feat: add mcp-servers/**

This commit is contained in:
Sun-ZhenXing
2025-10-23 09:08:07 +08:00
parent ece59b42bf
commit f603ed5db9
57 changed files with 3061 additions and 95 deletions

View File

@@ -0,0 +1,8 @@
# Docker image version
PLAYWRIGHT_VERSION=latest
# Host port override
PLAYWRIGHT_PORT_OVERRIDE=8000
# Timezone
TZ=UTC

View File

@@ -0,0 +1,67 @@
# Playwright MCP Server
Playwright MCP Server provides browser automation and web scraping capabilities through the Model Context Protocol.
## Features
- 🌐 **Browser Automation** - Automate browser operations
- 📸 **Screenshot Capture** - Capture web page screenshots
- 🔍 **Web Scraping** - Intelligent web content extraction
- 📝 **Form Filling** - Automated form filling
- 🎭 **Multi-Browser** - Support for Chromium, Firefox, WebKit
- 🔐 **Cookie & Session** - Cookie and session management
## Environment Variables
| Variable | Default | Description |
| -------------------------- | -------- | -------------------- |
| `PLAYWRIGHT_VERSION` | `latest` | Docker image version |
| `PLAYWRIGHT_PORT_OVERRIDE` | `8000` | Service port |
| `TZ` | `UTC` | Timezone |
## Quick Start
### 1. Configure Environment
Create a `.env` file:
```env
PLAYWRIGHT_VERSION=latest
PLAYWRIGHT_PORT_OVERRIDE=8000
TZ=Asia/Shanghai
```
### 2. Start Service
```bash
docker compose up -d
```
### 3. Verify Service
```bash
curl http://localhost:8000/health
```
## Resource Requirements
- Minimum memory: 512MB
- Recommended memory: 2GB
- Shared memory: 2GB (configured)
## Common Use Cases
1. **Web Screenshots** - Automatically visit and capture screenshots
2. **Data Scraping** - Extract data from dynamic web pages
3. **UI Testing** - Automated UI testing scenarios
4. **Form Automation** - Batch fill and submit forms
## References
- [Playwright Official Site](https://playwright.dev/)
- [MCP Documentation](https://modelcontextprotocol.io/)
- [Docker Hub - mcp/playwright](https://hub.docker.com/r/mcp/playwright)
## License
MIT License

View File

@@ -0,0 +1,67 @@
# Playwright MCP Server
Playwright MCP Server 是一个基于 Playwright 的模型上下文协议MCP服务器提供浏览器自动化和网页抓取功能。
## 功能特性
- 🌐 **浏览器自动化** - 自动化浏览器操作
- 📸 **截图捕获** - 捕获网页截图
- 🔍 **网页抓取** - 智能提取网页内容
- 📝 **表单填写** - 自动化表单填写
- 🎭 **多浏览器支持** - 支持 Chromium、Firefox、WebKit
- 🔐 **Cookie 和会话管理** - Cookie 和会话管理
## 环境变量
| 变量 | 默认值 | 说明 |
| -------------------------- | -------- | --------------- |
| `PLAYWRIGHT_VERSION` | `latest` | Docker 镜像版本 |
| `PLAYWRIGHT_PORT_OVERRIDE` | `8000` | 服务端口 |
| `TZ` | `UTC` | 时区 |
## 快速开始
### 1. 配置环境
创建 `.env` 文件:
```env
PLAYWRIGHT_VERSION=latest
PLAYWRIGHT_PORT_OVERRIDE=8000
TZ=Asia/Shanghai
```
### 2. 启动服务
```bash
docker compose up -d
```
### 3. 验证服务
```bash
curl http://localhost:8000/health
```
## 资源需求
- 最小内存512MB
- 推荐内存2GB
- 共享内存2GB已配置
## 常见使用场景
1. **网页截图** - 自动访问并捕获截图
2. **数据抓取** - 从动态网页提取数据
3. **UI 测试** - 自动化 UI 测试场景
4. **表单自动化** - 批量填写和提交表单
## 参考链接
- [Playwright 官方网站](https://playwright.dev/)
- [MCP 文档](https://modelcontextprotocol.io/)
- [Docker Hub - mcp/playwright](https://hub.docker.com/r/mcp/playwright)
## 许可证
MIT License

View File

@@ -0,0 +1,42 @@
x-default: &default
restart: unless-stopped
logging:
driver: json-file
options:
max-size: 100m
max-file: "3"
services:
playwright:
<<: *default
image: mcp/playwright:${PLAYWRIGHT_VERSION:-latest}
environment:
- MCP_HOST=0.0.0.0
- TZ=${TZ:-UTC}
ports:
- "${PLAYWRIGHT_PORT_OVERRIDE:-8000}:8000"
# 如果需要访问本地文件,可以挂载卷
volumes:
- playwright_data:/app/data
healthcheck:
test: ["CMD", "wget", "--spider", "-q", "http://localhost:8000/health"]
interval: 30s
timeout: 10s
retries: 3
start_period: 20s
deploy:
resources:
limits:
cpus: '2.00'
memory: 2G
reservations:
cpus: '0.5'
memory: 512M
# Playwright 需要额外的权限来运行浏览器
security_opt:
- seccomp:unconfined
shm_size: '2gb'
volumes:
playwright_data:
driver: local