← Back to Skills Marketplace
467
Downloads
0
Stars
5
Active Installs
1
Versions
Install in OpenClaw
/install ai-employee-team
Description
AI员工协作技能包 - 多AI角色配置、任务自动分配、进度监控、结果汇报。适合想要自动化运营的团队。
README (SKILL.md)
AI员工协作技能包
一句话介绍
把多个AI角色组织成虚拟团队,自动分配任务、监控进度、汇报结果。
解决什么问题?
- 一个人管不过来:多个项目并行 → AI团队自动分工
- 任务分配不均:不知道谁该做什么 → 智能任务分配
- 进度追踪困难:到处问进度 → 自动状态汇报
- 协作效率低:沟通成本高 → 结构化工作流
核心功能
- 🤖 多AI角色配置:老板、产品、技术、测试、运营...
- 📋 任务智能分配:根据角色自动派发任务
- 📊 状态实时监控:查看每个员工的工作状态
- 📝 结果自动汇报:日报、周报、任务完成通知
- 🔄 工作流编排:定义任务流转规则
- 💾 记忆持久化:AI员工有长期记忆
快速开始
安装
cd ai-employee-skill
npm install
配置员工
编辑 config/employees.json:
{
"employees": [
{
"id": "boss",
"name": "老板AI",
"role": "战略决策",
"skills": ["战略分析", "资源分配", "绩效评估"]
},
{
"id": "pm",
"name": "产品AI",
"role": "产品规划",
"skills": ["需求分析", "产品设计", "文档撰写"]
}
]
}
启动公司
npm start
命令列表
| 命令 | 说明 |
|---|---|
/hire \x3C角色> |
招聘新AI员工 |
/fire \x3Cid> |
解雇AI员工 |
/assign \x3C任务> |
分配任务 |
/status |
查看团队状态 |
/report [日报/周报] |
生成工作报告 |
/meeting |
召开团队会议 |
文件结构
ai-employee-skill/
├── SKILL.md # 技能说明
├── README.md # 产品文档
├── TUTORIAL.md # 使用教程
├── package.json # 依赖配置
├── config.json # 系统配置
├── scripts/ # 核心脚本
│ ├── employee-manager.js # 员工管理
│ ├── task-dispatcher.js # 任务分发
│ ├── status-monitor.js # 状态监控
│ ├── report-generator.js # 报告生成
│ └── meeting-coordinator.js # 会议协调
├── templates/ # 员工模板
│ ├── boss-ai.md
│ ├── product-ai.md
│ ├── tech-ai.md
│ ├── test-ai.md
│ └── operation-ai.md
└── examples/ # 示例场景
├── startup-team.json
└── content-team.json
员工角色说明
| 角色 | 职责 | 核心能力 |
|---|---|---|
| 老板AI | 战略决策 | 市场分析、资源分配、绩效评估 |
| 产品AI | 产品规划 | 需求分析、产品设计、PRD撰写 |
| 技术AI | 技术开发 | 架构设计、编码实现、代码审查 |
| 测试AI | 质量保障 | 测试计划、自动化测试、Bug追踪 |
| 运营AI | 增长运营 | 内容创作、用户运营、数据分析 |
| 财务AI | 财务管理 | 成本核算、ROI分析、预算规划 |
| HR-AI | 人力资源 | 招聘管理、绩效评估、培训发展 |
工作流程
┌─────────────────────────────────────────────────────────────┐
│ 任务流程 │
├─────────────────────────────────────────────────────────────┤
│ │
│ ┌────────┐ ┌────────┐ ┌────────┐ ┌────────┐ │
│ │ 老板AI │ → │ 产品AI │ → │ 技术AI │ → │ 测试AI │ │
│ │ 下达 │ │ 需求 │ │ 开发 │ │ 验收 │ │
│ └────────┘ └────────┘ └────────┘ └────────┘ │
│ ↓ ↓ ↓ │
│ ┌─────────────────────────────────────────────────┐ │
│ │ 运营AI / 财务AI │ │
│ │ 发布 + 数据 + 汇报 │ │
│ └─────────────────────────────────────────────────┘ │
│ │
└─────────────────────────────────────────────────────────────┘
适用人群
- 独立开发者:一人即团队
- 小型创业团队:人手不够AI来凑
- 自媒体运营:内容批量生产
- 项目管理者:多项目并行管理
- 自动化爱好者:AI协作实践
定价
- 基础版:¥199(5个员工角色 + 基础工作流)
- 专业版:¥399(无限角色 + 自定义工作流 + API接口)
- 企业版:¥999(私有部署 + 定制开发 + 技术支持)
开发者:AI-Company 联系:通过ClawHub
Usage Guidance
This package appears inconsistent: the docs promise a full orchestration system but the archive only contains a single employee-manager script and docs. Before installing or running anything: 1) ask the publisher for the full source or an upstream repository (package.json and the other scripts) and verify integrity; 2) inspect any package.json and all scripts that would be installed by 'npm install' (look for network calls, exec/spawn, remote downloads); 3) run in an isolated/sandbox environment (or container) if you must test; 4) be cautious about supplying LLM API keys — the skill references models but does not declare how keys are used or stored; 5) treat local memory files under memory/ as potentially sensitive and review read/write behavior. The current artifact is more likely incomplete or poorly packaged than overtly malicious, but the mismatches warrant caution.
