Rove runs on your own machine — the client starts it, so there is no endpoint to ping.
Read public web pages and run goal-based browsing tasks with Rove's sandboxed browser agent.
Today is the operative word: we check Rove every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add rove -e 'ROVE_API_KEY=<ROVE_API_KEY>' -- npx -y @rovezi/mcp
{
"mcpServers": {
"rove": {
"command": "npx",
"args": [
"-y",
"@rovezi/mcp"
],
"env": {
"ROVE_API_KEY": "<ROVE_API_KEY>"
}
}
}
}
[mcp_servers.rove]
command = "npx"
args = ["-y", "@rovezi/mcp"]
env = { ROVE_API_KEY = "<ROVE_API_KEY>" }
{
"mcpServers": {
"rove": {
"command": "npx",
"args": [
"-y",
"@rovezi/mcp"
],
"env": {
"ROVE_API_KEY": "<ROVE_API_KEY>"
}
}
}
}
{
"servers": {
"rove": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@rovezi/mcp"
],
"env": {
"ROVE_API_KEY": "<ROVE_API_KEY>"
}
}
}
}
Put your own value in place of
<ROVE_API_KEY> (Your per-user Rove API key (rove_live_…)).
The author’s registry entry marks it as required.
Optional settings from the registry entry, left out of the block above:
ROVE_BASE_URL (Rove API base URL (default https://www.rovezi.com)).
Websites as APIs: read pages, run site tasks via learned APIs, cloud browser with saved logins.
Read public web content with Evercraft; search and browser automation stay behind safety gates.
Cloud-based web access with real browsers and JS rendering by ScrapingAnt
Drive the browsers of running ZennoPoster tasks: open pages, read and click elements, screenshots.
Hosted browser for agents: open pages, search the web, extract data, read PDFs and take screenshots.
Browser agents and stealth cloud browsers for AI: run web tasks, fill forms, extract data.
Cloud browsers for AI agents: get past bot detection, sign in with MFA, run web tasks at scale.
Save how to read a public web page once, then fetch fresh items as JSON without a browser.
Answers built from our own checks of this server.