maestro-flow 0.5.33 → 0.5.34
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/.agents/agents/role-design-author.md +5 -3
- package/.agents/skills/maestro/SKILL.md +8 -1
- package/.agents/skills/maestro-help/SKILL.md +39 -10
- package/.agents/skills/maestro-help/index/catalog.json +118 -53
- package/.agents/skills/maestro-help/phases/01-parse-intent.md +9 -6
- package/.agents/skills/maestro-help/phases/02-search-present.md +55 -15
- package/.agents/skills/maestro-help/phases/03-workflow-guide.md +35 -2
- package/.agents/skills/maestro-ralph/SKILL.md +9 -2
- package/.agents/skills/maestro-ralph-beta/SKILL.md +7 -2
- package/.agents/skills/maestro-ralph-execute/SKILL.md +1 -0
- package/.agents/skills/manage-harvest/SKILL.md +3 -0
- package/.agents/skills/manage-knowledge-audit/SKILL.md +29 -0
- package/.agents/skills/odyssey-debug/SKILL.md +7 -1
- package/.agents/skills/odyssey-improve/SKILL.md +7 -1
- package/.agents/skills/odyssey-planex/SKILL.md +11 -1
- package/.agents/skills/odyssey-review-test-fix/SKILL.md +7 -1
- package/.agents/skills/odyssey-ui/SKILL.md +7 -1
- package/.agents/skills/quality-review/SKILL.md +3 -0
- package/.agy/agents/role-design-author.md +5 -3
- package/.agy/skills/maestro/SKILL.md +8 -1
- package/.agy/skills/maestro-help/SKILL.md +39 -10
- package/.agy/skills/maestro-help/index/catalog.json +118 -53
- package/.agy/skills/maestro-help/phases/01-parse-intent.md +9 -6
- package/.agy/skills/maestro-help/phases/02-search-present.md +55 -15
- package/.agy/skills/maestro-help/phases/03-workflow-guide.md +35 -2
- package/.agy/skills/maestro-ralph/SKILL.md +9 -2
- package/.agy/skills/maestro-ralph-beta/SKILL.md +7 -2
- package/.agy/skills/maestro-ralph-execute/SKILL.md +1 -0
- package/.agy/skills/manage-harvest/SKILL.md +3 -0
- package/.agy/skills/manage-knowledge-audit/SKILL.md +29 -0
- package/.agy/skills/odyssey-debug/SKILL.md +7 -1
- package/.agy/skills/odyssey-improve/SKILL.md +7 -1
- package/.agy/skills/odyssey-planex/SKILL.md +11 -1
- package/.agy/skills/odyssey-review-test-fix/SKILL.md +7 -1
- package/.agy/skills/odyssey-ui/SKILL.md +7 -1
- package/.agy/skills/quality-review/SKILL.md +3 -0
- package/.claude/agents/role-design-author.md +5 -3
- package/.claude/commands/maestro-ralph-beta.md +7 -2
- package/.claude/commands/maestro-ralph-execute.md +1 -0
- package/.claude/commands/maestro-ralph.md +9 -2
- package/.claude/commands/maestro.md +8 -1
- package/.claude/commands/manage-harvest.md +3 -0
- package/.claude/commands/manage-knowledge-audit.md +29 -0
- package/.claude/commands/odyssey-debug.md +7 -1
- package/.claude/commands/odyssey-improve.md +7 -1
- package/.claude/commands/odyssey-planex.md +11 -1
- package/.claude/commands/odyssey-review-test-fix.md +7 -1
- package/.claude/commands/odyssey-ui.md +7 -1
- package/.claude/commands/quality-review.md +3 -0
- package/.claude/skills/maestro-help/SKILL.md +39 -10
- package/.claude/skills/maestro-help/index/catalog.json +118 -53
- package/.claude/skills/maestro-help/phases/01-parse-intent.md +9 -6
- package/.claude/skills/maestro-help/phases/02-search-present.md +55 -15
- package/.claude/skills/maestro-help/phases/03-workflow-guide.md +35 -2
- package/.codex/skills/manage-harvest/SKILL.md +2 -1
- package/.codex/skills/quality-review/SKILL.md +4 -2
- package/README.md +196 -87
- package/README.zh-CN.md +198 -79
- package/dashboard/dist-server/dashboard/src/server/wiki/graph-analysis.js +9 -4
- package/dashboard/dist-server/dashboard/src/server/wiki/graph-analysis.js.map +1 -1
- package/dashboard/dist-server/dashboard/src/server/wiki/search.d.ts +8 -5
- package/dashboard/dist-server/dashboard/src/server/wiki/search.js +22 -42
- package/dashboard/dist-server/dashboard/src/server/wiki/search.js.map +1 -1
- package/dashboard/dist-server/dashboard/src/server/wiki/wiki-indexer.d.ts +1 -4
- package/dashboard/dist-server/dashboard/src/server/wiki/wiki-indexer.js +28 -42
- package/dashboard/dist-server/dashboard/src/server/wiki/wiki-indexer.js.map +1 -1
- package/dashboard/dist-server/dashboard/src/server/wiki/wiki-indexer.test.js +22 -22
- package/dashboard/dist-server/dashboard/src/server/wiki/wiki-indexer.test.js.map +1 -1
- package/dashboard/dist-server/dashboard/src/server/wiki/writer-stress.test.js +12 -12
- package/dashboard/dist-server/dashboard/src/server/wiki/writer-stress.test.js.map +1 -1
- package/dashboard/dist-server/dashboard/src/server/wiki/writer.js +10 -3
- package/dashboard/dist-server/dashboard/src/server/wiki/writer.js.map +1 -1
- package/dashboard/dist-server/src/graph/kg/db/types.d.ts +2 -1
- package/dashboard/dist-server/src/graph/kg/db/types.js.map +1 -1
- package/dashboard/dist-server/src/graph/kg/extraction/knowledge-extractor-registry.d.ts +22 -0
- package/dashboard/dist-server/src/graph/kg/extraction/knowledge-extractor-registry.js +64 -0
- package/dashboard/dist-server/src/graph/kg/extraction/knowledge-extractor-registry.js.map +1 -0
- package/dashboard/dist-server/src/graph/kg/extraction/orchestrator.js +12 -98
- package/dashboard/dist-server/src/graph/kg/extraction/orchestrator.js.map +1 -1
- package/dashboard/dist-server/src/hooks/keyword-spec-injector.js +5 -2
- package/dashboard/dist-server/src/hooks/keyword-spec-injector.js.map +1 -1
- package/dashboard/dist-server/src/hooks/wiki-role-loader.js +2 -1
- package/dashboard/dist-server/src/hooks/wiki-role-loader.js.map +1 -1
