maestro-flow 0.4.25 → 0.5.0

package/.codex/skills/maestro-execute/SKILL.md

@@ -164,6 +164,9 @@ Each wave generates `wave-{N}.csv` with extra `prev_context` column populated fr
 8. **Max 3 Fix Attempts**: Per task, auto-fix convergence failures up to 3 times, then mark blocked
 9. **Breakpoint Resume**: Always detect completed tasks and skip them on re-run
 10. **DO NOT STOP**: Continuous execution until all waves complete or user explicitly stops
+11. **Invariant violation = BLOCK** — violating any invariant above blocks the current operation.
+12. **Evidence required in task summaries** — task summaries MUST include: files actually modified (not just planned targets), convergence criteria verification results (pass/fail with evidence), any deviations from plan with rationale. "Task completed successfully" without evidence is INVALID.
+13. **Artifact verification before completion** — for each completed task, .summaries/TASK-{NNN}-summary.md MUST exist with concrete evidence. EXC artifact MUST be registered in state.json. If any missing: DO NOT report completion.
 </invariants>
 
 <execution>
@@ -233,7 +236,7 @@ If exit code is 1, present warnings and ask whether to proceed.
 
 7. **Load codebase + wiki context** (optional, passed to all agents):
    - If `.workflow/codebase/ARCHITECTURE.md` exists: read and include as `codebase_context` in agent instructions
-   - Run `maestro wiki search "<phase keywords>" --json 2>/dev/null`; if results: include top 5 entries as `wiki_context`
+   - Run `maestro search "<phase keywords>" --json 2>/dev/null`; if results: include top 5 entries as `wiki_context`
    - Both are optional — proceed without if unavailable
 
 7. **User validation**: Display task/wave breakdown. Skip if AUTO_YES.
@@ -347,24 +350,24 @@ Blocked/failed tasks cascade: mark all downstream dependents as `skipped` with e
 4. **Register EXC artifact in state.json**: Find matching plan artifact, create `{ id: "EXC-{next_id}", type: "execute", milestone, phase, scope, path, status: "completed", depends_on: plan_artifact.id, harvested: false, created_at, completed_at }`. `milestone` MUST come from D-007 `phase_slugs` reverse lookup (numeric phase) — inherit from matching plan artifact if available, otherwise reverse-lookup directly.
 
 5. **Extract incremental specs**: Read `.summaries/`, use `maestro spec add` CLI:
-   - Learnings/pitfalls → `maestro spec add learning "<title>" "<content>" --keywords ... --source execute:{PLAN_DIR}`
-   - Design rationale → `maestro spec add coding "<title>" "<content>" --keywords ...`
-   - Root cause/workaround → `maestro spec add debug "<title>" "<content>" --keywords ...`
+   - Learnings/pitfalls → `maestro spec add learning "<title>" "<content>" --keywords ... --description "<summary>" --source execute:{PLAN_DIR}`
+   - Design rationale → `maestro spec add coding "<title>" "<content>" --keywords ... --description "<summary>"`
+   - Root cause/workaround → `maestro spec add debug "<title>" "<content>" --keywords ... --description "<summary>"`
    Mark artifact `harvested: true`
 
 6. **Post-task Knowledge Inquiry**: After each task completes, evaluate inquiry triggers:
 
    - **Execution deviation**: If task summary mentions approach change, dependency swap, or plan deviation:
      → Prompt: "TASK-{NNN} deviated from the plan. Record as architecture constraint?"
-     → On confirm: `maestro spec add arch "<decision>" "<rationale>" --keywords ... --source execute:{PLAN_DIR}`
+     → On confirm: `maestro spec add arch "<decision>" "<rationale>" --keywords ... --description "<summary>" --source execute:{PLAN_DIR}`
 
    - **Retry success**: If task required >=2 retries before completion:
      → Prompt: "TASK-{NNN} succeeded after {N} retries. Document this fix pattern?"
-     → On confirm: `maestro spec add debug "<pattern>" "<content>" --keywords ... --source execute:{PLAN_DIR}`
+     → On confirm: `maestro spec add debug "<pattern>" "<content>" --keywords ... --description "<summary>" --source execute:{PLAN_DIR}`
 
    - **Implicit knowledge**: If task summary contains design rationale ("chose X because", "rejected Y due to"):
      → Prompt: "Design decision detected. Record as a learning?"
-     → On confirm: `maestro spec add learning "<decision>" "<rationale>" --keywords ... --source execute:{PLAN_DIR}`
+     → On confirm: `maestro spec add learning "<decision>" "<rationale>" --keywords ... --description "<summary>" --source execute:{PLAN_DIR}`
 
    Use `request_user_input` for prompts:
    ```json
@@ -375,7 +378,7 @@ Blocked/failed tasks cascade: mark all downstream dependents as `skipped` with e
 
 8. **Auto-sync** (if config.json.codebase.auto_sync_after_execute == true): detect changed files, trigger codebase doc update.
 
