Lets AI assistants run Google Lighthouse on any URL to measure performance metrics, with device emulation and throttling.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add lighthouse -- npx -y lighthouse-mcp~/.codex/config.toml
[mcp_servers.lighthouse]
command = "npx"
args = ["-y", "lighthouse-mcp"]
# codex mcp add lighthouse -- npx -y lighthouse-mcp~/.cursor/mcp.json
{
"mcpServers": {
"lighthouse": {
"command": "npx",
"args": [
"-y",
"lighthouse-mcp"
]
}
}
}.vscode/mcp.json
{
"servers": {
"lighthouse": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"lighthouse-mcp"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"lighthouse": {
"command": "npx",
"args": [
"-y",
"lighthouse-mcp"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"lighthouse": {
"command": "npx",
"args": [
"-y",
"lighthouse-mcp"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"lighthouse": {
"command": "npx",
"args": [
"-y",
"lighthouse-mcp"
]
}
}
}