- package/dashboard/dist-server/src/tools/domain-loader.js +22 -16
- package/dashboard/dist-server/src/tools/domain-loader.js.map +1 -1
- package/dashboard/dist-server/src/tools/spec-entry-parser.d.ts +6 -1
- package/dashboard/dist-server/src/tools/spec-entry-parser.js +44 -13
- package/dashboard/dist-server/src/tools/spec-entry-parser.js.map +1 -1
- package/dashboard/dist-server/src/tools/spec-keyword-index.d.ts +2 -0
- package/dashboard/dist-server/src/tools/spec-keyword-index.js +4 -9
- package/dashboard/dist-server/src/tools/spec-keyword-index.js.map +1 -1
- package/dashboard/dist-server/src/tools/spec-loader.js +3 -11
- package/dashboard/dist-server/src/tools/spec-loader.js.map +1 -1
- package/dashboard/dist-server/src/types/index.d.ts +14 -0
- package/dashboard/dist-server/src/utils/frontmatter.d.ts +15 -0
- package/dashboard/dist-server/src/utils/frontmatter.js +76 -0
- package/dashboard/dist-server/src/utils/frontmatter.js.map +1 -0
- package/dist/src/commands/domain.d.ts.map +1 -1
- package/dist/src/commands/domain.js +8 -2
- package/dist/src/commands/domain.js.map +1 -1
- package/dist/src/commands/install.d.ts.map +1 -1
- package/dist/src/commands/install.js +6 -2
- package/dist/src/commands/install.js.map +1 -1
- package/dist/src/commands/knowhow.d.ts.map +1 -1
- package/dist/src/commands/knowhow.js +3 -26
- package/dist/src/commands/knowhow.js.map +1 -1
- package/dist/src/commands/search.d.ts +1 -1
- package/dist/src/commands/search.d.ts.map +1 -1
- package/dist/src/commands/search.js +43 -41
- package/dist/src/commands/search.js.map +1 -1
- package/dist/src/commands/spec.d.ts.map +1 -1
- package/dist/src/commands/spec.js +106 -6
- package/dist/src/commands/spec.js.map +1 -1
- package/dist/src/commands/uninstall.d.ts.map +1 -1
- package/dist/src/commands/uninstall.js +12 -3
- package/dist/src/commands/uninstall.js.map +1 -1
- package/dist/src/config/cli-tools-config.d.ts.map +1 -1
- package/dist/src/config/cli-tools-config.js +2 -0
- package/dist/src/config/cli-tools-config.js.map +1 -1
- package/dist/src/core/extension-loader.d.ts.map +1 -1
- package/dist/src/core/extension-loader.js +4 -0
- package/dist/src/core/extension-loader.js.map +1 -1
- package/dist/src/core/install-executor.d.ts.map +1 -1
- package/dist/src/core/install-executor.js +2 -0
- package/dist/src/core/install-executor.js.map +1 -1
- package/dist/src/graph/analyzers/fs-analyzer.js +2 -2
- package/dist/src/graph/analyzers/fs-analyzer.js.map +1 -1
- package/dist/src/graph/kg/db/node-kind-registry.d.ts +18 -0
- package/dist/src/graph/kg/db/node-kind-registry.d.ts.map +1 -0
- package/dist/src/graph/kg/db/node-kind-registry.js +42 -0
- package/dist/src/graph/kg/db/node-kind-registry.js.map +1 -0
- package/dist/src/graph/kg/db/types.d.ts +2 -1
- package/dist/src/graph/kg/db/types.d.ts.map +1 -1
- package/dist/src/graph/kg/db/types.js.map +1 -1
- package/dist/src/graph/kg/extraction/knowledge-extractor-registry.d.ts +23 -0
- package/dist/src/graph/kg/extraction/knowledge-extractor-registry.d.ts.map +1 -0
- package/dist/src/graph/kg/extraction/knowledge-extractor-registry.js +64 -0
- package/dist/src/graph/kg/extraction/knowledge-extractor-registry.js.map +1 -0
- package/dist/src/graph/kg/extraction/orchestrator.d.ts.map +1 -1
- package/dist/src/graph/kg/extraction/orchestrator.js +12 -98
- package/dist/src/graph/kg/extraction/orchestrator.js.map +1 -1
- package/dist/src/hooks/keyword-spec-injector.d.ts.map +1 -1
- package/dist/src/hooks/keyword-spec-injector.js +5 -2
- package/dist/src/hooks/keyword-spec-injector.js.map +1 -1
- package/dist/src/hooks/wiki-role-loader.d.ts.map +1 -1
- package/dist/src/hooks/wiki-role-loader.js +2 -1
- package/dist/src/hooks/wiki-role-loader.js.map +1 -1
- package/dist/src/index.d.ts +6 -0
- package/dist/src/index.d.ts.map +1 -1
- package/dist/src/index.js +4 -0
- package/dist/src/index.js.map +1 -1
- package/dist/src/tools/domain-loader.d.ts.map +1 -1
- package/dist/src/tools/domain-loader.js +22 -16
- package/dist/src/tools/domain-loader.js.map +1 -1
- package/dist/src/tools/spec-conflict-marker.d.ts +55 -0
- package/dist/src/tools/spec-conflict-marker.d.ts.map +1 -0
- package/dist/src/tools/spec-conflict-marker.js +228 -0
- package/dist/src/tools/spec-conflict-marker.js.map +1 -0
- package/dist/src/tools/spec-entry-parser.d.ts +6 -1
- package/dist/src/tools/spec-entry-parser.d.ts.map +1 -1
- package/dist/src/tools/spec-entry-parser.js +44 -13
- package/dist/src/tools/spec-entry-parser.js.map +1 -1
- package/dist/src/tools/spec-keyword-index.d.ts +2 -0
- package/dist/src/tools/spec-keyword-index.d.ts.map +1 -1
- package/dist/src/tools/spec-keyword-index.js +4 -9
- package/dist/src/tools/spec-keyword-index.js.map +1 -1
- package/dist/src/tools/spec-loader.d.ts.map +1 -1
- package/dist/src/tools/spec-loader.js +3 -11
- package/dist/src/tools/spec-loader.js.map +1 -1
- package/dist/src/tools/spec-writer.d.ts.map +1 -1
- package/dist/src/tools/spec-writer.js +2 -6
- package/dist/src/tools/spec-writer.js.map +1 -1
- package/dist/src/tools/store-knowhow.d.ts.map +1 -1
- package/dist/src/tools/store-knowhow.js +6 -52
- package/dist/src/tools/store-knowhow.js.map +1 -1
- package/dist/src/tui/uninstall-ui/UninstallFlow.d.ts.map +1 -1
- package/dist/src/tui/uninstall-ui/UninstallFlow.js +3 -1
- package/dist/src/tui/uninstall-ui/UninstallFlow.js.map +1 -1
- package/dist/src/types/index.d.ts +14 -0
- package/dist/src/types/index.d.ts.map +1 -1
- package/dist/src/utils/frontmatter.d.ts +16 -0
- package/dist/src/utils/frontmatter.d.ts.map +1 -0
- package/dist/src/utils/frontmatter.js +76 -0
- package/dist/src/utils/frontmatter.js.map +1 -0
- package/dist/src/utils/migration-registry.d.ts.map +1 -1
- package/dist/src/utils/migration-registry.js +14 -1
- package/dist/src/utils/migration-registry.js.map +1 -1
- package/package.json +2 -2
- package/workflows/claude-instructions.md +48 -45
- package/workflows/codex-instructions.md +25 -19
|
@@ -123,18 +123,45 @@ const taskPatterns = [
|
|
|
123
123
|
4. `/quality-auto-test [phase] --re-run` — 重跑失败场景
|
|
124
124
|
```
|
|
125
125
|
|
|
126
|
+
**Odyssey 长周期循环** (深度自主):
|
|
127
|
+
|
|
128
|
+
```markdown
|
|
129
|
+
## Odyssey 长周期循环
|
|
130
|
+
|
|
131
|
+
适用于需要深度考古、多轮诊断/修复/验证、知识泛化和持久化的复杂任务。
|
|
132
|
+
每个 odyssey 命令都是自含闭环,自主循环直到完成。
|
|
133
|
+
|
|
134
|
+
### 调试类
|
|
135
|
+
1. `/odyssey-debug "问题描述"` — 考古→诊断→修复→确认→泛化→知识持久化
|
|
136
|
+
|
|
137
|
+
### 代码改进
|
|
138
|
+
1. `/odyssey-improve "改进目标"` — 多维审计→深度诊断→定向修复→验证→泛化
|
|
139
|
+
|
|
140
|
+
### 审查修复
|
|
141
|
+
1. `/odyssey-review-test-fix "审查范围"` — 考古→探索→多维审查→修复→泛化
|
|
142
|
+
|
|
143
|
+
### 需求迭代
|
|
144
|
+
1. `/odyssey-planex "需求描述"` — 计划→执行→严格验证→修复循环(直到验收通过)
|
|
145
|
+
|
|
146
|
+
### UI 优化
|
|
147
|
+
1. `/odyssey-ui "优化目标"` — 视觉调研→多维审计→发散探索→修复→验证
|
|
148
|
+
```
|
|
149
|
+
|
|
126
150
|
#### Step 4.3: 工作流全景图
|
|
127
151
|
|
|
128
152
|
对需要全景视角的用户,展示 Mermaid 图:
|
|
129
153
|
|
|
130
154
|
```
|
|
131
|
-
上游起源: brainstorm(发散) | blueprint(收敛) → 可选
|
|
155
|
+
上游起源: brainstorm(发散) | blueprint(收敛) | grill(压力测试) → 可选
|
|
132
156
|
理解层: analyze "topic"(宏观) → scope_verdict 路由
|
|
133
157
|
编排层: roadmap(可选,仅 scope_verdict=large 时建议)
|
|
134
158
|
执行层: plan → execute → quality → milestone-audit → milestone-complete
|
|
135
159
|
快速渠道: maestro-quick → (直接完成)
|
|
136
160
|
Issue 闭环: discover → create → analyze --gaps → plan --gaps → execute → close
|
|
137
|
-
全自动:
|
|
161
|
+
全自动: /maestro -y → (自动路由)
|
|
162
|
+
Odyssey: odyssey-debug / odyssey-improve / odyssey-planex / odyssey-ui → (自主循环)
|
|
163
|
+
并行加速: swarm-workflow / universal-workflow → (多 agent 并发)
|
|
164
|
+
智能导航: /maestro-next → (检测状态推荐下一步)
|
|
138
165
|
```
|
|
139
166
|
|
|
140
167
|
#### Step 4.4: 关键概念说明
|
|
@@ -156,6 +183,11 @@ Issue 闭环: discover → create → analyze --gaps → plan --gaps → execute
|
|
|
156
183
|
- **Spec**: 项目约束规则(coding/arch/debug/test),自动注入到工作流
|
|
157
184
|
- **Wiki**: 知识图谱,存储详细技术文档
|
|
158
185
|
- **Ralph**: 自适应决策引擎,动态调整执行链
|
|
186
|
+
- **Odyssey**: 长周期自主循环命令族(debug/improve/planex/ui/review-test-fix),自含考古→诊断→修复→泛化→知识持久化
|
|
187
|
+
- **Swarm Workflow**: 将任务路由到固定 Workflow 脚本进行多 agent 并发执行
|
|
188
|
+
- **Universal Workflow**: 动态生成对抗性 Workflow 脚本,支持复用和持久化
|
|
189
|
+
- **Grill**: 在 brainstorm 之前的压力测试,用代码库现实检验想法/需求
|
|
190
|
+
- **Next**: 智能导航,检测工作流状态并推荐最优下一步命令
|
|
159
191
|
```
|
|
160
192
|
|
|
161
193
|
### Mode 5: Beginner Onboarding
|
|
@@ -180,6 +212,7 @@ Issue 闭环: discover → create → analyze --gaps → plan --gaps → execute
|
|
|
180
212
|
| 7 | `/maestro-plan` | 规划 | 分析完成后生成执行计划,支持 `--from analyze:ANL-xxx` 直达 |
|
|
181
213
|
| 8 | `/maestro-execute` | 执行 | 计划完成后,执行实现 |
|
|
182
214
|
| 9 | `/maestro-quick` | 快速任务 | 简单任务跳过管线 |
|
|
215
|
+
| 10 | `/maestro-next` | 智能导航 | 不确定下一步时,自动检测状态推荐 |
|
|
183
216
|
```
|
|
184
217
|
|
|
185
218
|
#### Step 5.2: 快速入门路径
|
|
@@ -356,6 +356,7 @@ Generate steps from `session.lifecycle_position` to `milestone-complete`.
|
|
|
356
356
|
| brainstorm | `maestro-brainstorm "{intent}" --from grill:{grill_id}` *(if grill ran)* / `maestro-brainstorm "{intent}"` *(otherwise)* | *(same)* | — | all |
|
|
357
357
|
| blueprint | `maestro-blueprint "{intent}"` | *(same)* | — | all |
|
|
358
358
|
| init | `maestro-init` | *(same)* | — | all |
|
|
359
|
+
| spec-setup | `spec-setup` | *(same)* | — | all (**仅当 `.workflow/specs/` 不存在时插入**) |
|
|
359
360
|
| analyze-macro | `maestro-analyze "{intent}"` | *(same)* | `post-analyze-scope` | all |
|
|
360
361
|
| roadmap | `maestro-roadmap --from analyze:{analyze_macro_id}` | *(same)* | — | all |
|
|
361
362
|
| analyze | `maestro-analyze {phase}` | `maestro-analyze` | — | all |
|
|
@@ -374,6 +375,7 @@ Generate steps from `session.lifecycle_position` to `milestone-complete`.
|
|
|
374
375
|
**Build rules (按顺序应用):**
|
|
375
376
|
|
|
376
377
|
0. **planning_mode 选列**:`unified` → invoke_skill(unified) 列;`independent` → invoke_skill(independent) 列
|
|
378
|
+
0.5. **specs 预检**:当 `lifecycle_position ∉ {grill, brainstorm, blueprint, init}` 且 `.workflow/specs/` 目录不存在时,在链路最前面插入 `spec-setup` 步骤(stage=`spec-setup`,无 decision)。确保下游 analyze/plan/execute 可获得项目约束规则注入
|
|
377
379
|
1. **起点**:从 `session.lifecycle_position` 开始
|
|
378
380
|
2. **跳过已完成**:跳过当前 milestone+phase 下已有 completed artifact 的 stage(按 `session.phase` 过滤);unified 按 milestone 过滤
|
|
379
381
|
3. **quality_mode 过滤**:按 `session.quality_mode` 排除不匹配 stage
|
|
@@ -408,7 +410,7 @@ Generate steps from `session.lifecycle_position` to `milestone-complete`.
|
|
|
408
410
|
1. Validate: 所有 step 的 `command_scope != "missing"`;否则 raise E006 + 列出缺失 skill
|
|
409
411
|
2. Write `.workflow/.maestro/ralph-{YYYYMMDD-HHmmss}/status.json` (Appendix: Session Schema)
|
|
410
412
|
3. Display chain overview:每步显示 `{index}. {skill} [{type}] [{command_scope}]`
|
|
411
|
-
4.