-9. **Display completion report**: Phase, completed/blocked counts, wave progress, paths to `.summaries/` and `.task/`, next step suggestions (`maestro-verify`, `manage-status`).
+9. **Display completion report**: Phase, completed/blocked counts, wave progress, paths to `.summaries/` and `.task/`, next step suggestions (`quality-review`, `manage-status`).
 
 ### Shared Discovery Board Protocol
 

package/.codex/skills/maestro-fork/SKILL.md

@@ -125,6 +125,10 @@ Presence of `worktree-scope.json` signals "inside a worktree" — used by E003 v
 - Sync → resume work in worktree
 </execution>
 
+<invariants>
+**Artifact verification** — worktree-scope.json, scoped state.json, and worktrees.json registry update MUST all complete. If any missing: DO NOT report completion.
+</invariants>
+
 <error_codes>
 | Code | Severity | Condition | Recovery |
 |------|----------|-----------|----------|

package/.codex/skills/maestro-grill/SKILL.md

@@ -42,7 +42,7 @@ $ARGUMENTS -- topic/plan text for interactive mode, or --from source for upstrea
 **Produced files**: `grill-report.md`, `terminology.md`, `context-package.json`
 
 ### Role Knowledge
-`maestro wiki search "{topic keywords}"` → load relevant entries before grilling.
+`maestro search "{topic keywords}"` → load relevant entries before grilling.
 `maestro spec load --category arch` → load architecture constraints.
 </context>
 
@@ -83,12 +83,18 @@ Resume routing:
 - More branches to walk → `$maestro-grill "{topic}" -c`
 </execution>
 
+<invariants>
+1. **Invariant violation = BLOCK** — violating any invariant above blocks the current operation.
+2. **Code-grounded questions required** — grill questions MUST reference specific code (file:line) when challenging the user's proposal. Generic questions without code grounding are INVALID. If codebase scan failed, flag ALL locked decisions as LOW CONFIDENCE.
+3. **Artifact verification before completion** — verify grill-report.md (with Branch Log + Q&A + synthesis), terminology.md (≥5 terms), and context-package.json all exist before reporting completion. If any missing: DO NOT report completion.
+</invariants>
+
 <error_codes>
 | Code | Severity | Condition | Recovery |
 |------|----------|-----------|----------|
 | E001 | error | No topic/plan and no --from/--continue flag | Prompt user for topic text |
 | E002 | error | --session ID not found | Show available sessions |
-| W001 | warning | Codebase scan failed or returned empty | Continue without code grounding, note limitation |
+| W001 | warning | Codebase scan failed or returned empty | Retry once. If still fails: flag ALL subsequent locked decisions as LOW CONFIDENCE |
 | W002 | warning | CLI exploration timeout in auto mode | Skip question, mark as open |
 | W003 | warning | Max branch depth reached without resolution | Force synthesis, offer continuation |
 </error_codes>

package/.codex/skills/maestro-help/SKILL.md

@@ -95,7 +95,7 @@ Mode 3 (Smart Recommendations) 需要检测项目状态：
 4. 如有对应 guide 文档，读取相关段落
 
 Guide 映射：
-- analyze/plan/execute/verify → `guide/command-usage-guide.md`
+- analyze/plan/execute → `guide/command-usage-guide.md`
 - init/roadmap/blueprint → `guide/quick-start-guide.md`
 - ralph → `guide/maestro-ralph-guide.md`
 - maestro (协调器) → `guide/maestro-coordinator-guide.md`
@@ -115,9 +115,7 @@ Guide 映射：
 | roadmap 完成，phase=pending | `/maestro-analyze 1` | 微观分析：Phase 级深入探索 |
 | analyze (微观) 完成 | `/maestro-plan 1` | Phase 级规划 |
 | plan 完成 | `/maestro-execute` | 开始执行 |
-| execute 完成 | `/maestro-verify` | 验证成果 |
-| verify 有 gaps | `/maestro-analyze --gaps` | 重新分析 |
-| verify 通过 | `/quality-review` | 进入质量管线 |
+| execute 完成 | `/quality-review` | 进入质量管线 |
 | quality 全通过 | `/maestro-milestone-audit` | 里程碑审计 |
 | 所有 Phase 完成 | `/maestro-milestone-complete` | 关闭里程碑 |
 
@@ -126,16 +124,16 @@ Guide 映射：
 根据任务类型推荐工作流和命令序列：
 
 **Path A — 完整新项目**:
-- `brainstorm` → `blueprint`(可选) → `analyze "topic"`(宏观) → `roadmap` → `analyze 1`(微观) → `plan 1` → `execute` → `verify`
+- `brainstorm` → `blueprint`(可选) → `analyze "topic"`(宏观) → `roadmap` → `analyze 1`(微观) → `plan 1` → `execute` → `quality-review`
 
 **Path B — 旧项目大功能**:
-- `analyze "feature X"`(宏观) → `roadmap` → `analyze 1`(微观) → `plan 1` → `execute` → `verify`
+- `analyze "feature X"`(宏观) → `roadmap` → `analyze 1`(微观) → `plan 1` → `execute` → `quality-review`
 
 **Path C — 中等功能（跳过 roadmap）**:
-- `analyze "feature X"` → `plan --from analyze:ANL-xxx` → `execute` → `verify`
+- `analyze "feature X"` → `plan --from analyze:ANL-xxx` → `execute` → `quality-review`
 
 **Path D — 小改动**:
-- `plan "fix auth bug"` → `execute` → `verify`
+- `plan "fix auth bug"` → `execute` → `quality-review`
 - 快速: `/maestro-quick "功能描述"`
 - 全自动: `/maestro -y "功能描述"`
 
@@ -165,7 +163,6 @@ Guide 映射：
 | `/maestro-roadmap` | 路线图编排 | scope_verdict=large 时，Milestone > Phase 分解 |
 | `/maestro-plan` | 规划 | 分析完成后，支持 `--from analyze:ANL-xxx` 直达 |
 | `/maestro-execute` | 执行 | 计划完成后 |
-| `/maestro-verify` | 验证 | 执行完成后 |
 | `/maestro-quick` | 快速任务 | 简单任务跳过管线 |
 
 快速上手路径：

package/.codex/skills/maestro-help/catalog.json

@@ -13,8 +13,7 @@
     {"name": "maestro-analyze", "command": "/maestro-analyze", "category": "pipeline", "description": "双层分析 — 宏观(文本参数)探索影响面+scope_verdict / 微观(数字参数)Phase 级深入", "source": "../../commands/maestro-analyze.md"},
     {"name": "maestro-plan", "command": "/maestro-plan", "category": "pipeline", "description": "任务规划 — 支持 --from analyze:ANL-xxx 直达、碰撞检测、gap 修复计划", "source": "../../commands/maestro-plan.md"},
     {"name": "maestro-execute", "command": "/maestro-execute", "category": "pipeline", "description": "任务执行 — wave 并行执行、delegate 编排、自动提交", "source": "../../commands/maestro-execute.md"},
-    {"name": "maestro-verify", "command": "/maestro-verify", "category": "pipeline", "description": "验证确认 — 成果检查、gap 发现、下一阶段建议", "source": "../../commands/maestro-verify.md"},
-    {"name": "maestro-quick", "command": "/maestro-quick", "category": "core", "description": "快速任务 — 跳过 Phase 管线直接完成，支持 --full/--discuss 模式", "source": "../../commands/maestro-quick.md"},
+{"name": "maestro-quick", "command": "/maestro-quick", "category": "core", "description": "快速任务 — 跳过 Phase 管线直接完成，支持 --full/--discuss 模式", "source": "../../commands/maestro-quick.md"},
     {"name": "maestro-overlay", "command": "/maestro-overlay", "category": "core", "description": "Overlay 管理 — 创建、查看、删除命令补丁", "source": "../../commands/maestro-overlay.md"},
     {"name": "maestro-amend", "command": "/maestro-amend", "category": "core", "description": "修正补丁 — 从工作流信号自动生成 overlay", "source": "../../commands/maestro-amend.md"},
     {"name": "maestro-fork", "command": "/maestro-fork", "category": "worktree", "description": "Worktree 分叉 — 基于 milestone 创建隔离开发环境", "source": "../../commands/maestro-fork.md"},
@@ -154,7 +153,6 @@
     "/maestro-roadmap",
     "/maestro-plan",
     "/maestro-execute",
-    "/maestro-verify",
     "/maestro-quick"
   ],
   "workflows": {
@@ -168,16 +166,16 @@
       }
     },
     "paths": {
-      "A_full_project": "brainstorm → blueprint(可选) → analyze 宏观 → roadmap → analyze 微观 → plan → execute → verify",
-      "B_large_feature": "analyze 宏观 → roadmap → analyze 微观 → plan → execute → verify",
-      "C_medium_feature": "analyze 宏观 → plan --from analyze:ANL-xxx → execute → verify",
-      "D_small_fix": "plan → execute → verify",
+      "A_full_project": "brainstorm → blueprint(可选) → analyze 宏观 → roadmap → analyze 微观 → plan → execute → quality-review",
+      "B_large_feature": "analyze 宏观 → roadmap → analyze 微观 → plan → execute → quality-review",
+      "C_medium_feature": "analyze 宏观 → plan --from analyze:ANL-xxx → execute → quality-review",
+      "D_small_fix": "plan → execute → quality-review",
       "E_spec_only": "blueprint → (供人阅读)",
       "F_explore_only": "brainstorm → (供人决策)"
     },
     "main_pipeline": {
       "name": "主干管线",
-      "phases": ["analyze", "plan", "execute", "verify"],
+      "phases": ["analyze", "plan", "execute"],
       "quality": ["quality-auto-test", "quality-review", "quality-test"],
       "close": ["maestro-milestone-audit", "maestro-milestone-complete"]
     },

