Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Maestro Flow 命令帮助系统。搜索命令、浏览技能、工作流推荐、新手引导。Triggers on "maestro-help", "帮助", "命令", "怎么用", "skill", "workflow", "maestro 怎么用".
.claude/skills/catlog22-maestro-help/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 137% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 552% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 197% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 261% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 250% | 0% |
Maestro Flow 命令帮助系统，提供命令搜索、技能浏览、工作流推荐、新手引导功能。
/maestro-help, /maestro-help search <keyword>, /maestro-help skills, /maestro-help guide┌──────────────────────────────────────────────────────────────────┐
│ Maestro Help (SKILL.md) — Orchestrator │
│ → Parse intent → Route to mode → Execute phase → Present │
└────────────────────────┬─────────────────────────────────────────┘
│
┌────────────────────┼────────────────────────┐
↓ ↓ ↓
┌──────────┐ ┌──────────────┐ ┌──────────┐
│ Phase 1 │ │ Phase 2 │ │ Phase 3 │
│ Parse │─────→│ Search & │────────→│ Workflow │
│ Intent │ │ Present │ │ Guide │
└──────────┘ └──────────────┘ └──────────┘
↑ ↗ │
└──────┘ ↓
(refine search) present guideindex/catalog.json，不做硬编码guide/ 目录中的参考文档Single source of truth: index/catalog.json
| Field | Purpose | |-------|---------| | commands[] | 64 个 slash 命令，含分类和描述 | | skills[] | 23 个 Skill（含 10 个选装 scholar-，标 optional: true），含分类和描述 | | agents[] | 24 个 Agent，含分类和描述 | | cli_commands[] | 21 个终端命令 | | guide_files[] | 17 个 Guide 文档索引（planned，尚未创建） | | essential_commands[] | 10 个核心命令（新手用） | | workflows | 主干管线、Companion 轻量入口、Issue 闭环、初始化路径 |
Triggers: "搜索命令", "find command", "search", 命令名关键词
Process:
Ref: phases/01-parse-intent.md — 解析搜索意图catalog.json commands] + cli_commands]Triggers: "怎么用", "how to use", "详情", 具体命令名
Process:
catalog.jsonsource path（从 catalog 相对路径）Triggers: "下一步", "what's next", "推荐", "继续"
Process:
Triggers: "工作流", "workflow", "怎么开始", "用什么流程"
Process:
Ref: phases/03-workflow-guide.mdTriggers: "新手", "getting started", "常用命令", "入门"
Process:
catalog.json essential_commands]Triggers: "skill", "agent", "技能", "有哪些 skill", "团队"
Process:
Ref: phases/02-search-present.mdcatalog.json skills] 或 agents]Triggers: "终端命令", "CLI", "maestro 命令", "terminal"
Process:
catalog.json cli_commands]Input: $ARGUMENTS (free text)
Phase 1: Parse Intent
└─ Ref: phases/01-parse-intent.md
├─ 分析关键词确定 operation mode
├─ 提取搜索词 / 命令名 / 分类过滤
└─ Output: { mode, query, category?, context? }
Phase 2: Search & Present (Mode 1/2/3/6/7)
└─ Ref: phases/02-search-present.md
├─ 查询 catalog.json
├─ 按模式过滤和排序
├─ 读取 source 文件（Mode 2）
└─ Output: 格式化结果
Phase 3: Workflow Guide (Mode 4/5)
└─ Ref: phases/03-workflow-guide.md
├─ 检测项目状态
├─ 匹配工作流模板
├─ 生成推荐命令序列
└─ Output: 引导信息Phase Reference Documents (read on-demand):
| Phase | Document | Purpose | |-------|----------|---------| | 1 | phases/01-parse-intent.md | 意图解析和模式路由 | | 2 | phases/02-search-present.md | 搜索和呈现 | | 3 | phases/03-workflow-guide.md | 工作流推荐和引导 |
$ARGUMENTS → Parse:
├─ "search <keyword>" → Mode 1: Command Search
├─ 命令名 (如 "analyze") → Mode 2: Documentation
├─ "下一步" / "next" → Mode 3: Smart Recommendations
├─ "工作流" / "workflow" → Mode 4: Workflow Guide
├─ "新手" / "入门" → Mode 5: Beginner Onboarding
├─ "skill" / "agent" → Mode 6: Skill & Agent Browsing
├─ "CLI" / "终端" → Mode 7: CLI Reference
├─ 空参数 → Mode 5: Beginner Onboarding
└─ 其他自由文本 → Mode 1: Command Search (fuzzy)> 裸名称为 first-tier step：经 /maestro "<意图>" 自动路由，或 maestro run create <step> [args...] --session {session_id} ... 直接执行（v3 无独立 run prepare 命令——prepare 指引内嵌于 birth packet 的 guidance）；/ 前缀为独立命令。
| 命令 | 用途 | |------|------| | /maestro | 智能协调器，自动路由 | | /maestro-init | 项目初始化 | | brainstorm | 头脑风暴 — 发散探索，多角色创意 | | blueprint | 正式规格文档化 — 7-phase 收敛规格链 | | roadmap | 路线图编排 — 消费上游 context，纯 Milestone > Phase 分解 | | /maestro-companion | 轻量任务直接执行 | | /maestro-overlay | Overlay 管理 — 自然语言创建，或 --amend 从信号自动生成修正补丁 | | grill | 压力测试 — 对计划或需求进行代码库现实性压力测试 | | /maestro-next | 智能导航 — 检测状态并推荐下一步最优命令 | | /maestro-ralph --engine swarm | Swarm 并行加速器 — 多 agent 并发执行 | | /maestro-ralph --engine universal | 动态对抗工作流生成器 |
| 命令 | 用途 | |------|------| | analyze | 双层分析 — 宏观(文本参数)探索影响面 / 微观(数字参数)Phase 级深入 | | plan | 任务规划 — 支持 --from analyze:ANL-xxx 直达 | | execute | 任务执行 |
| 命令 | 用途 | |------|------| | review | 代码审查 | | auto-test | 自动测试 | | test | 业务测试 | | debug | 质量调试 | | retrospective | 复盘 |
| 命令 | 用途 | |------|------| | /maestro-issue | Issue 管理（意图驱动：报告/列出/关闭/关联） | | /maestro-issue discover | Issue 发现 | | /maestro-knowledge | 知识存储管理（意图驱动：audit/harvest/wiki/extractors/domain） | | /maestro-knowhow | KnowHow 沉淀（意图驱动：按类型捕获可复用知识） |
单入口 /maestro-odyssey <intent> --mode <name>（--mode 可省略，从 intent 关键词自动识别）：
| 模式 | 用途 | |------|------| | --mode debug | 长周期调试 — 考古、诊断、修复、泛化 | | --mode improve | 长周期代码改进 — 多维审计、深度诊断、定向修复 | | --mode review | 深度审查修复循环 | | --mode planex | 需求驱动迭代 — 计划/执行/验证/修复循环 | | --mode ui | 长周期 UI 优化 — 视觉调研、多维审计、修复 |
Roadmap > Milestone > Phase > Task上游起源层（并列，可选）
brainstorm（发散/轻量） | blueprint（收敛/重型） | grill（压力测试）
理解层
analyze 双层: 宏观(文本参数) → scope_verdict | 微观(数字参数) → Phase 级决策
编排层（可选）
roadmap — 消费上游 context，纯 Milestone > Phase 分解
执行层
plan → execute
Odyssey 长周期循环（独立路径）
maestro-odyssey --mode debug|improve|review|planex|ui
自适应引擎（高级）
ralph → 自运行决策循环
swarm-workflow / universal-workflow → 多 agent 并行执行| 路径 | 场景 | 命令序列 | |------|------|---------| | Path A | 完整新项目 | brainstorm → blueprint(可选) → analyze "topic" → roadmap → analyze 1 → plan 1 → execute | | Path B | 旧项目大功能 | analyze "feature X" → roadmap → analyze 1 → plan 1 → execute | | Path C | 中等功能 | analyze "feature X" → plan --from analyze:ANL-xxx → execute | | Path D | 小改动 | plan "fix auth bug" → execute | | Path E | 纯规格文档 | blueprint "project idea" → (供人阅读) | | Path F | 纯探索 | brainstorm "idea" → (供人决策) | | 轻量修复 | 已知简单问题 | /maestro-companion "修复描述" | | Bug 追踪 | Issue 闭环 | /maestro-issue discover → /maestro-issue create → analyze/plan/execute → close | | 全自动 | /maestro 入口 | /maestro -y "任务描述" | | 代码审查 | 质量管线 | review → auto-test → test | | 多 CLI 交叉验证 | Collab step | collab "需求描述" | | 长周期调试 | Odyssey 深度循环 | /maestro-odyssey "问题描述" --mode debug | | 长周期改进 | Odyssey 深度循环 | /maestro-odyssey "改进目标" --mode improve | | 需求迭代 | Odyssey 深度循环 | /maestro-odyssey "需求描述" --mode planex |
> 注：Path A/B 中 analyze 1 / plan 1 的数字 1 指 milestone 编号（第 1 个里程碑），下游以 milestone 为入口。
| 场景 | 处理 | |------|------| | 命令未找到 | 模糊搜索最近匹配，提示正确命令名 | | 项目未初始化 | 推荐先运行 /maestro-init | | Guide 文件不存在 | 跳过，仅提供 catalog 中的描述 | | 参数为空 | 默认进入 Beginner Onboarding 模式 |
guide/ — 17 个专题指南文档~/.maestro/workflows/delegate-usage.md~/.maestro/workflows/coding-philosophy.md~/.maestro/cli-tools.jsonoptional: true）| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 17,259 | 14,567 | -16% | 1 | 1 | 0% | 2,566 | 6,074 | +137% | 0 | 0 | — |
case-02 | fail→pass | 6,002 | 11,174 | +86% | 1 | 1 | 0% | 830 | 5,414 | +552% | 0 | 0 | — |
case-03 | fail→fail | 14,733 | 15,983 | +8% | 1 | 1 | 0% | 2,322 | 6,315 | +172% | 0 | 0 | — |
case-04 | fail→pass | 10,483 | 5,398 | -49% | 1 | 1 | 0% | 1,515 | 4,500 | +197% | 0 | 0 | — |
case-05 | fail→pass | 7,790 | 7,691 | -1% | 1 | 1 | 0% | 1,351 | 4,879 | +261% | 0 | 0 | — |
case-06 | fail→pass | 9,275 | 5,651 | -39% | 1 | 1 | 0% | 1,275 | 4,462 | +250% | 0 | 0 | — |
case-07 | fail→pass | 12,258 | 3,267 | -73% | 1 | 1 | 0% | 1,847 | 4,100 | +122% | 0 | 0 | — |
case-08 | fail→pass | 12,236 | 5,962 | -51% | 1 | 1 | 0% | 1,812 | 4,453 | +146% | 0 | 0 | — |
case-09 | fail→pass | 15,365 | 5,416 | -65% | 1 | 1 | 0% | 2,325 | 4,421 | +90% | 0 | 0 | — |
case-10 | pass→pass | 8,554 | 8,597 | +1% | 1 | 1 | 0% | 1,253 | 4,995 | +299% | 0 | 0 | — |
case-11 | fail→pass | 7,960 | 2,560 | -68% | 1 | 1 | 0% | 1,047 | 3,958 | +278% | 0 | 0 | — |
case-12 | fail→pass | 12,242 | 2,746 | -78% | 1 | 1 | 0% | 1,904 | 4,086 | +115% | 0 | 0 | — |
case-13 | fail→pass | 9,641 | 5,920 | -39% | 1 | 1 | 0% | 1,406 | 4,585 | +226% | 0 | 0 | — |
case-14 | fail→pass | 13,859 | 2,676 | -81% | 1 | 1 | 0% | 2,039 | 4,018 | +97% | 0 | 0 | — |
case-15 | fail→pass | 9,398 | 2,748 | -71% | 1 | 1 | 0% | 1,353 | 4,023 | +197% | 0 | 0 | — |
case-16 | fail→pass | 13,049 | 6,400 | -51% | 1 | 1 | 0% | 1,859 | 4,494 | +142% | 0 | 0 | — |
case-17 | fail→pass | 10,559 | 5,161 | -51% | 1 | 1 | 0% | 1,678 | 4,429 | +164% | 0 | 0 | — |
case-23 | fail→pass | 3,861 | 7,509 | +94% | 1 | 1 | 0% | 538 | 4,787 | +790% | 0 | 0 | — |
case-18 | fail→pass | 8,748 | 4,094 | -53% | 1 | 1 | 0% | 1,219 | 4,270 | +250% | 0 | 0 | — |
case-19 | fail→pass | 10,863 | 2,843 | -74% | 1 | 1 | 0% | 1,654 | 4,028 | +144% | 0 | 0 | — |
case-20 | fail→pass | 3,609 | 3,174 | -12% | 1 | 1 | 0% | 514 | 4,089 | +696% | 0 | 0 | — |
case-21 | pass→pass | 12,417 | 5,101 | -59% | 1 | 1 | 0% | 1,785 | 4,382 | +145% | 0 | 0 | — |
case-22 | pass→pass | 15,594 | 6,395 | -59% | 1 | 1 | 0% | 2,403 | 4,515 | +88% | 0 | 0 | — |
case-24 | fail→fail | 7,747 | 7,111 | -8% | 1 | 1 | 0% | 1,325 | 4,797 | +262% | 0 | 0 | — |
case-25 | fail→pass | 3,020 | 10,169 | +237% | 1 | 1 | 0% | 473 | 5,138 | +986% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 25 cases were attempted. The headline lift of +80 percentage points is the difference between those two pass rates over the 25 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.