~/Library/Application Support/Claude/claude_desktop_config.json · Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"klavis": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://api.klavis.ai/mcp/{server}?auth_token=YOUR_TOKEN"
]
}
}
}
打开 Claude Desktop → Settings → Developer → Edit Config。保存后重启应用。
~/.cursor/mcp.json · .cursor/mcp.json
{
"mcpServers": {
"klavis": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://api.klavis.ai/mcp/{server}?auth_token=YOUR_TOKEN"
]
}
}
}
Cursor 使用与 Claude Desktop 相同的 mcpServers 格式。项目级配置优先于全局。
VS Code → Cline → MCP Servers → Edit
{
"mcpServers": {
"klavis": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://api.klavis.ai/mcp/{server}?auth_token=YOUR_TOKEN"
]
}
}
}
点击 Cline 侧栏中的 MCP Servers 图标，然后选 "Edit Configuration"。
~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"klavis": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://api.klavis.ai/mcp/{server}?auth_token=YOUR_TOKEN"
]
}
}
}
格式与 Claude Desktop 相同。重启 Windsurf 生效。
~/.continue/config.json
{
"mcpServers": [
{
"name": "klavis",
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://api.klavis.ai/mcp/{server}?auth_token=YOUR_TOKEN"
]
}
]
}
Continue 使用服务器对象数组，而非映射。
~/.config/zed/settings.json
{
"context_servers": {
"klavis": {
"command": {
"path": "npx",
"args": [
"-y",
"mcp-remote",
"https://api.klavis.ai/mcp/{server}?auth_token=YOUR_TOKEN"
]
}
}
}
}
加入 context_servers。Zed 保存后热重载。
claude mcp add klavis -- npx -y mcp-remote https://api.klavis.ai/mcp/{server}?auth_token=YOUR_TOKEN
一行命令搞定。用 claude mcp list 验证，claude mcp remove 卸载。