package/.codex/skills/maestro-impeccable/SKILL.md

@@ -153,6 +153,8 @@ Layer 1+2 both did not match, but intent is to build/create a specific thing:
 2. Read workflow file before execution — never execute a command without loading its .md
 3. Interactive gates respected — teach, shape, craft retain user confirmation gates
 4. status.json before chain steps — session created before any chain step runs
+5. **Quality gate enforcement** — score MUST be parsed from critique/audit output (not assumed). P0 count MUST be extracted. Do NOT skip quality gate steps or mark as "passed" without parsing actual score.
+6. **Chain completion verification** — all non-skipped steps MUST be executed (TodoWrite all completed). status.json MUST be updated with status: "completed" and final scores.
 </invariants>
 
 ## Prerequisites

package/.codex/skills/maestro-init/SKILL.md

@@ -155,4 +155,7 @@ Display created files and next steps: `$maestro-blueprint` (full spec), `$maestr
 - [ ] `config.json` written with configuration
 - [ ] `specs/` initialized with convention files
 - [ ] Completion report displayed with next steps
+
+### Artifact Verification
+Before reporting completion, verify: .workflow/project.md, .workflow/state.json, .workflow/config.json all exist. If any missing: DO NOT report completion.
 </success_criteria>

package/.codex/skills/maestro-merge/SKILL.md

@@ -41,13 +41,17 @@ $ARGUMENTS — milestone number and optional flags.
 Follow '~/.maestro/workflows/merge.md' completely.
 
 **Knowledge inquiry on completion:**
-After successful merge, ask user once: "Record milestone learnings?" If yes, persist via `maestro spec add learning "<title>" "<insight>" --keywords <kw1>,<kw2>`.
+After successful merge, ask user once: "Record milestone learnings?" If yes, persist via `maestro spec add learning "<title>" "<insight>" --keywords <kw1>,<kw2> --description "<summary>"`.
 
 **Next steps:**
 - View dashboard → `$manage-status`
 - Audit milestone → `$maestro-milestone-audit`
 </execution>
 
+<invariants>
+**Phase order enforcement** — git merge MUST complete before artifact sync. Do NOT sync artifacts until git merge succeeds — prevents partial state corruption.
+</invariants>
+
 <error_codes>
 | Code | Severity | Condition | Recovery |
 |------|----------|-----------|----------|

package/.codex/skills/maestro-milestone-audit/SKILL.md

@@ -44,7 +44,7 @@ id,title,description,scope,check_targets,deps,wave,status,findings,gaps_found,se
 | `check_targets` | Input | Specific verification commands/grep patterns |
 | `deps` | Input | Dependencies (empty — all wave 1) |
 | `wave` | Computed | Wave number (always 1 — single parallel wave) |
-| `status` | Lifecycle | `pending` (initial) → `pass`/`fail`/`warning`/`failed` (set by merge step from worker's `result_status`) |
+| `status` | Lifecycle | `pending` (initial) → `completed`/`failed` (set by merge step from worker's `result_status`) |
 | `findings` | Lifecycle | Detailed findings per dimension (max 500 chars; merged) |
 | `gaps_found` | Lifecycle | Semicolon-separated list of integration gaps (merged) |
 | `severity` | Lifecycle | `critical` / `warning` / `info` per gap (merged) |
@@ -67,6 +67,9 @@ id,title,description,scope,check_targets,deps,wave,status,findings,gaps_found,se
 2. **Non-blocking warnings** — missing analyze is warning, missing execute is error
 3. **Integration check is required** — always spawn checker via CSV wave
 4. **Clear verdict** — PASS or FAIL with specific reasons
+5. **Invariant violation = BLOCK** — violating any invariant above blocks the current operation.
+6. **Evidence required on audit checks** — every audit check result MUST cite what was examined and what was found. PASS: "Phase 1 chain complete: ANL-001 → PLN-001 → EXC-001". FAIL: "Phase 2 missing EXC artifact". Do NOT mark checks as PASS without verifying the artifact exists and contains expected content.
+7. **Artifact verification before completion** — audit-report.md with clear PASS/FAIL verdict MUST exist before reporting completion. If missing: DO NOT report completion.
 </invariants>
 
 <execution>
@@ -105,18 +108,19 @@ spawn_agents_on_csv({
     type: "object",
     properties: {
       id:            { type: "string" },
-      result_status: { type: "string", enum: ["pass", "fail", "warning", "failed"] },
+      result_status: { type: "string", enum: ["completed", "failed"] },
+      audit_verdict: { type: "string", enum: ["pass", "fail", "warning"], description: "Audit check outcome" },
       findings:      { type: "string", maxLength: 500 },
       gaps_found:    { type: "string", description: "Semicolon-separated list of gaps" },
       severity:      { type: "string", enum: ["critical", "warning", "info", ""] },
       error:         { type: "string" }
     },
-    required: ["id", "result_status", "findings", "severity"]
+    required: ["id", "result_status", "audit_verdict", "findings", "severity"]
   }
 })
 ```
 
-4. Merge results into master `tasks.csv`: map `result_status` → master `status` column, copy `findings`, `gaps_found`, `severity`, `error`. Delete temporary files (`wave-1.csv`, `wave-1-results.csv`) after merge.
+4. Merge results into master `tasks.csv`: map `result_status` → master `status` column, copy `audit_verdict`, `findings`, `gaps_found`, `severity`, `error`. Delete temporary files (`wave-1.csv`, `wave-1-results.csv`) after merge.
 
 #### Integration Checker Worker Contract (AUDIT_INTEGRATION_INSTRUCTION)
 
@@ -134,9 +138,9 @@ REQUIRED STEPS:
   5. Call report_agent_job_result EXACTLY ONCE
 
 TERMINATION CONTRACT (mandatory — NO worker may end without calling report_agent_job_result):
-  - Pass path  → no gaps found → result_status=pass, severity="info"
-  - Warning path → minor gaps → result_status=warning, severity="warning"
-  - Fail path → critical contract drift or broken dependencies → result_status=fail, severity="critical"
+  - Pass path  → no gaps found → result_status=completed, audit_verdict=pass, severity="info"
+  - Warning path → minor gaps → result_status=completed, audit_verdict=warning, severity="warning"
+  - Fail path → critical contract drift or broken dependencies → result_status=completed, audit_verdict=fail, severity="critical"
   - Failure path → cannot read scope, tool error → result_status=failed with error message
   - Timeout path → near 600s, finalize current findings → report with what was collected
   - NEVER skip report_agent_job_result.
@@ -144,7 +148,8 @@ TERMINATION CONTRACT (mandatory — NO worker may end without calling report_age
 OUTPUT (must match output_schema):
   {
     "id": "<your row id>",
-    "result_status": "pass" | "warning" | "fail" | "failed",
+    "result_status": "completed" | "failed",
+    "audit_verdict": "pass" | "warning" | "fail",
     "findings": "<one-sentence summary, max 500 chars>",
     "gaps_found": "<semicolon-separated list of gaps, each with file:line; empty if pass>",
     "severity": "critical" | "warning" | "info" | "",
@@ -157,7 +162,7 @@ CONSTRAINTS:
   - Do NOT call spawn_agents_on_csv (no recursion).
 ```
 5. Parse `gaps_found` from all workers — aggregate into `.workflow/milestones/{milestone}/audit-report.md`
-6. Any worker with `result_status == fail` and `severity == critical` → milestone verdict = FAIL
+6. Any worker with `audit_verdict == fail` and `severity == critical` → milestone verdict = FAIL
 
 ### Step 6: Verdict
 

package/.codex/skills/maestro-milestone-complete/SKILL.md

@@ -27,6 +27,7 @@ $maestro-milestone-complete --force "M1"  # skip audit check
 3. **Learnings are mandatory** — always extract before archiving
 4. **Clean after archive** — remove scratch dirs only after successful copy
 5. **Advance state** — always set next milestone or mark project complete
+6. **Prerequisite enforcement** — audit report MUST be verified as PASS before archival proceeds. Artifact verification: milestones/{M}/artifacts/ directory, milestone_history in state.json, project.md update all MUST exist before reporting completion.
 </invariants>
 
 <execution>
@@ -77,8 +78,12 @@ Snapshot `roadmap.md` as `roadmap-snapshot.md` in the milestone archive.
 
