agent-web-combo
Agent Web Combo
一个面向 AI Agent 的联网工具组合:搜索发现 + 内容提取 + Kitesurf/Playwright 渲染 + agent-reach 平台读取。
搜索:Tavily / Exa / AnySearch
渲染:Cloudflare Kitesurf + Playwright(CDP)
平台内容:agent-reach(可选)
接入方式:CLI 和 MCP Server
功能
多搜索源切换:
tavily/exa/anysearch搜索结果的页面内容提取
Kitesurf 渲染页面、提取正文、截图
完整流水线:搜索 → 提取 → 渲染 → 截图
MCP Server:Claude Code / Cursor / OpenCode / DSH Desktop 等可直接调用
Related MCP server: myscrape
架构
搜索发现:Tavily / Exa / AnySearch
↓
内容提取:AnySearch extract / Tavily Extract / Exa Contents
↓
需要渲染/截图/JS:Kitesurf + Playwright (connectOverCDP)
↓
平台类内容/登录态:agent-reach (可选)项目结构
agent-web-combo/
├── combo/
│ ├── __main__.py # CLI 入口
│ ├── cli.py # 命令行
│ ├── config.py # 环境变量配置
│ ├── search.py # Tavily / Exa / AnySearch
│ ├── kitesurf.py # Kitesurf + Playwright
│ ├── agent_reach.py # agent-reach CLI
│ └── pipeline.py # 组合流水线
├── combo_mcp.py # MCP Server 入口
├── requirements.txt
├── pyproject.toml
└── .env.example安装
pip install -r requirements.txt
playwright install chromium # 仅本地调试/备选浏览器时需要;连 Kitesurf 不需要配置
cp .env.example .env按需填写:
# Cloudflare / Kitesurf(需要 Browser Run 权限)
CF_ACCOUNT_ID=你的账号ID
CF_API_TOKEN=你的API Token
# 搜索源(至少一个;AnySearch 可不填 Key)
TAVILY_API_KEY=tvly-xxx
EXA_API_KEY=xxx
ANYSEARCH_API_KEY=as_sk_xxx
.env已被.gitignore忽略,请勿提交。
CLI 使用
# 搜索
python -m combo search "Cloudflare Kitesurf" --provider anysearch --max-results 5
# 渲染单个页面 + 截图
python -m combo render "https://example.com" --screenshot output.png
# 完整流水线
python -m combo research "Kitesurf vs Tavily" \
--provider auto \
--max-results 5 \
--extract \
--render \
--render-limit 3 \
--screenshot-dir shots
# 检查 agent-reach
python -m combo agent-reachMCP 使用
启动 MCP Server:
python combo_mcp.py暴露工具:
工具 | 作用 |
| 搜索 |
| Kitesurf 渲染/截图 |
| 搜索 + 提取 + 渲染 |
| 检查 agent-reach 状态 |
各客户端配置见 README.mcp.md。
安全说明
所有 API Key 只放在本地
.env.gitignore已忽略.env、截图、缓存MCP 配置中不写明文密钥
公开发布前请确认没有提交
.env或截图文件
说明
AnySearch 走 MCP
tools/callHTTP 接口,匿名可用Kitesurf 通过 Cloudflare Browser Run 的 CDP endpoint 连接
Kitesurf 目前是 Beta,适合截图、HTML 提取、DOM 操作等 Agent 常见任务
This server cannot be deployed
Maintenance
Related MCP Connectors
Scrape, crawl and search the web for AI agents via MCP.
MCP server for Firecrawl — web search, scraping, and biomedical/arXiv paper search.
Docs: https://docs.keenable.ai/mcp-server Keenable is a free, remote MCP server that gives agents access to the web index. Search the web with ranked results and date/site filters, then fetch any indexed page as clean markdown. Works out of the box with no account or API key.
Free web search for AI agents. No API key required. Hosted MCP in active development.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceA minimal MCP server for agent-friendly web extraction and search. Offers two tools: fetching real pages with Playwright and Crawl4AI, and searching across 7 engines with automatic fallback.34AGPL 3.0
- AlicenseNot gradedqualityCmaintenanceA self-contained web-research MCP server that lets local LLM agents search, fetch, and synthesize web content using tools like web_search, web_fetch, and web_research.2MIT
- AlicenseNot gradedqualityDmaintenanceMCP server that wraps Playwright to give AI agents eyes on the web, enabling browser search, navigation, extraction, and interaction with intelligent LLM-based DOM extraction and skill caching.14 npmMIT
- FlicenseNot gradedqualityDmaintenanceMCP server that enables AI agents to search the web and extract clean Markdown content, with support for JavaScript rendering, structured data extraction, and screenshots.1-