|
|
413
|
+
4. **Goal Prompt 显示(强制)**:`task_decomposition` 存在时,chain overview 之后**必须逐字显示** Goal Prompt block(Appendix),不得省略或摘要。显示后继续 handoff,不阻塞
|
|
412
414
|
|
|
413
415
|
### A_DELEGATE_EVALUATE
|
|
414
416
|
|
|
@@ -686,7 +688,11 @@ decision:post-goal-audit {retry+1}
|
|
|
686
688
|
```
|
|
687
689
|
📋 任务分解完成。可随时复制以下 /goal 设定终止条件(执行过程中输入即可):
|
|
688
690
|
|
|
689
|
-
/goal
|
|
691
|
+
/goal 完成以下子目标:
|
|
692
|
+
{for each G in task_decomposition:}
|
|
693
|
+
- {G.id}: {G.goal} — 完成条件: {G.done_when}
|
|
694
|
+
{end for}
|
|
695
|
+
直到 {session_dir}/status.json 的 task_decomposition[*] 与 steps[*] 全部 completion_confirmed=true 才停。每轮以 status.json 为唯一行动手册,通过 /maestro-ralph-execute 推进 step;decision 节点由其自动 handoff 回 ralph 评估。禁止手动执行 skill 或修改 boundary_contract.out_of_scope。
|
|
690
696
|
```
|
|
691
697
|
|
|
692
698
|
`/goal` 由用户输入;ralph 输出提示词后继续 handoff,不阻塞。
|
|
@@ -729,6 +735,7 @@ decision:post-goal-audit {retry+1}
|
|
|
729
735
|
- [ ] post-goal-audit decision 仅在 decomposed 时插入,位于 milestone-complete 之前
|
|
730
736
|
- [ ] Unmet sub-goals 动态 grow steps[](goal_ref tagged);max retries → escalate
|
|
731
737
|
- [ ] planning_mode 显式决定;unified=无 `{phase}`, independent=带 `{phase}`
|
|
738
|
+
- [ ] specs 预检:lifecycle_position 在 init 之后 + `.workflow/specs/` 不存在 → 链路最前面插入 `spec-setup`
|
|
732
739
|
- [ ] Chain 必须以 `milestone-complete` 结尾
|
|
733
740
|
- [ ] Decision nodes 由 maestro delegate --role analyze 评估
|
|
734
741
|
- [ ] Ralph 不执行 step,只 evaluate;invoke_skill("maestro-ralph-execute") handoff
|
|
@@ -51,6 +51,7 @@ Remaining → intent (new session)
|
|
|
51
51
|
9. **Decomposition is outcome-oriented** — sub-goals 为可观测交付,禁止 lifecycle 复刻;`/goal` 由用户输入
|
|
52
52
|
10. **planning_mode governs arg granularity** — `unified` → skill args 无 `{phase}`;`independent` → 含 `{phase}`
|
|
53
53
|
11. **task_decomposition 驱动 steps[] 动态生长** — `post-goal-audit` 按 unmet 子目标插入 scoped mini-loop;字段累加,既有字段不删不改
|
|
54
|
+
12. **禁止以上下文消耗为由中断自调用链** — harness 自动处理 context compression(消息摘要),模型无需判断上下文剩余空间。自调用的唯一合法终止条件是:router 命中终止条件(S_COMPLETE / S_FALLBACK / END)、session paused、或 decision 末尾的 A_PAUSE_ESCALATE。以"上下文不足"、"已连续完成 N 个 step"、"避免 context overflow"等理由中断属于 invariant violation
|
|
54
55
|
</invariants>
|
|
55
56
|
|
|
56
57
|
<router>
|
|
@@ -449,7 +450,7 @@ narrow → derive defaults from intent + codebase, skip questions.
|
|
|
449
450
|
1. 校验所有 step 的 `command_scope != "missing"`;否则 raise E006 + 列出缺失 skill
|
|
450
451
|
2. Write `.workflow/.maestro/ralph-{YYYYMMDD-HHmmss}/status.json`
|
|
451
452
|
3. Display chain overview:每步显示 `{index}. {skill} [{command_scope}]`
|
|
452
|
-
4.
|
|
453
|
+
4. **Goal Prompt 显示(强制)**:`task_decomposition` 存在时,chain overview 之后**必须逐字显示** Goal Prompt block(Appendix),不得省略或摘要。显示后继续进入 tick
|
|
453
454
|
|
|
454
455
|
### A_LOCATE_SESSION
|
|
455
456
|
|
|
@@ -841,7 +842,11 @@ decomposition 产出后,链路概览之后逐字显示:
|
|
|
841
842
|
```
|
|
842
843
|
📋 任务分解完成。可随时复制以下 /goal 设定终止条件:
|
|
843
844
|
|
|
844
|
-
/goal
|
|
845
|
+
/goal 完成以下子目标:
|
|
846
|
+
{for each G in task_decomposition:}
|
|
847
|
+
- {G.id}: {G.goal} — 完成条件: {G.done_when}
|
|
848
|
+
{end for}
|
|
849
|
+
直到 {session_dir}/status.json 的 task_decomposition[*] 与 steps[*] 全部 completion_confirmed=true 才停。每轮以 status.json 为唯一行动手册,通过 /maestro-ralph-beta 推进 tick;decision 节点由 ralph 内联评估。禁止手动执行 skill 或修改 boundary_contract.out_of_scope。
|
|
845
850
|
```
|
|
846
851
|
|
|
847
852
|
### Error Codes
|
|
@@ -55,6 +55,7 @@ HARD RULES:
|
|
|
55
55
|
7. **status.json 每步骤后由 CLI 原子写盘** — resume-safe
|
|
56
56
|
8. **STATUS 枚举受限** — 仅 `DONE | DONE_WITH_CONCERNS | NEEDS_RETRY | BLOCKED`;`NEEDS_CONTEXT` 已废除
|
|
57
57
|
9. **CLI 输出禁止截断** — `maestro ralph next` 的 stdout 包含完整 skill prompt,必须全量捕获。**严禁** `| head`、`| tail`、`2>&1 | head -N` 等任何截断管道。Bash timeout 可加长但不可截断输出
|
|
58
|
+
10. **禁止以上下文消耗为由中断自调用链** — harness 自动处理 context compression(消息摘要),模型无需判断上下文剩余空间。自调用链的唯一合法终止条件是:全部 `completion_confirmed`、session paused、或 decision handoff 到 ralph。以"上下文不足"、"已连续完成 N 个 step"、"避免 context overflow"等理由中断属于 invariant violation
|
|
58
59
|
</invariants>
|
|
59
60
|
|
|
60
61
|
<state_machine>
|
|
@@ -46,6 +46,7 @@ Follow '~/.maestro/workflows/harvest.md' Stages 1-8 in order.
|
|
|
46
46
|
2. **Dedup before write** — check harvest-log.jsonl and existing stores before each write.
|
|
47
47
|
3. **Never modify source artifacts** — harvest is purely extractive.
|
|
48
48
|
4. **Dedup contract with parallel writers** — when appending to `issues.jsonl`, set `source: "harvest"` on each row so concurrent writers (e.g. `manage-issue-discover` with `source: "discover"`) can be distinguished and deduplicated.
|
|
49
|
+
5. **Conflict pre-check on spec routing** — when routing to spec (Stage 6b), compare new entry against existing specs with same keywords/category. If semantic conflict detected, set `confidence="low"` on the new entry and log conflict note. Use `maestro spec conflict mark` if contradiction is clear.
|
|
49
50
|
|
|
50
51
|
Extraction patterns, classification rules, routing infrastructure, and fragment ID scheme defined in workflow harvest.md.