 **Write**: Append to `.workflow/specs/learnings.md` using `<spec-entry>` closed-tag format:
 ```
-<spec-entry category="learning" keywords="kw1, kw2" date="YYYY-MM-DD" source="milestone-complete:{milestone}">
+<spec-entry category="learning" keywords="kw1, kw2" date="YYYY-MM-DD" title="{title}" description="{one-line summary}" source="milestone-complete:{milestone}">
+
+### {title}
+
 Learning content here.
+
 </spec-entry>
 ```
 

package/.codex/skills/maestro-milestone-release/SKILL.md

@@ -46,6 +46,10 @@ Changelog: {N} entries
 ```
 </execution>
 
+<invariants>
+**Prerequisite enforcement** — milestone MUST be completed (audit PASS + milestone-complete run). Working tree MUST be clean. Artifact verification: CHANGELOG.md entry, version manifest update, release commit, git tag (unless --no-tag) all MUST exist before reporting completion.
+</invariants>
+
 <error_codes>
 | Code | Severity | Condition | Recovery |
 |------|----------|-----------|----------|

package/.codex/skills/maestro-next/SKILL.md

@@ -130,8 +130,7 @@ ls -la .workflow/.maestro/ 2>/dev/null | head -5  # 进行中的 session
 | 有 roadmap，未启动 phase | analyze | `maestro-analyze {phase}` |
 | 最新 artifact = analyze | plan | `maestro-plan {phase}` |
 | 最新 artifact = plan | execute | `maestro-execute {phase}` |
-| 最新 artifact = execute | verify | `maestro-verify {phase}` |
-| verify passed | review | `quality-review {phase}` |
+| 最新 artifact = execute | review | `quality-review {phase}` |
 | review verdict=PASS | test-gen | `quality-auto-test {phase}` |
 | 测试全绿 | milestone-audit | `maestro-milestone-audit` |
 | 当前 milestone 全 phase 完成 | milestone-complete | `maestro-milestone-complete` |
@@ -140,7 +139,7 @@ ls -la .workflow/.maestro/ 2>/dev/null | head -5  # 进行中的 session
 **Maestro Lifecycle 主线：**
 ```
 brainstorm → blueprint → init → analyze-macro → roadmap
-   → [per phase] analyze → plan → execute → verify
+   → [per phase] analyze → plan → execute
    → [quality gate] review → auto-test → test
    → milestone-audit → milestone-complete → milestone-release
 ```
@@ -175,7 +174,7 @@ brainstorm → blueprint → init → analyze-macro → roadmap
 | 分析 / analyze / 多维度调研 | `maestro-analyze` |
 | 规划 / plan / 任务分解 | `maestro-plan` |
 | 实现 / 执行 / execute | `maestro-execute` |
-| 验证 / verify / 验收 | `maestro-verify` |
+| 验证 / verify / 验收 | `quality-review` |
 | 调试 / debug / 排查 / bug | `quality-debug` |
 | 审查 / review / 代码审查 | `quality-review` |
 | 测试 / test / UAT | `quality-test` / `quality-auto-test` |

package/.codex/skills/maestro-plan/SKILL.md

@@ -88,7 +88,7 @@ Write resolved milestone into PLN artifact registration and `plan.json.milestone
 **Session**: `.workflow/.csv-wave/{YYYYMMDD}-plan-P{N}-{slug}/`
 **Scratch**: `.workflow/scratch/{YYYYMMDD}-plan-P{N}-{slug}/` (.task/ subdir)
 
-**Pre-load** (optional): context-package.json (via `--from`, takes precedence), context.md (prior analyze), conclusions.json, codebase ARCHITECTURE.md, `maestro wiki search`, `maestro spec load --category arch`, team preflight `maestro collab preflight`.
+**Pre-load** (optional): context-package.json (via `--from`, takes precedence), context.md (prior analyze), conclusions.json, codebase ARCHITECTURE.md, `maestro search`, `maestro spec load --category arch`, team preflight `maestro collab preflight`.
 </context>
 
 <csv_schema>
@@ -108,6 +108,9 @@ Wave 1: N exploration rows (parallel). Wave 2: 1 planning row (sequential).
 3. **Discovery board append-only**: Never modify/delete
 4. **Skip on failure**: If all explorations fail, planner proceeds with available context
 5. **DO NOT STOP**: Continuous until all waves complete
+6. **Invariant violation = BLOCK** — violating any invariant above blocks the current operation.
+7. **Verifiable convergence criteria required** — every task MUST have convergence.criteria[] with grep-verifiable conditions (no subjective language like "well-structured" or "properly implemented"). If any task lacks verifiable criteria: DO NOT report completion — fix the criteria first.
+8. **Artifact verification before completion** — plan.json and .task/TASK-*.json files MUST exist. PLN artifact MUST be registered in state.json. If any missing: DO NOT report completion.
 </invariants>
 
 <state_machine>
@@ -226,7 +229,7 @@ Collision detection against same-milestone plans.
