Token导航 LogoToken导航

Agent Skill · 研究检索

teamclaw-orchestrator(团队之爪协调器)

teamclaw-orchestrator 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

teamclaw-orchestrator 是研究检索类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为操作浏览器;许可证为 MIT-0,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 26,800。

总安装 3,501周安装 143GitHub Stars 公开资料未说明下载 1,121核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
1.0.0

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:teamclaw-orchestrator(团队之爪协调器)
来源仓库:https://github.com/topcheer/teamclaw-orchestrator
安装命令:
openclaw skills install teamclaw-orchestrator
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install teamclaw-orchestrator

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

动态组建虚拟软件团队完成功能开发或缺陷修复。根据任务类型自动匹配开发者、测试员等角色。实时同步进度并支持人工干预调整策略。teamclaw-orchestrator 属于研究检索类 Skill,可作为该场景下的辅助能力补充。适用于快速原型开发与紧急 bug 响应场景。建议限制单次最大并发代理数量以防过载。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install teamclaw-orchestrator;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要操作浏览器,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

teamclaw-orchestrator 怎么安装?

在终端执行 openclaw skills install teamclaw-orchestrator。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

teamclaw-orchestrator 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

teamclaw-orchestrator 需要什么权限?

权限级别是操作浏览器。该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

teamclaw-orchestrator 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

teamclaw-orchestrator 用的人多吗?

公开数据是累计安装 3,501 次,近一周安装 143 次,下载 1,121 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

teamclaw-orchestrator 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
teamclaw-orchestrator

团队之爪协调器 · 本页

3,501—操作浏览器MIT-0通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

name
teamclaw
description
>
Triggers
build me ...", "create a ...", "team status", "assign to team", "teamclaw",
version
1.0.0
metadata
openclaw
author
TeamClaws
homepage
https://github.com/topcheer/teamclaw
links
homepage
https://github.com/topcheer/teamclaw
repository
https://github.com/topcheer/teamclaw
documentation
https://github.com/topcheer/teamclaw/blob/main/README.md
changelog
https://github.com/topcheer/teamclaw/releases

TeamClaw — AI Team Orchestration

Orchestrate a virtual software team through natural conversation. Submit requirements, monitor progress, review deliverables, and answer clarifications — all without leaving your chat.

When to Use This Skill

Activate when the user:

  • Wants to build, create, or implement something (e.g., "build me a todo app", "create an auth system")
  • Asks to delegate work to a team (e.g., "have the team work on this", "assign this to the developer")
  • Wants to check team status (e.g., "what are the workers doing?", "show me task progress")
  • Needs to review results (e.g., "show me what the developer built", "check the QA report")
  • Has pending clarifications to answer (e.g., "are there questions from the team?")
  • Mentions teamclaw by name

Prerequisites

TeamClaw must be running in controller mode. Detect the controller URL:

# Default controller URL
TEAMCLAW_URL="http://127.0.0.1:9527"

# Health check — verify the controller is running
curl -sf "$TEAMCLAW_URL/api/v1/health"

If the health check fails, tell the user to start TeamClaw first (install the teamclaw-setup skill for guidance).

Core Workflow

1. Submit a Requirement

When the user describes what they want built, submit it to the controller intake:

curl -s -X POST "$TEAMCLAW_URL/api/v1/controller/intake" \
  -H "Content-Type: application/json" \
  -d '{"message": "<user requirement here>"}'

The controller will:

  1. Analyze the requirement
  2. Decompose it into tasks for appropriate roles (developer, architect, QA, etc.)
  3. Assign tasks to available workers
  4. Return a controller run with tracking info

Response fields:

  • controllerRun.id — run identifier for tracking
  • controllerRun.status — running, completed, failed
  • result — the controller agent's orchestration output

Important: The intake call may take 30-180 seconds as the controller agent plans and decomposes the work. Use --max-time 300 with curl for large requirements.

