browserless-mcp

Browserless MCP Server

Browserless-mcp connects AI clients to Browserless for web search, page extraction, site crawling, and browser interaction. It serves research, site-audit, and automation workflows, with tools for checking account usage and diagnosing requests.

Server facts

browserless-mcp

Project

License
Not specified

Tech

Runtime / language
node
Transports
stdio
Remote available
No
Authentication
Not specified

Activity

GitHub stars
8
Last commit
Data as of

Tags

  • scraping
  • web automation

Overview

Web research and browser automation share a Browserless connection in browserless-mcp. Search can retrieve web, news, or image results and extract content from the pages it finds. For a known target, the scraper produces Markdown or HTML, while mapping discovers site URLs and crawling collects page content with depth and path controls. Lighthouse audits provide performance, accessibility, and SEO measurements. Page exports preserve native formats and can package associated assets for offline use; custom Puppeteer functions cover interactions or extraction that need JavaScript.

For interactive tasks, browserless_agent maintains a browser session and uses page snapshots to guide clicks, typing, scrolling, and other actions. Supporting skills address mechanics such as consent banners, shadow DOM, dynamic content, and file transfers, alongside discoverable recipes for particular sites. Saved authentication profiles let other tools reuse an existing signed-in state. Operational tools expose account details, consumption, request logs, and session information without modifying them, including access to recorded replays. Clients can connect to the hosted service using an API token, or run the server locally against a Browserless instance.

browserless-mcp setup notes

  • browserless-mcp records stdio only; confirm a network transport upstream before planning remote access.
  • browserless-mcp has no recorded license; confirm reuse terms in its repository.

Related MCP servers