|
|
51
52
|
|
|
@@ -60,6 +61,8 @@ Extraction patterns, classification rules, routing infrastructure, and fragment
|
|
|
60
61
|
| Wiki graph needs linking | `/manage-wiki connect --fix` |
|
|
61
62
|
| Issues created | `/manage-issue list --source harvest` |
|
|
62
63
|
| Specs extracted | `/spec-load --role implement` |
|
|
64
|
+
| Specs extracted (冲突审查) | `/manage-knowledge-audit --scope spec` — 新写入的 spec 可能与现有条目矛盾 |
|
|
65
|
+
| Spec 冲突标记已存在 | `maestro spec conflict list` — 查看当前冲突状态 |
|
|
63
66
|
| Full phase retrospective | `/quality-retrospective` |
|
|
64
67
|
</completion>
|
|
65
68
|
|
|
@@ -45,10 +45,12 @@ Follow `~/.maestro/workflows/knowledge-audit.md` Stages 1-8 in order.
|
|
|
45
45
|
**GATE 1: Load → Detect** (Stages 1-2 → Stage 4)
|
|
46
46
|
- REQUIRED: Scope 解析通过,互斥标志校验完成。
|
|
47
47
|
- REQUIRED: 三存储按 scope 加载完成。
|
|
48
|
+
- REQUIRED: 加载已有冲突标记: `maestro spec conflict list` → 合并到 finding 池。
|
|
48
49
|
- BLOCKED if scope 非法或存储不可读: E001/E002。
|
|
49
50
|
|
|
50
51
|
**GATE 2: Detect → Decision** (Stage 4 → Stage 5)
|
|
51
52
|
- REQUIRED: Finding 池按 P0/P1/P2 分级输出。
|
|
53
|
+
- REQUIRED: 已标记 `contested` 的条目自动归入 P0 finding(来源: conflict-marker)。
|
|
52
54
|
- REQUIRED: 未 harvest 的 artifact 删除前触发抢救确认(W002)。
|
|
53
55
|
- BLOCKED if finding 为空: 无需淘汰,直接输出报告。
|
|
54
56
|
|
|
@@ -63,6 +65,31 @@ Follow `~/.maestro/workflows/knowledge-audit.md` Stages 1-8 in order.
|
|
|
63
65
|
- **Deprecate over delete**: 文本存储首选 `status="deprecated"`,保留历史。
|
|
64
66
|
- **Purge 仅 artifact**: `--purge` 不作用于 spec/knowhow。
|
|
65
67
|
- **Rescue before delete**: 未抽取 artifact 删除前强制提示先 `/manage-harvest`。
|
|
68
|
+
|
|
69
|
+
### Conflict Resolution Integration
|
|
70
|
+
|
|
71
|
+
四态决策(扩展自三态 keep/deprecate/delete):
|
|
72
|
+
|
|
73
|
+
| 动作 | 适用场景 | 执行 |
|
|
74
|
+
|------|---------|------|
|
|
75
|
+
| `keep` | 内容正确,无需变更 | 写 audit-log ignore 记录 |
|
|
76
|
+
| `contest` | 矛盾真实存在,需进一步审查 | `maestro spec conflict mark <file> <line> --note "<evidence>"` |
|
|
77
|
+
| `deprecate` | 内容过时或被取代 | 注入 `status="deprecated"` + `maestro spec conflict clear <file> <line>` |
|
|
78
|
+
| `delete` | 内容明确错误 | 移除 entry + `maestro spec conflict clear <file> <line>` |
|
|
79
|
+
|
|
80
|
+
**关键**: deprecate/delete 执行时,如果目标条目有 conflict-marker,必须同步调用 `maestro spec conflict clear` 清除标记,避免悬空冲突。
|
|
81
|
+
|
|
82
|
+
### Code-as-Truth 校验(审查核心原则)
|
|
83
|
+
|
|
84
|
+
**代码是唯一真理源。** Spec/knowhow 中的任何声明,必须与代码实际行为一致。
|
|
85
|
+
|
|
86
|
+
当 detector 发现 spec 条目声称某行为/规则时:
|
|
87
|
+
1. **代码校验**: grep/read 代码中相关实现,确认 spec 声明是否与代码一致
|
|
88
|
+
2. **不一致处理**:
|
|
89
|
+
- 代码正确、spec 过时 → `deprecate` 或 `delete` spec 条目
|
|
90
|
+
- 代码正确、spec 不完整 → `contest` 并建议补充
|
|
91
|
+
- 代码有 bug、spec 正确 → `keep` spec,生成 issue 修代码
|
|
92
|
+
3. **禁止**: 仅凭 spec 文本判断正确性。每个 finding 的 evidence 必须包含代码引用(文件:行号)
|
|
66
93
|
</execution>
|
|
67
94
|
|
|
68
95
|
<completion>
|
|
@@ -73,6 +100,8 @@ Follow `~/.maestro/workflows/knowledge-audit.md` Stages 1-8 in order.
|
|
|
73
100
|
| 复审淘汰记录 | 查看 `audit-report-{date}.md` |
|
|
74
101
|
| 抢救未抽取 artifact | `/manage-harvest <artifact-id>` |
|
|
75
102
|
| 验证 spec 现状 | `/spec-load --role implement` |
|
|
103
|
+
| 查看冲突标记 | `maestro spec conflict list` |
|
|
104
|
+
| 清除已解决冲突 | `maestro spec conflict clear-all <file>` |
|
|
76
105
|
| 周期巡检 | `--scope all --report` |
|
|
77
106
|
</completion>
|
|
78
107
|
|
|
@@ -57,6 +57,8 @@ Treats every bug as a learning signal with exhaustive iteration until root cause
|
|
|
57
57
|
- 关键判断用不同 `--role`(analyze / review / explore)获取多视角
|
|
58
58
|
- 修复前后各做一次 CLI review 确认
|
|
59
59
|
- 不同阶段可调用不同工具,综合多方意见再行动
|
|
60
|
+
|
|
61
|
+
4. **禁止以上下文消耗为由中断** — harness 自动处理 context compression,以"上下文不足"或"已执行 N 个阶段"为由中断属于纪律违反;必须完整走完状态机直到 S_RECORD → END
|
|
60
62
|
</execution_discipline>
|
|
61
63
|
|
|
62
64
|
<context>
|
|
@@ -392,7 +394,11 @@ Goals: {done}/{total} confirmed ({skipped} skipped)
|
|
|
392
394
|
```
|
|
393
395
|
📋 Debug Odyssey 会话已创建。可随时复制以下 /goal 设定终止条件:
|
|
394
396
|
|
|
395
|
-
/goal
|
|
397
|
+
/goal 完成以下目标:
|
|
398
|
+
{for each G in phase_goals where status != "skipped":}
|
|
399
|
+
- {G.id}: {G.goal} — 完成条件: {G.done_when}
|
|
400
|
+
{end for}
|
|
401
|
+
穷尽迭代:直到根因确认(或明确 INCONCLUSIVE)且修复验证通过
|
|
396
402
|
且泛化扫描穷尽(所有 hits 已分类处理)且 phase_goals_all_done=true 才停。
|
|
397
403
|
假设失败时扩范围→换视角→升级工具,不轻易放弃。
|
|
398
404
|
泛化发现的同类 bug 全部修复或创建 issue,不允许遗留。
|
|
@@ -47,6 +47,8 @@ Baseline-first approach with exhaustive iteration until zero remaining actionabl
|
|
|
47
47
|
- survey 阶段: `--role explore` 发现依赖/复杂度热点
|
|
48
48
|
- audit/diagnose: `--role analyze` 获取多视角分析
|
|
49
49
|
- fix 前后: `--role review` 确认改进正确性
|
|
50
|
+
|
|
51
|
+
4. **禁止以上下文消耗为由中断** — harness 自动处理 context compression,以"上下文不足"或"已执行 N 个阶段"为由中断属于纪律违反;必须完整走完状态机直到 S_RECORD → END
|
|
50
52
|
</execution_discipline>
|
|
51
53
|
|
|
52
54
|
<context>
|
|
@@ -431,7 +433,11 @@ Goals: {done}/{total} ({skipped} skipped)
|
|
|
431
433
|
```
|
|
432
434
|
📋 Improve Odyssey 会话已创建。可随时复制以下 /goal 设定终止条件:
|
|
433
435
|
|
|
434
|
-
/goal
|
|
436
|
+
/goal 完成以下目标:
|
|
437
|
+
{for each G in phase_goals where status != "skipped":}
|
|
438
|
+
- {G.id}: {G.goal} — 完成条件: {G.done_when}
|
|
439
|
+
{end for}
|
|
440
|
+
穷尽迭代:直到 session.json 的 audit_result 中所有 findings 均已处理(fix/issue/decision)
|
|
435
441
|
且 phase_goals_all_done=true 才停。修复按 severity 逐轮迭代,每轮修复后 re-verify 修改区域。
|
|
436
442
|
Baseline metrics 必须在修复前采集,修复后必须与 baseline 对比确认改进。
|
|
437
443
|
遇到 phase=decision 的 pending 必须 ask_user。不允许"只报告不处理"。
|
|
@@ -56,6 +56,8 @@ plan → execute → verify → fix gaps → iterate until ALL criteria pass.
