
# Bootstrap Cross-Check Report
|
|
65
|
+
|
|
66
|
+
> 由 workflow-bootstrap B5-B 主代理机械比对生成(v0.42.0)。确定性数据为锚,语义结论为被检对象。
|
|
67
|
+
|
|
68
|
+
## 疑点清单
|
|
69
|
+
|
|
70
|
+
| # | 比对项 | 确定性证据 | 语义结论 | 冲突描述 | 建议动作 |
|
|
71
|
+
|---|--------|-----------|---------|---------|---------|
|
|
72
|
+
| 1 | 技术栈 | ... | ... | ... | ... |
|
|
73
|
+
|
|
74
|
+
## 结论
|
|
75
|
+
|
|
76
|
+
cross-check: <PASS | N/A(未选 B)| deferred(疑点留档)>
|
|
77
|
+
```
|
|
78
|
+
|
|
79
|
+
不阻断(advisory)。`cross-check: deferred` 时提示用户可后续补核对。
|
|
80
|
+
|
|
81
|
+
## C 真实性核对:用户确认门
|
|
82
|
+
|
|
83
|
+
**触发**:用户选 C,或选 B 后主代理提示"是否进入 C 真实性核对?"。
|
|
84
|
+
|
|
85
|
+
**两种模式**:
|
|
86
|
+
- **C 跟随 B(推荐)**:针对 cross-check-report.md 疑点清单,主代理用 AskUserQuestion 逐条(或按比对项分组)确认:
|
|
87
|
+
- **C 独立(未选 B)**:无 cross-check-report.md 可核对,主代理对 baseline.md 的全部语义结论做真实性抽查(每维度挑 1-2 个代表性结论),用户逐条确认"属实 / 修正 / 忽略"。
|
|
88
|
+
|
|
89
|
+
**确认流程**(两模式共用):
|
|
90
|
+
|
|
91
|
+
| 用户选择 | 动作 |
|
|
92
|
+
|---------|------|
|
|
93
|
+
| 确认(属实) | 修正 baseline.md 对应内容,provenance 标注 `user-confirmed` |
|
|
94
|
+
| 忽略 | 保留疑点,标记 `deferred`,不阻断 |
|
|
95
|
+
| 补充说明 | 按用户描述修正,provenance 标注 `user-confirmed` |
|
|
96
|
+
|
|
97
|
+
**核对完成**:更新 baseline.md 顶部校验状态 `cross-check: PASS`(无剩余疑点)或 `deferred`(有忽略项)。
|
|
98
|
+
|
|
99
|
+
## 与其他选项关系
|
|
100
|
+
|
|
101
|
+
- A(重建)与 B(交叉验证)可组合:先 B 校验基线可信,再 A 重建——重建消费的是校验后的基线,降低重建偏差
|
|
102
|
+
- D/E/F 为原 A/B/C 平移,语义不变,仅位置后移
|