 | Condition | Recovery |
 |-----------|----------|
 | No args and no roadmap and no analyze artifact in state.json | Provide phase number, topic, or run analyze first |
-| --gaps but no gap source | Run maestro-verify first |
+| --gaps but no gap source | Run maestro-execute first |
 | Planning agent fails | Retry once with simplified context |
 | Plan-checker exceeds 3 rounds | Accept with warnings |
 </error_codes>

package/.codex/skills/maestro-quick/SKILL.md

@@ -51,7 +51,7 @@ Extract from arguments:
 
 Read `.workflow/state.json` and `.workflow/project.md` if they exist. If `.workflow/` does not exist, create minimal scratch structure anyway (quick works without full init).
 
-Load coding specs + tools if available: `maestro spec load --category coding`. If the task involves frontend/UI work (description contains component, page, style, layout, CSS, HTML, frontend), also run `maestro spec load --category ui`. Browse wiki: `maestro wiki list --category coding`, load relevant entries via `maestro wiki load <id>`. All optional — proceed without if unavailable.
+Load coding specs + tools if available: `maestro spec load --category coding`. If the task involves frontend/UI work (description contains component, page, style, layout, CSS, HTML, frontend), also run `maestro spec load --category ui`. Browse wiki: `maestro search --category coding`, load relevant entries via `maestro wiki load <id>`. All optional — proceed without if unavailable.
 
 ### Step 3: Create Scratch Directory
 
@@ -106,6 +106,9 @@ Display report: task description, scratch path, status (completed/completed-with
 
 </execution>
 
+### Artifact Verification
+Before reporting completion, verify: plan.json exists AND .summaries/TASK-*-summary.md exists for each task. Task summaries MUST include concrete evidence of completion. If missing: DO NOT report completion.
+
 <error_codes>
 
 | Code | Severity | Description | Recovery |

package/.codex/skills/maestro-ralph/SKILL.md

@@ -78,6 +78,8 @@ Remaining                        → intent
 10. **planning_mode governs arg granularity** — `unified` → skill args 无 `{phase}`；`independent` → 含 `{phase}`
 11. **task_decomposition 驱动 steps[] 动态生长** — `post-goal-audit` 按 unmet 子目标插入 scoped mini-loop；字段可选/累加，既有字段不删不改
 12. **Platform** — `session.platform = "codex"`；CLI 调用一律带 `--platform codex`
+13. **Invariant violation = BLOCK** — violating any invariant above blocks the current operation. Do NOT bypass for "efficiency" or "clear intent" reasons. Especially invariants about ralph never executing steps and completion_confirmed by CLI.
+14. **Delegate fallback must be marked** — when A_DELEGATE_EVALUATE verdict parse fails and falls back to "fix", MUST record `parse_failed: true, confidence_score: 0` in decisions.ndjson. Subsequent steps inherit LOW CONFIDENCE flag.
 </invariants>
 
 <state_machine>
@@ -160,7 +162,7 @@ S_DISPATCH:
   → END             DO: $maestro-ralph-execute
 
 S_DECISION_EVAL: (decision 节点 == `step.decision` 非空，下述 gate 名取自该字段)
-  → S_APPLY_VERDICT WHEN: quality-gate (post-verify, post-business-test, post-review, post-test)
+  → S_APPLY_VERDICT WHEN: quality-gate (post-execute, post-business-test, post-review, post-test)
                      DO: A_DELEGATE_EVALUATE
   → S_APPLY_VERDICT WHEN: goal-gate (post-goal-audit)
                      DO: A_GOAL_AUDIT_EVALUATE
@@ -267,8 +269,7 @@ resolve_milestone(phase_number):
 | phase 已存在 + 无任何 artifact | `analyze` |
 | phase 已存在 + 最新 artifact = analyze | `plan` |
 | phase 已存在 + 最新 artifact = plan | `execute` |
-| phase 已存在 + 最新 artifact = execute | `verify` |
-| phase 已存在 + 最新 artifact = verify | → refine from result files |
+| phase 已存在 + 最新 artifact = execute | `review` |
 
 **关键不变量**：artifact 过滤按 `session.phase`，不读 `state.json.current_phase`。`phase_is_new` → 直接 `analyze`。
 
@@ -287,12 +288,10 @@ resolve_milestone(phase_number):
 | `medium` / `small` | analyze-macro → plan --from analyze:{ANL_ID} → execute → ...（跳过 roadmap + analyze-phase） |
 | `unknown` | 默认走 large 路径，post-analyze-scope 决策节点再纠正 |
 
-**Refine from verify results:**
+**Refine from review results:**
 
 | Condition | Position |
 |-----------|----------|
-| verification.json: passed==false or gaps[] | `verify-failed` |
-| passed==true, no review.json | `business-test` |
 | review.json: verdict=="BLOCK" | `review-failed` |