|
|
|
56
56
|
- 计划阶段: `--role analyze` 获取任务分解建议
|
|
57
57
|
- 修复前后: `--role review` 确认无回归
|
|
58
58
|
- verify 阶段: cli-review 类型标准自动 delegate
|
|
59
|
+
|
|
60
|
+
4. **禁止以上下文消耗为由中断** — harness 自动处理 context compression,以"上下文不足"或"已执行 N 个阶段"为由中断属于纪律违反;必须完整走完状态机直到 S_RECORD → END
|
|
59
61
|
</execution_discipline>
|
|
60
62
|
|
|
61
63
|
<context>
|
|
@@ -515,7 +517,15 @@ Write understanding.md §6, generalization_stats. Mark G5 done.
|
|
|
515
517
|
```
|
|
516
518
|
📋 Planex Odyssey 会话已创建。可随时复制以下 /goal 设定终止条件(执行过程中输入即可):
|
|
517
519
|
|
|
518
|
-
/goal
|
|
520
|
+
/goal 完成以下目标:
|
|
521
|
+
{for each G in phase_goals where status != "skipped":}
|
|
522
|
+
- {G.id}: {G.goal} — 完成条件: {G.done_when}
|
|
523
|
+
{end for}
|
|
524
|
+
验收标准:
|
|
525
|
+
{for each AC in acceptance_criteria:}
|
|
526
|
+
- {AC.id}: {AC.criterion} (验证方式: {AC.verify_method})
|
|
527
|
+
{end for}
|
|
528
|
+
穷尽迭代:直到 acceptance_criteria[*] 全部 status==passed
|
|
519
529
|
且 phase_goals_all_done=true 才停。verify 失败自动 fix→re-verify 循环。
|
|
520
530
|
每轮修复后重新验证,新发现的标准违反继续修,不超过 max_iterations。
|
|
521
531
|
遇到 phase=decision 的 pending 必须 ask_user,不得自行 resolve。
|
|
@@ -46,6 +46,8 @@ fix ALL findings → confirm → generalize → discover → persist. Zero-resid
|
|
|
46
46
|
3. **多 CLI 辅助** — 利用 `maestro delegate` 调用多个 CLI 工具交叉验证
|
|
47
47
|
- 关键判断用不同 `--role`(analyze / review / explore)获取多视角
|
|
48
48
|
- 修复前后各做一次 CLI review 确认
|
|
49
|
+
|
|
50
|
+
4. **禁止以上下文消耗为由中断** — harness 自动处理 context compression,以"上下文不足"或"已执行 N 个阶段"为由中断属于纪律违反;必须完整走完状态机直到 S_RECORD → END
|
|
49
51
|
</execution_discipline>
|
|
50
52
|
|
|
51
53
|
<context>
|
|
@@ -366,7 +368,11 @@ Goals: {done}/{total} ({skipped} skipped)
|
|
|
366
368
|
```
|
|
367
369
|
📋 Review-Test-Fix Odyssey 会话已创建。可随时复制以下 /goal 设定终止条件(执行过程中输入即可):
|
|
368
370
|
|
|
369
|
-
/goal
|
|
371
|
+
/goal 完成以下目标:
|
|
372
|
+
{for each G in phase_goals where status != "skipped":}
|
|
373
|
+
- {G.id}: {G.goal} — 完成条件: {G.done_when}
|
|
374
|
+
{end for}
|
|
375
|
+
穷尽迭代:直到 session.json 的 review_result.remaining_actionable == 0
|
|
370
376
|
且 confirmation.verdict == "confirmed" 且 phase_goals_all_done == true 才停。
|
|
371
377
|
修复按 severity 逐轮迭代(critical→high→medium→low),每轮修复后 re-review 修改区域。
|
|
372
378
|
发现新问题追加到当前轮继续。遇到 phase=decision 的 pending 必须 ask_user。
|
|
@@ -47,6 +47,8 @@ fix → verify → generalize → discover → persist. Every pixel is a learnin
|
|
|
47
47
|
- survey 阶段: `--role explore` 发现设计系统用法
|
|
48
48
|
- audit/diverge: `--role analyze` 获取多视角创意
|
|
49
49
|
- fix 前后: `--role review` 确认视觉正确性
|
|
50
|
+
|
|
51
|
+
4. **禁止以上下文消耗为由中断** — harness 自动处理 context compression,以"上下文不足"或"已执行 N 个阶段"为由中断属于纪律违反;必须完整走完状态机直到 S_RECORD → END
|
|
50
52
|
</execution_discipline>
|
|
51
53
|
|
|
52
54
|
<context>
|
|
@@ -391,7 +393,11 @@ Goals: {done}/{total} ({skipped} skipped)
|
|
|
391
393
|
```
|
|
392
394
|
📋 UI Odyssey 会话已创建。可随时复制以下 /goal 设定终止条件(执行过程中输入即可):
|
|
393
395
|
|
|
394
|
-
/goal
|
|
396
|
+
/goal 完成以下目标:
|
|
397
|
+
{for each G in phase_goals where status != "skipped":}
|
|
398
|
+
- {G.id}: {G.goal} — 完成条件: {G.done_when}
|
|
399
|
+
{end for}
|
|
400
|
+
穷尽迭代:直到 session.json 的 audit + diverge findings 均已处理(fix/issue/decision)
|
|
395
401
|
且 phase_goals_all_done=true 才停。修复按 impact×severity 逐轮迭代。
|
|
396
402
|
每轮修复后重审修改区域,新发现追加继续修。
|
|
397
403
|
遇到 phase=decision 的 pending 必须 ask_user。不允许"只报告不处理"。
|
|
@@ -52,6 +52,7 @@ Each artifact's type determines its outputs at `.workflow/{a.path}/`:
|
|
|
52
52
|
- Codebase docs: `.workflow/codebase/ARCHITECTURE.md` → component boundaries, layer rules
|
|
53
53
|
- Wiki constraints: `maestro search "architecture constraint" --json` → documented decisions
|
|
54
54
|
- Specs: `maestro spec load --category review` → review standards, checklists, knowhow tools
|
|
55
|
+
- Conflict state: `maestro spec conflict list` → 当前已标记冲突的 spec 条目(review 时优先关注)
|
|
55
56
|
- Role knowledge: `maestro search --category review` → select relevant → `maestro wiki load`
|
|
56
57
|
|
|
57
58
|
**Output**: `REVIEW_DIR = .workflow/scratch/{YYYYMMDD}-review-P{N}-{slug}/` (P{N} = phase number, enables directory-level identification as state.json fallback)
|
|
@@ -75,6 +76,7 @@ Follow '~/.maestro/workflows/review.md' completely.
|
|
|
75
76
|
- REQUIRED: review.json written with findings, severity distribution, and verdict.
|
|
76
77
|
- REQUIRED: Issues auto-created based on level thresholds.
|
|
77
78
|
- REQUIRED: index.json updated with review status.
|
|
79
|
+
- REQUIRED: Spec conflict check — if any finding directly contradicts a loaded spec entry (code behavior ≠ spec rule), suggest `maestro spec conflict mark` on the spec entry. Code is the single source of truth.
|
|
78
80
|
|
|
79
81
|
**Output writes to REVIEW_DIR** (not EXEC_DIR):
|
|
80
82
|
- `REVIEW_DIR/review.json` — findings, severity distribution, verdict
|
|
@@ -129,6 +131,7 @@ maestro ralph complete <idx> --status {STATUS} [--evidence {path}]
|
|
|
129
131
|
| PASS verdict | `/quality-test {phase}` |
|
|
130
132
|
| WARN verdict (non-blocking issues) | `/quality-test {phase}` (proceed with caveats) |
|
|
131
133
|
| BLOCK verdict (critical findings) | `/maestro-plan {phase} --gaps` (fix first) |
|
|
134
|
+
| Spec contradictions found | `maestro spec conflict list` → `/manage-knowledge-audit --scope spec` |
|
|
132
135
|
| Want code cleanup | `/quality-refactor {phase}` |
|
|
133
136
|
</completion>
|
|
134
137
|
|
|
@@ -189,9 +189,11 @@ All behavioral statements MUST use MUST / SHOULD / MAY / MUST NOT / SHOULD NOT.
|
|
|
189
189
|
- [ ] No interrogative sentences (all declarative)
|
|
190
190
|
- [ ] system-architect: §3 contains "Data Model" and "State Machine" headings
|
|
191
191
|
|
|
192
|
-
##
|
|
192
|
+
## Completion Sequence (MANDATORY ORDER)
|
|
193
193
|
|
|
194
|
-
|
|
194
|
+
**Step A — Write files**: Use the Write tool to create every file under `output_dir/`. Then verify with Glob that `analysis.md` and each `analysis-F-*.md` exist on disk. If any Write call fails, emit `TASK BLOCKED` and stop.
|
|
195
|
+
|
|
196
|
+
**Step B — Report summary**: ONLY AFTER Step A files are confirmed on disk, emit this text (the orchestrator reads this summary to track progress — it does NOT substitute for the files):
|
|
195
197
|
|
|
196
198
|
```
|
|
197
199
|
TASK COMPLETE
|
|
@@ -204,7 +206,7 @@ rfc_keyword_count: {N}
|
|
|
204
206
|
total_lines: {sum across all files}
|
|
205
207
|
```
|
|
206
208
|
|
|
207
|
-
If blocked: report
|
|
209
|
+
Never emit `TASK COMPLETE` without files on disk. If blocked: report with `TASK BLOCKED` prefix.
|
|
208
210
|
|
|
209
211
|
## NEVER
|
|
210
212
|
|
|
@@ -49,6 +49,7 @@ $ARGUMENTS — user intent text, or special keywords.
|
|
|
49
49
|
10. **每个 step 必须 `completion_confirmed: true`** — 由 `maestro ralph complete N --status DONE|DONE_WITH_CONCERNS` 写入
|
|
50
50
|
11. **schema** — `ralph_protocol_version: "1"` 标记 CLI-driven session;新增字段全部可选
|
|
51
51
|
12. **Invariant violation = BLOCK** — 违反上述任一 invariant 即阻断当前操作,不可绕过。特别是 invariant 1(dispatch via ralph-execute)和 invariant 2(session before execution)和 invariant 10(completion_confirmed 由 CLI 写入)为硬约束。
|
|
52
|
+
14. **禁止以上下文消耗为由中断执行** — harness 自动处理 context compression,以"上下文不足"或"避免 context overflow"为由中断属于 invariant violation
|
|
52
53
|
13. **Classification evidence** — S_CLASSIFY 的 chain 选择决策 MUST 记录到 status.json 的 `classification_rationale` 字段:匹配了哪个 pattern、排除了哪些备选、confidence level。无记录的分类不可进入 S_CREATE。
|
|
53
54
|
</invariants>
|
|
54
55
|
|
|
@@ -157,11 +158,16 @@ S_FALLBACK:
|
|
|
157
158
|
```
|
|
158
159
|
📋 任务分解完成。可随时复制下面一行设定目标(执行过程中输入即可):
|
|
159
160
|
|
|
160
|
-
/goal
|
|
161
|
+
/goal 完成以下子目标:
|
|
162
|
+
{for each G in task_decomposition:}
|
|
163
|
+
- {G.id}: {G.goal} — 完成条件: {G.done_when}
|
|
164
|
+
{end for}
|
|
165
|
+
达成条件: {session_dir}/status.json 中 task_decomposition[*].status == "done" 且 task_decomposition[*].completion_confirmed == true 且 steps[*].completion_confirmed == true。未达成时:阅读 {session_dir}/status.json 取得 execution_criteria / boundary_contract / task_decomposition / steps 作为行动手册,调用 /maestro-ralph continue 推进;严禁手动执行 skill 或越界修改 status.json.boundary_contract.out_of_scope。
|
|
161
166
|
```
|
|
162
167
|
|
|
163
168
|
### A_CREATE_SESSION
|
|
164
169
|
|
|
170
|
+
0. **Specs 预检**:当 chain 包含 `analyze-macro` / `analyze` / `plan` / `execute` 等执行 stage 且 `.workflow/specs/` 目录不存在时,在 steps 最前面插入 `spec-setup`(stage=`spec-setup`,无 decision)。确保下游可获得项目约束规则注入。chain ∈ {grill, brainstorm, blueprint, init, status, quick} 时跳过
|
|
165
171
|
1. Read `.workflow/state.json` 获取 phase / milestone(含 D-007 反查 `phase_slugs`);读最新 macro analyze artifact 注入 `scope_verdict` + `analyze_macro_id`(如存在);读最新 blueprint artifact 注入 `blueprint_id`
|
|
166
172
|
2. Create `.workflow/.maestro/maestro-{YYYYMMDD-HHMMSS}/status.json`(与 ralph 共用 schema):
|
|
167
173
|
```json
|
|
@@ -221,6 +227,7 @@ S_FALLBACK:
|
|
|
221
227
|
- [ ] plan 支持 `{phase}` / `--from analyze:{ANL_ID}` / `--from blueprint:{BLP_ID}` 三路径;`source_artifact_ref` 写入 step
|
|
222
228
|
