Crawlio Browser MCP server
{
"mcpServers": {
"crawlio-browser": {
"command": "npx",
"args": [
"-y",
"crawlio-browser@latest"
]
}
}
}{
"mcpServers": {
"crawlio-browser": {
"command": "npx",
"args": [
"-y",
"crawlio-browser@latest"
]
}
}
}{
"mcpServers": {
"crawlio-browser": {
"command": "npx",
"args": [
"-y",
"crawlio-browser@latest"
]
}
}
}{
"servers": {
"crawlio-browser": {
"command": "npx",
"args": [
"-y",
"crawlio-browser@latest"
]
}
}
}{
"mcpServers": {
"crawlio-browser": {
"command": "npx",
"args": [
"-y",
"crawlio-browser@latest"
]
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsoninstall npx -y crawlio-browserversion 1.11.0
What it does
MCP server that connects an AI agent to a live Chrome browser via Chrome DevTools Protocol, exposing 150 tools for browser automation, page analysis, and data extraction. Designed for developers who need agents to interact with single-page applications, authenticated sessions, and dynamically rendered content.
- Automate interactions with web applications that require JavaScript rendering and real user sessions, such as logging in and navigating authenticated pages.
- Extract structured data from dynamic web content, including tracking pixel analysis and technographic fingerprinting.
- Audit website SEO properties and capture evidence from live pages that static crawlers cannot access.
Tools this server exposes
| Tool | Does |
|---|---|
| page_navigation_and_interaction | Navigate to URLs, click elements, submit forms, and interact with page content. |
| content_extraction_and_capture | Extract text, HTML, structured data, and capture screenshots from the current page. |
| DOM_querying_and_inspection | Query and inspect DOM elements, retrieve element properties and states. |
| browser_state_management | Read and manage cookies, localStorage, sessionStorage, and other browser state. |
| network_and_request_analysis | Analyze network traffic, XHR requests, and HTTP headers from the page. |
| SEO_auditing | Audit SEO properties including metadata, structured data, and page accessibility. |
| technographic_analysis | Identify technologies, frameworks, and libraries used on the page. |
| tracking_and_pixel_analysis | Detect and analyze tracking pixels and third-party measurement scripts. |
| evidence_tracking | Capture and track evidence from findings with confidence scores. |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.