 | review.json: verdict!="BLOCK" | `test` |
 | uat.md: all passed | `milestone-audit` |
@@ -302,7 +301,7 @@ resolve_milestone(phase_number):
 
 决定下游质量管线长度。读 `session.quality_mode_override`（CLI 标志 `--quality`），无则按规则推断：
 
-| Condition | Mode | Pipeline (verify 之后) |
+| Condition | Mode | Pipeline (execute 之后) |
 |-----------|------|-------------------------|
 | Has `specs/REQ-*.md` + 当前 phase 业务范围明确 | `full` | business-test → review → test-gen → test |
 | Default | `standard` | review → test-gen (当 coverage<80%) → test |
@@ -389,8 +388,7 @@ Generate steps from `session.lifecycle_position` to `milestone-complete`.
 | roadmap | `maestro-roadmap --from analyze:{analyze_macro_id}` | *(same)* | — | all |
 | analyze | `maestro-analyze {phase}` | `maestro-analyze` | — | all |
 | plan | `maestro-plan {phase}` *(scope=phase)* / `maestro-plan --from analyze:{analyze_macro_id}` *(scope=standalone)* / `maestro-plan --from blueprint:{blueprint_id}` *(scope=standalone)* | `maestro-plan` | — | all |
-| execute | `maestro-execute {phase}` | `maestro-execute` | — | all |
-| verify | `maestro-verify {phase}` | `maestro-verify` | `post-verify` | all |
+| execute | `maestro-execute {phase}` | `maestro-execute` | `post-execute` | all |
 | business-test | `quality-auto-test {phase}` | `quality-auto-test` | `post-business-test` | full only |
 | review | `quality-review {phase}` | `quality-review` | `post-review` | all (quick: append `--tier quick`) |
 | test-gen | `quality-auto-test {phase}` | `quality-auto-test` | — | full / standard if coverage<80% |
@@ -446,7 +444,7 @@ Generate steps from `session.lifecycle_position` to `milestone-complete`.
 3. Map result files:
    | Decision | Files |
    |----------|-------|
-   | post-verify | verification.json |
+   | post-execute | verification.json |
    | post-business-test | .tests/auto-test/report.json |
    | post-review | review.json |
    | post-test | uat.md, .tests/test-results.json |
@@ -618,9 +616,9 @@ Runs only when `task_decomposition` present.
     "index": 0,
     "skill": "",                  // 执行 step 有值；decision 节点为空字符串/null
     "args": "",
-    "stage": "",                  // brainstorm|blueprint|init|analyze-macro|roadmap|analyze|plan|execute|verify|...
+    "stage": "",                  // brainstorm|blueprint|init|analyze-macro|roadmap|analyze|plan|execute|...
     "scope": null,                // "phase"|"standalone"|"milestone"|null（plan 等需要）
-    "decision": null,             // 非 null → decision 节点（值为 gate 名，如 "post-verify"）；null → 执行 step
+    "decision": null,             // 非 null → decision 节点（值为 gate 名，如 "post-execute"）；null → 执行 step
     "retry_count": 0,             // decision 节点专用
     "max_retries": 2,             // decision 节点专用
     "command_scope": "global|project|missing|null",  // 执行 step；decision 节点固定 null
@@ -658,13 +656,12 @@ Runs only when `task_decomposition` present.
 
 所有插入的执行 step 按 A_BUILD_STEPS 规则 9 解析 `command_path` + `command_scope`；`decision:*` 条目为 decision 节点（`step.decision` 字段）。
 
-**post-verify:**
+**post-execute:**
 ```
 quality-debug "{gap_summary}"
 maestro-plan --gaps {phase}
 maestro-execute {phase}
-maestro-verify {phase}
-decision:post-verify {retry+1}
+decision:post-execute {retry+1}
 ```
 
 **post-business-test:**
@@ -672,8 +669,7 @@ decision:post-verify {retry+1}
 quality-debug --from-business-test "{gap_summary}"
 maestro-plan --gaps {phase}
 maestro-execute {phase}
-maestro-verify {phase}
-decision:post-verify {retry: 0}
+decision:post-execute {retry: 0}
 quality-auto-test {phase}
 decision:post-business-test {retry+1}
 ```
@@ -692,8 +688,7 @@ decision:post-review {retry+1}
 quality-debug --from-uat "{gap_summary}"
 maestro-plan --gaps {phase}
 maestro-execute {phase}
-maestro-verify {phase}
-decision:post-verify {retry: 0}
+decision:post-execute {retry: 0}
 quality-auto-test {phase}
 decision:post-business-test {retry: 0}
 quality-review {phase}
@@ -708,7 +703,6 @@ decision:post-test {retry+1}
 # for each unmet sub-goal G{n}, scoped to target_phase:
 maestro-plan --gaps {target_phase} "G{n}: {gap}"     [goal_ref: G{n}]
 maestro-execute {target_phase}                       [goal_ref: G{n}]
-maestro-verify {target_phase}                        [goal_ref: G{n}]
 # after all unmet groups inserted:
 decision:post-goal-audit {retry+1}
 ```