← Back to Skills Marketplace
164
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install agent-team-creator
Description
Agent团队创建和管理工具 - 用于创建多Agent协作团队，分配角色和任务，协调复杂工作流。支持创建CEO、分析师、研究员等角色，实现智能任务分发和结果整合。
README (SKILL.md)
Agent团队创建器
创建和管理多Agent协作团队，让复杂任务由多个专业Agent共同完成。
功能特点
- 👥 多角色团队 - 创建CEO、分析师、研究员等专业角色
- 🎯 任务分发 - 智能将任务分配给最适合的Agent
- 🔄 结果整合 - 自动汇总多个Agent的分析结果
- 📊 工作流管理 - 支持串行、并行、条件分支等复杂工作流
- 🎮 实时协调 - CEO Agent实时协调团队成员
使用场景
- 投资分析团队: CEO + 市场观察员 + 投资分析师 + 风控官
- 内容创作团队: 主编 + 研究员 + 撰稿人 + 编辑
- 代码审查团队: 技术负责人 + 安全专家 + 性能优化师
使用方法
创建团队
node scripts/create-team.mjs --name "投资分析团队" --roles "ceo,market-watcher,investment-analyst"
分配任务
node scripts/assign-task.mjs --team "投资分析团队" --task "分析英维克股票"
查看团队状态
node scripts/team-status.mjs --team "投资分析团队"
团队角色模板
| 角色 | 职责 | 适用场景 |
|---|---|---|
| CEO | 决策中枢、任务分配、结果整合 | 所有团队 |
| 市场观察员 | 宏观分析、板块追踪 | 投资分析 |
| 投资分析师 | 个股研究、估值分析 | 投资分析 |
| 风控官 | 风险评估、止损建议 | 投资分析 |
| 主编 | 内容策略、质量把控 | 内容创作 |
| 研究员 | 资料收集、数据分析 | 内容创作、投资分析 |
示例：创建投资分析团队
const team = {
name: "投资分析团队",
members: [
{ role: "ceo", agentId: "ceo", responsibility: "决策中枢" },
{ role: "market-watcher", agentId: "market-watcher", responsibility: "市场分析" },
{ role: "investment-analyst", agentId: "investment-analyst", responsibility: "个股研究" },
{ role: "risk-controller", agentId: "risk-controller", responsibility: "风险控制" }
],
workflow: "parallel-analysis-then-decision"
};
License
MIT
Usage Guidance
This skill is an instruction-only guide that expects Node scripts (node scripts/*.mjs) to exist, but the package you installed contains no code files. Before running any of the example commands, manually inspect the referenced repository or scripts: clone the GitHub homepage, open the scripts (create-team.mjs, assign-task.mjs, team-status.mjs), and review what they do (network calls, file I/O, environment access). Only run scripts you trust, preferably in a sandbox or isolated environment. If you need automated behavior, prefer a skill that bundles source or a signed release; if the repository is intended as the source, verify the repo URL, commit history, and author identity first.
Capability Analysis
Type: OpenClaw Skill
Name: agent-team-creator
Version: 1.0.0
The skill bundle contains metadata and documentation for a multi-agent orchestration tool. The SKILL.md file describes standard functionality for creating and managing agent teams and does not contain any malicious instructions, prompt injections, or suspicious execution patterns. All content is consistent with the stated purpose of the tool.
Capability Assessment
Purpose & Capability
The skill is a multi-agent team/orchestration tool and declares node as a required binary, which is coherent. However, the SKILL.md's runtime commands call scripts (node scripts/*.mjs) that are not present in the skill bundle. That mismatch (instructions expecting local scripts but no code provided) is an inconsistency: either the skill is only documentation pointing to an external repo or it's incomplete.
Instruction Scope
Instructions explicitly tell the agent/user to run node scripts (create-team.mjs, assign-task.mjs, team-status.mjs). Those commands will execute arbitrary JavaScript if the scripts exist. The SKILL.md does not instruct reading unrelated system files or env vars, but because the actual scripts are absent, there's no way to verify what those scripts would do — they could perform network access, read credentials, or modify files.
Install Mechanism
There is no install spec and no files to write to disk from the skill itself. Instruction-only skills are low-risk in terms of automated installs. The homepage points to a GitHub repo (which is a normal location to host code), but the repo contents are not bundled here, so manual review would be required before running anything from that repo.
Credentials
The skill declares no required environment variables, no credentials, and no config paths. That is proportionate to the SKILL.md presented. The main risk is that the referenced Node scripts (not included) might request credentials or read environment variables when executed — this cannot be determined from the provided materials.
Persistence & Privilege
always is false and the skill does not request persistent presence or elevated privileges. There are no instructions to modify other skills or system-wide settings in the SKILL.md.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install agent-team-creator - After installation, invoke the skill by name or use
/agent-team-creator - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of agent-team-creator
- Create and manage multi-agent collaborative teams for complex workflows.
- Supports role assignment (e.g., CEO, analyst, researcher) and intelligent task distribution.
- Automates integration of results from different agents.
- Enables workflow management (serial, parallel, conditional branching).
- Designed for various scenarios, including investment analysis, content creation, and code review teams.
Metadata
Frequently Asked Questions
What is Claw Team Builder?
Agent团队创建和管理工具 - 用于创建多Agent协作团队，分配角色和任务，协调复杂工作流。支持创建CEO、分析师、研究员等角色，实现智能任务分发和结果整合。 It is an AI Agent Skill for Claude Code / OpenClaw, with 164 downloads so far.
How do I install Claw Team Builder?
Run "/install agent-team-creator" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Claw Team Builder free?
Yes, Claw Team Builder is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Claw Team Builder support?
Claw Team Builder is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Claw Team Builder?
It is built and maintained by ryan-wuxl (@ryan-wuxl); the current version is v1.0.0.
More Skills