-
Notifications
You must be signed in to change notification settings - Fork 967
Description
Being Blocked by bot detection.
Till date I was using this MCP server programatically with open ai agents SDK to create an AI agent.
I avoided bot detection while running in headless with the following config -
"mcpServers": {
"chrome-devtools": {
"command": "npx -y chrome-devtools-mcp@latest",
"env": {},
"args": [
"--headless=true",
"--isolated=true",
"--viewport=1280x900",
"--chromeArg=--disable-blink-features=AutomationControlled",
"--chromeArg=--user-agent=Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
]
}
But now those websites have updated their bot measures and even with "--headless=false" this config triggers a bot response.
Describe the solution you'd like
ZFC Digital's puppeteer-real-browser package is stealthy and avoids bot triggers.
So is there any way to get the mcp server to use/rely on that package to avoid bot detection? Or is there any other way to configure the chromedevtools mcp server to avoid bot measures?
Any advice is greatly appreciated.
Describe alternatives you've considered
There is another mcp literally called the puppeteer-real-browser-mcp-server which does not trigger bot measures but is not as reliable as the ChromeDevTools mcp.
Additional context
No response