2. Monitor Progress

Check team overview
curl -s "$TEAMCLAW_URL/api/v1/team/status"

Key response fields:

  • workers — map of all workers with their status (idle, busy, offline)
  • tasks — map of all tasks with status (pending, assigned, in-progress, completed, failed)
  • pendingClarificationCount — number of unanswered questions from workers
  • controllerRuns — orchestration run history
List tasks
curl -s "$TEAMCLAW_URL/api/v1/tasks"
Get task details with execution history
curl -s "$TEAMCLAW_URL/api/v1/tasks/<taskId>/execution"

Returns the task with full execution log (lifecycle events, progress updates, errors).

List workers
curl -s "$TEAMCLAW_URL/api/v1/workers"

3. Review Results

When a task completes, its result contract contains structured deliverables:

# Get the task to see its resultContract
curl -s "$TEAMCLAW_URL/api/v1/tasks/<taskId>"

The resultContract includes:

  • summary — what was accomplished
  • deliverables[] — list of files, notes, or artifacts produced
  • discoveredPatterns[] — patterns noticed during execution
  • suggestedNextSteps[] — recommended follow-up actions

4. Handle Clarifications

Workers may need clarification to proceed. Check and answer them:

# List pending clarifications
curl -s "$TEAMCLAW_URL/api/v1/clarifications"

# Answer a clarification
curl -s -X POST "$TEAMCLAW_URL/api/v1/clarifications/<clarificationId>/answer" \
  -H "Content-Type: application/json" \
  -d '{"answer": "<your answer>", "answeredBy": "user"}'

5. Create Individual Tasks

For fine-grained control, create tasks directly:

curl -s -X POST "$TEAMCLAW_URL/api/v1/tasks" \
  -H "Content-Type: application/json" \
  -d '{
    "title": "Implement user login",
    "description": "Create a login form with email/password auth",
    "priority": "high",
    "assignedRole": "developer"
  }'

Valid roles: pm, architect, developer, qa, release-engineer, infra-engineer, devops, security-engineer, designer, marketing.

Valid priorities: low, medium, high.

6. Send Messages to the Team

# Direct message to a role
curl -s -X POST "$TEAMCLAW_URL/api/v1/messages/direct" \
  -H "Content-Type: application/json" \
  -d '{
    "from": "user",
    "toRole": "developer",
    "content": "Please also add input validation"
  }'

# Broadcast to all workers
curl -s -X POST "$TEAMCLAW_URL/api/v1/messages/broadcast" \
  -H "Content-Type: application/json" \
  -d '{
    "from": "user",
    "content": "Deadline moved up — prioritize core features"
  }'

Response Presentation

When presenting results to the user:

  1. After intake submission: Summarize what the controller planned — how many tasks, which roles assigned, estimated workflow.
  1. For status checks: Show a concise table of workers and tasks:
   Workers: 3 active (developer: busy, qa: idle, architect: idle)
   Tasks: 5 total (2 completed, 1 in-progress, 2 pending)
   Clarifications: 1 pending
  1. For completed tasks: Highlight the deliverables, key files changed, and suggested next steps.
  1. For clarifications: Present the question clearly and ask the user to provide an answer.

Web UI

TeamClaw also provides a web dashboard for visual monitoring:

Open in browser: $TEAMCLAW_URL/ui

Mention this to users who prefer a visual overview.

Troubleshooting

IssueSolution
Health check failsEnsure TeamClaw plugin is enabled in controller mode: openclaw plugins list
No workers availableCheck processModel config; single-process creates workers automatically
Intake times outIncrease taskTimeoutMs in TeamClaw config; ensure AI model is responsive
Task stuck in pendingNo idle worker for the assigned role; check GET /api/v1/workers
Clarification blockingAnswer pending clarifications via POST /api/v1/clarifications/:id/answer

Read references for detailed API docs

Read references/api-quick-ref.md for the complete endpoint reference when you need exact request/response formats.

平台分布

OpenClaw

93.09%
按下载量换算1,044

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills