12 web scraping tools as MCP server for AI agents (Claude Desktop, ChatGPT, Cursor). Reddit, Amazon, eBay, Google Maps, Yelp, YouTube, TikTok, Indeed, Trustpilot, contact finder, SaaS pricing.
接入配置
选择客户端，复制配置，再按项目文档补齐环境变量。
claude mcp add multi-scraper --env APIFY_TOKEN='<APIFY_TOKEN>' -- npx -y multi-scraper-mcp~/.codex/config.toml
[mcp_servers.multi-scraper]
command = "npx"
args = ["-y", "multi-scraper-mcp"]
env = { APIFY_TOKEN = "<APIFY_TOKEN>" }
# codex mcp add multi-scraper --env APIFY_TOKEN='<APIFY_TOKEN>' -- npx -y multi-scraper-mcp~/.cursor/mcp.json
{
"mcpServers": {
"multi-scraper": {
"command": "npx",
"args": [
"-y",
"multi-scraper-mcp"
],
"env": {
"APIFY_TOKEN": "<APIFY_TOKEN>"
}
}
}
}.vscode/mcp.json
{
"servers": {
"multi-scraper": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"multi-scraper-mcp"
],
"env": {
"APIFY_TOKEN": "<APIFY_TOKEN>"
}
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"multi-scraper": {
"command": "npx",
"args": [
"-y",
"multi-scraper-mcp"
],
"env": {
"APIFY_TOKEN": "<APIFY_TOKEN>"
}
}
}
}claude_desktop_config.json
{
"mcpServers": {
"multi-scraper": {
"command": "npx",
"args": [
"-y",
"multi-scraper-mcp"
],
"env": {
"APIFY_TOKEN": "<APIFY_TOKEN>"
}
}
}
}.trae/mcp.json
{
"mcpServers": {
"multi-scraper": {
"command": "npx",
"args": [
"-y",
"multi-scraper-mcp"
],
"env": {
"APIFY_TOKEN": "<APIFY_TOKEN>"
}
}
}
}需要的环境变量
APIFY_TOKEN
把占位值换成你自己的密钥；不要把密钥提交到代码仓库。