
MCP server for AI clients to scrape sites with HTTP and stealth browser sessions
Runs locally- 17 tools
- Open source
- 104K GitHub stars
- Web search and scraping

MCP server for AI clients to scrape sites with HTTP and stealth browser sessions
Runs locallyBrowser MCP server that gives agents numbered text pages instead of screenshots
Runs locallyRun WCAG 2.2 accessibility audits and vision simulations on any URL from an agent
Runs locallyRun recorded app workflows from your agent and check each save before moving on
Runs locallyMCP server that gives a coding agent a Linux desktop it can see, click and type on
Runs locally
Records your voice and a browser session, then hands agents a causal evidence package
Runs locally
Persistent memory for Claude Code, Codex and OpenCode, stored in SQLite on your machine
Runs locallyBrowser actions for AI agents that are classified, approved and logged
Runs locallyShare VS Code's built-in browser with Claude Code, Copilot, Codex and other agents
Runs locally
Desktop GUI automation through accessibility trees, with replayable tests
Runs locallyA visible desktop browser, one workspace per project, that coding agents drive over MCP
Runs locallyLocal npm runtime for Kadmo agents: workflows and Chrome automation on your computer
Runs locallyLet an agent read Safari history, tabs and reading list, with writes switched off
Runs locally
MCP server that drives your real Chrome in per-agent background tabs on Windows
Runs locallyCapture browser flows into reviewable Java tests and diagnose failures from evidence
Runs locallyMCP server that lets an assistant operate any VNC-reachable desktop by screenshot
Runs locallyBrowser automation over raw CDP in one Python file, with an MCP server for AI agents
Runs locallyBrowser MCP that records website routes and replays them without model calls
Runs locallyRun 56 web performance snippets in headless Chromium and read Core Web Vitals results
Runs locallyFetch bot-protected and JavaScript-heavy pages as Markdown or template-parsed JSON
Not checked yet
Finds why Playwright tests fail, proposes fixes and verifies them before keeping the edit
Runs locallyLet an agent browse sites where you are already logged in, with encrypted local sessions
Not checked yetFree lookup an agent can query to see if a site permits AI agents to browse, shop or book
Not checked yetRun your agent CLI in your real, signed-in browser from a side panel or over MCP
Runs locallyGive an AI agent a real iPhone or Android phone, with one-tap approval before sends
Runs locallyMCP server that clones a web page from its URL into a verified offline HTML copy
Not checked yetBrowse login-protected and JavaScript-heavy pages through your own Chrome, Edge or Brave
Runs locallyLook up whether a website lets AI agents in, and the steps that worked, before visiting
Not checked yetAccessibility-tree desktop control for coding agents on macOS, Windows and Linux
Runs locally
Let an agent drive your logged-in Chrome, then replay the compiled task at zero tokens
Not checked yetSee every kind of AI for browser automation: tools, agents, MCP servers and skills.Ranked: the best AI for browser automation.