Capability Analysis
Type: OpenClaw Skill
Name: ai-employee-team
Version: 1.0.1
The `scripts/employee-manager.js` file contains a path traversal vulnerability in its `hire` method. Specifically, the `template` parameter, if controlled by a malicious actor (e.g., via prompt injection to the AI agent), can be used to read arbitrary files from the file system (e.g., `../../../../etc/passwd`) and write their content into the skill's local `memory` directory. Additionally, the `writeMemory` function allows appending arbitrary content to local memory files, which could be used to store malicious data within the skill's scope. While there is no evidence of intentional data exfiltration to external endpoints or other direct malicious activities, these vulnerabilities pose a significant information disclosure risk and local data manipulation capability.
Capability Assessment
Purpose & Capability
SKILL.md/README describe a multi-component AI team framework (task dispatcher, status monitor, report generator, meeting coordinator, templates, package.json, npm start, etc.), but the bundle only contains one script (scripts/employee-manager.js) plus docs. The claimed capabilities exceed what's actually provided.
Instruction Scope
Runtime instructions tell the user to run 'npm install' and 'npm start', edit config files, and use many CLI commands and endpoints that are not present in the supplied files. The SKILL.md expects a working project layout (package.json, many scripts) that is missing — following the docs could fail or run unknown code if additional files are fetched later.
Install Mechanism
There is no formal install spec (lowest-risk). However SKILL.md instructs running 'npm install' and 'npm start' even though package.json is not included in the bundle. That mismatch is suspicious: either required files were omitted from the package or the docs are outdated.
Credentials
The skill requests no environment variables or credentials (proportionate). SKILL.md references using LLM models (e.g., 'gpt-4') and private deployment/pricing but does not declare any API keys or integration details — missing required credentials for real LLM use.
Persistence & Privilege
The skill does local file I/O (creates config/ and memory/ directories and writes memory files). 'always' is false and there are no instructions to modify other skills or system-wide settings. Local persistence is expected for the described memory feature.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install ai-employee-team - After installation, invoke the skill by name or use
/ai-employee-team - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
更新文档和销售材料
Metadata
Frequently Asked Questions
What is AI Employee Team?
AI员工协作技能包 - 多AI角色配置、任务自动分配、进度监控、结果汇报。适合想要自动化运营的团队。 It is an AI Agent Skill for Claude Code / OpenClaw, with 467 downloads so far.
How do I install AI Employee Team?
Run "/install ai-employee-team" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is AI Employee Team free?
Yes, AI Employee Team is completely free (open-source). You can download, install and use it at no cost.
Which platforms does AI Employee Team support?
AI Employee Team is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created AI Employee Team?
It is built and maintained by gdp6539 (@gdp6539); the current version is v1.0.1.
More Skills