Data as of
BrowserTools MCP for Claude Desktop: Setup & Config
Claude Desktop: BrowserTools MCP brings browser logs, screenshots, selected elements, and Lighthouse audits into compatible AI editors.
What you can do with BrowserTools in Claude Desktop
Browser evidence and Lighthouse audits in one MCP workflow, with an explicit inactive-project caveat.
- Investigate front-end failures using console and network evidence.
- Capture screenshots and inspect a selected DOM element.
Before you start: BrowserTools in Claude Desktop
- Chrome with the BrowserTools extension
- A compatible MCP client
- Node.js tooling
- A separate browser-tools-server process
- The BrowserToolsMCP DevTools panel open
Run npx @agentdeskai/browser-tools-server@latest separately. The MCP package belongs in the client configuration. The project is no longer active.
BrowserTools Claude Desktop configuration
Choose Claude Desktop configuration scope
For BrowserTools MCP, open macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json. Merge under mcpServers via Settings → Developer → Edit Config when available.
Add BrowserTools MCP to Claude Desktop
Use the browser-tools-mcp entry below in mcpServers. Keep BrowserTools MCP's documented stdio connection fields and your other Claude Desktop entries.
Check BrowserTools MCP in Claude Desktop
BrowserTools MCP: Fully quit and reopen Claude Desktop; check developer diagnostics.
Config file · macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"browser-tools-mcp": {
"command": "npx",
"args": [
"@agentdeskai/browser-tools-mcp@latest"
],
"env": {}
}
}
}BrowserTools tools available to Claude Desktop
BrowserTools's stored tool list documents 7 tools; compare it with Claude Desktop's connected tool list, because the stored list is not a live check.
- runAccessibilityAudit
When Claude Desktop requests runAccessibilityAudit from BrowserTools, review the action before approving it.
Check the page for accessibility problems, including WCAG-related issues.
- runPerformanceAudit
When Claude Desktop requests runPerformanceAudit from BrowserTools, review the action before approving it.
Inspect page loading and rendering performance.
- runSEOAudit
When Claude Desktop requests runSEOAudit from BrowserTools, review the action before approving it.
Review page-level search optimization factors.
- runBestPracticesAudit
When Claude Desktop requests runBestPracticesAudit from BrowserTools, review the action before approving it.
Check the page against web development practices.
- runAuditMode
When Claude Desktop requests runAuditMode from BrowserTools, review the action before approving it.
- runNextJSAudit
When Claude Desktop requests runNextJSAudit from BrowserTools, review the action before approving it.
- runDebuggerMode
When Claude Desktop requests runDebuggerMode from BrowserTools, review the action before approving it.
Troubleshooting BrowserTools in Claude Desktop
- Fully quit and reopen Claude Desktop for BrowserTools MCP. If npx @agentdeskai/browser-tools-mcp@latest (Node.js/npm) is missing, check the desktop process's PATH, which may differ from your terminal.
- BrowserTools MCP uses stdio here: Claude Desktop must keep the configured process running; protocol output belongs on stdout and diagnostics on stderr.
- Known limitation: The project explicitly states that it is no longer active.
- Known limitation: Requires an extension and two distinct server components.
- General Claude Desktop troubleshooting lives in the /clients/claude-desktop guide.
BrowserTools MCP — details. Claude Desktop guide