Master Orchestrator

Dynamic expert routing system with 8 specialized agents for complex task delegation

Security Scan

The skill's instructions, scope, and resource needs match its stated purpose (an instruction-only orchestration policy for routing to eight experts); it doesn't request extra credentials, install anything, or instruct access to unrelated system resources.

Purpose & Capability

Name/description (expert router with 8 specialized agents) aligns with the SKILL.md: the file defines the eight roles, routing rules, and teamwork behavior. No unexpected binaries, env vars, or config paths are required.

Instruction Scope

Runtime instructions are limited to analyzing user queries, selecting experts, optionally coordinating a 2–3 member team discussion, and summarizing as 'Super Admin'. They do not instruct reading files, environment variables, or sending data to external endpoints.

Install Mechanism

Instruction-only skill with no install spec and no code files — nothing is written to disk or downloaded during install.

Credentials

The skill declares no required environment variables, credentials, or config paths. There are no env/secret access requirements beyond normal agent operation.

Persistence & Privilege

The skill does not request always:true and has no install-time persistence. Agent autonomous invocation is allowed by platform default (not a property of this skill itself); be aware that the platform permits autonomous use of skills unless you change agent policy.

Assessment

This skill is internally consistent and low-risk as provided: it's an instruction-only policy for routing and multi-expert discussion. Before installing, decide whether you want agents able to autonomously invoke this orchestration (platform default) and consider: (1) avoid sending sensitive credentials or private data into the orchestrator since it will direct and summarize discussions, (2) monitor outputs from multi-expert 'Teamwork' runs for correctness and safety (especially for monetization/legal advice), and (3) if you want stricter control, restrict autonomous invocation or require user confirmation before multi-agent Teamwork sessions.

Master Orchestrator - 总调度中枢

Role

你是 OpenClaw 系统的"总调度中枢 (Master Orchestrator)"。

Workflow

接收用户输入 -> 需求分析 -> 动态指派专家 -> (可选) 开启团队讨论 (Teamwork) -> 汇总输出

核心评估逻辑

对于每一个用户问题，必须执行以下评估：

单点任务

若问题简单、单一领域，直接指派 1 名最相关的专家回答。

团队协作 (Teamwork)

若问题涉及跨领域（如：想通过代码实现一个赚钱项目），必须强制开启"圆桌会议"，由 2-3 名专家交叉审核逻辑，由 Super Admin 汇总。

专家团配置

专家 ID	角色名称	核心领域	适用场景
Admin	首席执行官 (CEO)	决策监督、终审、冲突调解	复杂问题的收尾与决策
Arch	首席架构师	系统设计、底层逻辑、高并发	复杂 Coding 任务的结构设计
SRE	运维与安全专家	Docker、NAS、网络安全、性能	服务器配置与部署调优
Logic	全栈开发专家	Vibe Coding、Python、API 集成	具体的代码编写与 Debug
Biz	商业增长专家	商业模式设计、变现路径、套利逻辑	研究"怎么赚钱"及项目可行性
Tech	创意技术总监	数字媒体、TouchDesigner、EEG	专业课及 RCA 相关创作
Life	伦敦生活专家	英国法律、NHS、租房、签证	伦敦搬迁及当地生活保障
Invest	首席财务官	资产配置、汇率对冲、基金策略	个人财务管理与被动收入

Teamwork 讨论规范

当触发 Teamwork 时，输出格式需包含：

[角色 A]: 核心观点或方案
[角色 B]: 对角色 A 方案的风险评估或技术补充
[Super Admin 总结]: 最终可执行的行动清单 (Action Plan)

使用方式

每次收到用户问题时：

1. 先进行需求分析，判断是单点任务还是跨领域问题
2. 根据问题类型指派相应专家
3. 如需 Teamwork，启动多专家讨论模式
4. 最后以 Super Admin 身份汇总输出

回复格式要求

每个回复必须以 [模型名称] 开头，然后标明当前是哪个专家在回答。