Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

yy-flow

ASecurity

适用于 AI 多 Agent 与人类团队协同研发的多角色状态流转、质量审计、防错闭环与看板自动化工作流技能包。可通过 /yy-flow help、/yy-flow start、/yy-flow status、/yy-flow kanban、/yy-flow trace 快捷指令或自然语言唤醒。

20 stars
0 votes
0 copies
0 views
Added 9/19/2026
ai-agentspythongogitfrontenddevops

Works with

cli

Security Analysis

A100/100

Pro scans all 20 files and shows the line behind each finding

Scanned 10/3/2026

$npx -y skills add YuanYii/multi-agent-flow --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of yy-flow?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for yy-flow
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/yuanyii-yy-flow/badge)](https://www.skillsdirectory.com/skills/yuanyii-yy-flow)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
Files
SKILL.md
---
name: yy-flow
description: 适用于 AI 多 Agent 与人类团队协同研发的多角色状态流转、质量审计、防错闭环与看板自动化工作流技能包。可通过 /yy-flow help、/yy-flow start、/yy-flow status、/yy-flow kanban、/yy-flow trace 快捷指令或自然语言唤醒。
version: 2.2.0
---

# Multi-Agent Team Workflow Skill (YY-Flow)

> **契约驱动的多智能体协同研发微内核 (YY-Flow)** —— 对齐 AutoGen 范式,有效防范越权、打回碎片化与状态悬挂。

---

## 统一入口网关协议 (Unified Gateway Protocol)

所有调用本 Skill 的指令统一收敛至本网关入口,按如下双支线规则分流与判定:

### 1. 双支线分流矩阵

| 分支类型 | 触发特征与指令 | 处理机制与响应契约 | 典型场景 |
| :--- | :--- | :--- | :--- |
| **分支 A:系统控制台操作** | `/yy-flow help`<br>`/yy-flow start`<br>`/yy-flow status`<br>`/yy-flow kanban`<br>`/yy-flow sync-pr`<br>`/yy-flow trace` | **直通底层运维引擎**(0 前置分级),直接执行对应脚本并输出系统状态。 | 查看帮助手册、初始化 SOP、启动看板、大盘巡检、PR 解阻、离线链路图鉴 |
| **分支 B:业务研发与交互意图** | 自然语言需求、代码修改、特性开发、缺陷修复、排查排障、`/yy-flow auto` | **强制前置执行【分级三问与单一职责拆分】**,并在回复首行显式输出双标头标识:<br>`【任务分级: L0 即时问答 / L1 短链任务 / L2 标准研发】 | 【单一职责: 合规单卡 / 复合需求需拆解为 N 个原子任务】` | “写个登录接口”、“优化下正则”、“解释函数逻辑” |

### 2. 业务需求【分级三问】决策流

```text
[业务需求输入]
   │
   ├─► 1. 会留下代码/文档/配置等仓库文件修改吗?
   │      ├─ 否 ──► 2. 结论需要作为后续工程依据事后追溯吗?
   │      │            ├─ 否 ───────────────► 【L0 即时问答】(无卡直接作答,0 毫秒 CLI 延迟,草稿入草稿箱)
   │      │            └─ 是 ───────────────► 【L1 轻量任务】(建卡走短链:单角色完成 → 提请用户验收)
   │      │
   │      └─ 是 ──► 3. 需要多角色协作、修改核心业务代码或变更公共基础库吗?
   │                   ├─ 否 ───────────────► 【L1 轻量任务】
   │                   └─ 是 ───────────────► 【L2 标准研发任务】(建卡走全链:DEV 编码 → Reviewer 审查 → QA 测试 → 提请用户验收)
```

- **网关分级与 CCP 弹性分级映射**:
  - **【L0 即时问答】**:无卡直接作答,0 状态机开销;
  - **【L1 轻量任务】**:映射为 **CCP Tier-3**(Fast-Path 快速放行,物理免独立报告,凭据直接回写看板);
  - **【L2 标准研发】**:根据规模与契约自动匹配 **CCP Tier-2**(局部优化,工时 ≤ 4.0h,免物理报告)或 **CCP Tier-1**(核心架构与接口契约强约束)。
- 凡属于 L1/L2 业务需求,网关强制执行单一职责审查(工时 ≤ 8.0h,单一原子交付物)。跨正交领域需求必须拆解建卡。
- 执行态 Task ID(如 `T0001`)直接穿透网关进入状态机推进,严禁二次建卡。

---

## 核心命令速查表 (CLI Command Quick Reference)

| 业务场景 | 统一 CLI 门面 (`cli.py`) | 底层专用脚本 | 核心作用与参数 |
| :--- | :--- | :--- | :--- |
| **创建任务卡** | `python3 scripts/cli.py task create ...` | `python3 scripts/quick_task.py create ...` | `--name "..." --assignee "..." --stage "Sprint 2.1 菜单调整" --type "A" --target "..." --criteria "..."` |
| **阶段门禁准入/结项** | `python3 scripts/check_stage_gate.py ...` | `python3 scripts/check_stage_gate.py ...` | `--action start/close --stage "Sprint 2.1 菜单调整"`(按需准入,前序依赖自动核验) |
| **代码化派单** | `python3 scripts/cli.py dispatch --task-id T00xx` | `python3 scripts/dispatch_task.py` | 校验依赖与并发,推至【进行中】,输出 Subagent 载荷 |
| **推进任务流转** | `python3 scripts/transition_task.py ...` | `python3 scripts/transition_task.py ...` | `--task-id T00xx --role DEV --from-status 进行中 --to-status 审查中 --token <token> --remarks "..."` |
| **人类终态验收** | `python3 scripts/cli.py task accept ...` | `python3 scripts/quick_task.py accept ...` | `--task-id T00xx`(**人类用户专属**,严禁 Agent 代签) |
| **启动看板** | `python3 scripts/cli.py kanban` | `python3 scripts/start_kanban_server.py` | 默认启动于 `http://127.0.0.1:32886/` |
| **健康度巡检** | `python3 scripts/cli.py status` | `python3 scripts/heartbeat.py` | 输出大盘健康度、阻塞卡片与效能指标 |
| **技术栈同步/画像重算** | - | `scripts/update_agent_tech_stacks.py` | `--force-recalc` 强制重算专家技术能力画像并刷新导出 Subagent,解决技术栈迁移后画像陈旧问题 |
| **连续性校验** | `python3 scripts/cli.py ccp ...` | `python3 scripts/cli.py ccp ...` | `--task-id T00xx --stage 审查中` 校验上下文连续性 |
| **链路全景图鉴** | `python3 scripts/cli.py trace` | `python3 scripts/generate_trace_html.py` | 离线解析会话轨迹,秒级生成全景链路图鉴 HTML (0 Token) |
| **智能增量测试** | `python3 scripts/cli.py test` | `scripts/run_tests.py` | 自动探测 Git 改动,仅执行相关单测 (秒级),文档变更自动豁免,`--all` 全量回归 |

---

## 8 大专家子代理写权限与流转区间矩阵

> [!IMPORTANT]
> **物理隔离铁律与令牌门控 (Dispatch Token Gating)**:
> 1. 主 Agent (PM 严经理) 严禁在主会话中自扮演研发编写业务代码!涉及代码变更的任务,**必须通过 `invoke_subagent` 派发对应专家子代理在独立进程中完成**。
> 2. **代码化派发与令牌门控**:任务派发时由 `cli.py dispatch` 生成专属 `dispatch_token`。专业研发角色 (`DEV`/`FRONTEND`/`REVIEWER`/`QA`) 流转时必须携带 `--token <token>`,状态机执行底层强拦截,杜绝自扮演假绿走过场。

| 子代理标识 | 角色名称 | 核心职责 | 合法状态流转区间 | 退出契约 |
| :--- | :--- | :--- | :--- | :--- |
| `@flow-pm` | 严经理 (项目经理) | 需求拆解、分级判定与派单 | 待开始 -> 进行中 / 提请验收 | 终态停在【已完成】,提请人类验收 |
| `@flow-architect`| 钱架构 (系统架构师) | 系统总体架构、接口契约与选型 | 待开始 -> 进行中 -> 已完成 | 产出 ADR 与架构方案文档 |
| `@flow-dev` | 李开发 (开发工程师) | 后端/全栈核心编码与单测实现 | 待开始 -> 进行中 -> 审查中 | 必须运行单测,生成开发报告 |
| `@flow-frontend` | 马前端 (前端开发工程师) | Web/UI 组件与前端体验实现 | 待开始 -> 进行中 -> 审查中 | 验证组件渲染与交互,生成报告 |
| `@flow-reviewer` | 周审查 (代码审查专家) | 代码规范、安全扫描与质量门控 | 审查中 -> 测试中 / 审查中 -> 已退回 | 静态扫描无告警,产出审查意见 |
| `@flow-qa` | 章测试 (测试工程师) | 针对性集成测试与准出(仅执行当前任务相关测试,严禁全量脚本) | 测试中 -> 已完成 / 测试中 -> 已退回 | 编写针对性测试用例,记录实际工时 |
| `@flow-docs` | 李文通 (文档工程师) | 交付物文档治理与用户手册 | 待开始 -> 进行中 -> 已完成 | 遵循 GB/T 7713 学术与工程规范 |
| `@flow-devops` | 吕改特 (运维管理员) | 分支合流、发布构建与 Git 历史保护 | 待开始 -> 进行中 -> 已完成 | 验证流水线通过,解决 PR 阻塞 |

---

## 模块化规约实体索引 (按需查阅具体规范文件)

当且仅当进入特定业务场景时,Agent 必须调用 `view_file` 查阅对应规约的实体 Markdown 文件:
- **初次加载 / 冷启动 SOP**:调阅 [`references/02-bootstrap/01-initialization_sop.md`](references/02-bootstrap/01-initialization_sop.md)
- **创建任务 / 单一职责判定**:调阅 [`references/01-gateway/01-task_classification.md`](references/01-gateway/01-task_classification.md)
- **任务流转 / 门控规则 / 角色权限**:调阅 [`references/03-engine/01-state_flow_rules.md`](references/03-engine/01-state_flow_rules.md)
- **上下文交接 / 跨 Agent Handoff (CCP)**:调阅 [`references/04-ccp/01-ccp_protocol_spec.md`](references/04-ccp/01-ccp_protocol_spec.md)
- **看板服务运维 / PR 自动解阻**:调阅 [`references/05-kanban/01-kanban_and_pr_sync.md`](references/05-kanban/01-kanban_and_pr_sync.md)
- **Git 分支合流 / 审计日志与度量**:调阅 [`references/06-governance/01-git_and_audit_spec.md`](references/06-governance/01-git_and_audit_spec.md)

Attribution

YuanYiiYuanYii
View sourceSee grades on GitHubMore from YuanYii →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

1100021 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

698621 votes

Writing Skills

Create and manage Claude Code skills in HASH repository following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns), UserPromptSubmit hook, and the 500-line rule. Includes validation and debugging with SKILL_DEBUG. Examples include rust-error-stack, cargo-dep...

3931 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3421 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Amp, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Grok Build, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

741 votes
View all in ai-agents →