M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
COMPARE UP TO 4 SERVERS

Compare MCP servers

Compare scores, permissions, risks, and fit in one decision-focused table.

DimensionOpenTabsLet AI call real web APIs through your browser session.Playwright MCPMicrosoft's official Playwright-based browser automation MCP server
FMRS49 / 100 · D78 / 100 · B
Reliability6 / 2013 / 20
Security and permissions10 / 2012 / 20
Maintenance9 / 2018 / 20
Documentation12 / 2018 / 20
Setup experience12 / 2017 / 20
Best for
  • Users who want AI to call web APIs through an already authenticated browser session.
  • Users who want local operation with multiple service plugins through an MCP client.
  • Developers building browser-level workflows or custom plugins.
  • Scenarios where an AI assistant needs to actually drive a browser to complete a task
  • Web automation testing and end-to-end verification
Not for
  • Users who do not use Chrome or do not want to grant AI access to browser sessions.
  • Use cases requiring a cloud-hosted service or a documented OAuth/API-key setup.
  • Integration evaluations that require an evidenced tool inventory and transport protocol.
  • Simple static-text scraping (a lightweight scraper is a better fit)
  • Environments sensitive to browser resource usage that can't run a full browser engine
Required permissions
  • Requires installing and running the Node.js CLI.
  • Requires a Chrome extension to access browser tabs and bridge to the local server.
  • AI uses the permissions of accounts currently authenticated in the browser.
  • