edisonzerolamedisonzerolam/openclaw-skills/agent-team-orchestration/SKILL.md

agent-team-orchestration

Skill0 starsChanged 4 months agoMITView on GitHub

Orchestrate multi-agent teams with defined roles, task lifecycles, handoff protocols, and review workflows. 触发词：「组成专家小组」「团队协作」「子代理协作」「team」。不适用：单 agent 任务、一次性 sessions_spawn、简单 Q&A 转发。

3 other projects use a version of this.

Adopted by

Most recently taken up by

vsriram11/wealthtrajectorytheir own version, first seen May 16, 2026
GPTtang/skill-atlastheir own version, first seen Mar 22, 2026
casper7995/ai-agent-solution-architecttheir own version, first seen Feb 24, 2026

Other versions3 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 3
vsriram11/wealthtrajectory★ 8same name
GPTtang/skill-atlas★ 4same name
casper7995/ai-agent-solution-architect★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository5 other files it gives its agents.
- agent-methodologySkill
- agent-plannerSkill
- auditorSkill
- debugSkill
- qclaw-skill-creatorSkill

The file

Create a free account to mount

Files

SKILL.md

需求路由

def route_demand(demand: str) -> str:
    d = demand.lower()
    if any(kw in d for kw in ["PowerShell 错误", "self_heal"]): return "local"
    if any(kw in d for kw in ["retry", "skip"]): return "local"
    if any(kw in d for kw in ["规划团队", "团队设置"]): return "llm"
    if any(kw in d for kw in ["spawn", "派发"]): return "local"
    if any(kw in d for kw in ["健康检查", "评估"]): return "hybrid"
    if any(kw in d for kw in ["协作结果", "汇总"]): return "hybrid"
    return "llm"

Core Concepts

Roles

Role	Purpose	Model guidance
Orchestrator	Route work, track state, priority calls	High-reasoning
Builder	Produce artifacts (code, docs, configs)	Cost-effective
Reviewer	Verify quality, push back on gaps	High-reasoning
Ops	Cron, health checks, dispatching	Cheapest reliable

Task States

Inbox → Pre-task Discussion → Assigned → In Progress → Review → Consensus Check → Done | Failed

Orchestrator owns state transitions. Every transition gets a comment.

Handoffs

Handoff messages must include: What was done / Where artifacts are / How to verify / Known issues / What's next.

Reviews

Cross-role reviews prevent quality drift. Every artifact gets at least one set of eyes that didn't produce it.

Quick Start: Minimal 2-Agent Team

1. Define roles: Orchestrator (you) + Builder
2. Spawn builder with: Task ID, description, output path, handoff instructions
3. On completion: review artifacts, mark done, report
4. Add a reviewer: Builder → Reviewer checks → Orchestrator ships or returns

触发器路由矩阵

场景	工具	理由
2-3 Agent 长期协同，有交接	`team-brain.py launch`	生命周期管理 + checkpoint
≤5 Agent 一次性任务	`sessions_spawn`	轻量
并行同优先级子任务	`sessions_spawn × N`	并行分发
临时单次指令	直接 `sessions_spawn`	最简路径
cron 周期任务	`sessions_spawn` + cron	定时触发

与 agents-config.md 协同

用户说"组成专家小组"
  → agents-config.md 判断复杂度
    → T1-T2（≤2 子代理）→ 直接 sessions_spawn
    → T3-T5（3+ 子代理，需交接）→ team-brain.py

Computer Pitfalls

无输出路径：Agent 干活了你找不到结果 → 每次 spawn 指定精确路径
跳过 review：连续 3 次"小改动不审" = 质量崩塌
Agent 不汇报：静默 = 卡住，强制进度注释
能力不匹配：给无浏览器权限的 Agent 派浏览器任务
Orchestrator 自己干活：一旦动手干就失去了对整个团队的监督

参考文件

文件	读...
`references/team-setup.md`	定义角色/模型/workspace
`references/task-lifecycle.md`	任务状态/转换/注释
`references/communication.md`	通信/artifacts 路径
`references/patterns.md`	spec→build→test 等工作流
`references/tools-detail.md`	team-brain.py 等脚本详细参考
`references/enhancements.md`	v4.0 增强模块详情
`references/self-learning.md`	自学习 + 事实核查

Windows+WSL 环境说明

操作	状态
`sessions_spawn`	✅ 100% 稳定（默认方案）
`team-brain.py launch`	✅ 可用
`agent-team spawn subprocess`	❌ 0% 成功率（已废弃）

增强层索引

层级	文件	加载时机
L1	SKILL.md（核心流程）	首次加载
L2	`references/expert-knowledge-pool.md`	SKILL.md 后
L3	`references/knowledge/` (47 知识文件)	专家模式
L3	`scripts/*.py`（工具脚本）	按需调用
L4	`references/enhancements.md`（增强模块）	专家模式

故障与降级

场景	处理
子代理全部超时 → 0% 成功率	强制降级主会话执行
fact_check.py 不可达	回退到 exec 验证
team-brain.py 执行异常	回退到 sessions_spawn 直接派发
增强模块 feature flag false	跳过，不使用该模块

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

It worked
Partly
It didn't

What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free