- [ ] Broad lifecycle intents decomposed (≤3 boundary questions); narrow/single-step skip
|
|
223
229
|
- [ ] status.json 唯一真源;无 markdown 清单;post-goal-audit 节点在 decomposed 时追加;/goal 提示词以 status.json 为判据
|
|
230
|
+
- [ ] Specs 预检:chain 含执行 stage + `.workflow/specs/` 不存在 → steps 最前面插入 `spec-setup`
|
|
224
231
|
- [ ] Chain selected and confirmed (or auto-confirmed)
|
|
225
232
|
- [ ] Session dir created with status.json before execution; decomposition fields additive-only
|
|
226
233
|
- [ ] 执行 step 含 `command_scope` + `command_path` + `completion_confirmed`;decision step 由 `step.decision` 标识
|
|
@@ -50,11 +50,11 @@ Single source of truth: **[index/catalog.json](index/catalog.json)**
|
|
|
50
50
|
|
|
51
51
|
| Field | Purpose |
|
|
52
52
|
|-------|---------|
|
|
53
|
-
| `commands[]` |
|
|
54
|
-
| `skills[]` |
|
|
55
|
-
| `agents[]` |
|
|
53
|
+
| `commands[]` | 64 个 slash 命令,含分类和描述 |
|
|
54
|
+
| `skills[]` | 44 个 Skill,含分类和描述 |
|
|
55
|
+
| `agents[]` | 24 个 Agent,含分类和描述 |
|
|
56
56
|
| `cli_commands[]` | 21 个终端命令 |
|
|
57
|
-
| `guide_files[]` | 17 个 Guide
|
|
57
|
+
| `guide_files[]` | 17 个 Guide 文档索引(planned,尚未创建) |
|
|
58
58
|
| `essential_commands[]` | 10 个核心命令(新手用) |
|
|
59
59
|
| `workflows` | 主干管线、快速渠道、Issue 闭环、初始化路径 |
|
|
60
60
|
|
|
@@ -190,6 +190,11 @@ $ARGUMENTS → Parse:
|
|
|
190
190
|
| `/maestro-quick` | 快速任务 |
|
|
191
191
|
| `/maestro-overlay` | Overlay 管理 |
|
|
192
192
|
| `/maestro-amend` | 修正补丁 |
|
|
193
|
+
| `/maestro-grill` | 压力测试 — 对计划或需求进行代码库现实性压力测试 |
|
|
194
|
+
| `/maestro-next` | 智能导航 — 检测状态并推荐下一步最优命令 |
|
|
195
|
+
| `/maestro-companion` | 知识伴侣 — 上下文加载、洞察捕获、技能路由 |
|
|
196
|
+
| `/maestro-swarm-workflow` | Swarm 并行加速器 — 多 agent 并发执行 |
|
|
197
|
+
| `/maestro-universal-workflow` | 动态对抗工作流生成器 |
|
|
193
198
|
|
|
194
199
|
### 理解层 + 执行管线 (pipeline)
|
|
195
200
|
|
|
@@ -217,9 +222,23 @@ $ARGUMENTS → Parse:
|
|
|
217
222
|
| `/manage-issue` | Issue 管理 |
|
|
218
223
|
| `/manage-issue-discover` | Issue 发现 |
|
|
219
224
|
| `/manage-knowhow` | 知识管理 |
|
|
225
|
+
| `/manage-knowhow-capture` | 知识捕获 |
|
|
220
226
|
| `/manage-status` | 状态查看 |
|
|
221
227
|
| `/manage-wiki` | Wiki 管理 |
|
|
222
228
|
| `/manage-harvest` | 收获 |
|
|
229
|
+
| `/manage-codebase-rebuild` | 代码库重建 |
|
|
230
|
+
| `/manage-kg-extractors` | 知识图谱提取器管理 |
|
|
231
|
+
| `/manage-knowledge-audit` | 知识审计 |
|
|
232
|
+
|
|
233
|
+
### Odyssey 长周期循环 (odyssey)
|
|
234
|
+
|
|
235
|
+
| 命令 | 用途 |
|
|
236
|
+
|------|------|
|
|
237
|
+
| `/odyssey-debug` | 长周期调试 — 考古、诊断、修复、泛化 |
|
|
238
|
+
| `/odyssey-improve` | 长周期代码改进 — 多维审计、深度诊断、定向修复 |
|
|
239
|
+
| `/odyssey-review-test-fix` | 深度审查修复循环 |
|
|
240
|
+
| `/odyssey-planex` | 需求驱动迭代 — 计划/执行/验证/修复循环 |
|
|
241
|
+
| `/odyssey-ui` | 长周期 UI 优化 — 视觉调研、多维审计、修复 |
|
|
223
242
|
|
|
224
243
|
## Workflow Mapping
|
|
225
244
|
|
|
@@ -238,7 +257,7 @@ Roadmap > Milestone > Phase > Task
|
|
|
238
257
|
|
|
239
258
|
```
|
|
240
259
|
上游起源层(并列,可选)
|
|
241
|
-
brainstorm(发散/轻量) | blueprint(收敛/重型)
|
|
260
|
+
brainstorm(发散/轻量) | blueprint(收敛/重型) | grill(压力测试)
|
|
242
261
|
|
|
243
262
|
理解层
|
|
244
263
|
analyze 双层: 宏观(文本参数) → scope_verdict | 微观(数字参数) → Phase 级决策
|
|
@@ -248,6 +267,13 @@ Roadmap > Milestone > Phase > Task
|
|
|
248
267
|
|
|
249
268
|
执行层
|
|
250
269
|
plan → execute
|
|
270
|
+
|
|
271
|
+
Odyssey 长周期循环(独立路径)
|
|
272
|
+
odyssey-debug / odyssey-improve / odyssey-review-test-fix / odyssey-planex / odyssey-ui
|
|
273
|
+
|
|
274
|
+
自适应引擎(高级)
|
|
275
|
+
ralph / ralph-beta → 自运行决策循环
|
|
276
|
+
swarm-workflow / universal-workflow → 多 agent 并行执行
|
|
251
277
|
```
|
|
252
278
|
|
|
253
279
|
### 合法路径
|
|
@@ -265,6 +291,9 @@ Roadmap > Milestone > Phase > Task
|
|
|
265
291
|
| 全自动 | /maestro 入口 | `/maestro -y "任务描述"` |
|
|
266
292
|
| 代码审查 | 质量管线 | `/quality-review` → `/quality-auto-test` → `/quality-test` |
|
|
267
293
|
| 团队开发 | Team Lite | `/maestro-collab` |
|
|
294
|
+
| 长周期调试 | Odyssey 深度循环 | `/odyssey-debug "问题描述"` |
|
|
295
|
+
| 长周期改进 | Odyssey 深度循环 | `/odyssey-improve "改进目标"` |
|
|
296
|
+
| 需求迭代 | Odyssey 深度循环 | `/odyssey-planex "需求描述"` |
|
|
268
297
|
|
|
269
298
|
## Core Rules
|
|
270
299
|
|
|
@@ -292,9 +321,9 @@ Roadmap > Milestone > Phase > Task
|
|
|
292
321
|
|
|
293
322
|
## Statistics
|
|
294
323
|
|
|
295
|
-
- **Slash 命令**:
|
|
324
|
+
- **Slash 命令**: 64 个(13 个分类:core/pipeline/milestone/manage/quality/spec/learn/worktree/team/ralph/ui/tools/odyssey)
|
|
296
325
|
- **CLI 命令**: 21 个
|
|
297
|
-
- **Skills**:
|
|
298
|
-
- **Agents**:
|
|
299
|
-
- **Guide 文档**: 17
|
|
300
|
-
- **工作流路径**:
|
|
326
|
+
- **Skills**: 44 个(5 个分类:meta/team/knowledge/quality/scholar)
|
|
327
|
+
- **Agents**: 24 个(4 个分类:workflow/team/cli/ui)
|
|
328
|
+
- **Guide 文档**: 17 个(planned,尚未创建)
|
|
329
|
+
- **工作流路径**: 7 个合法路径 (Path A-G) + 4 个辅助流程
|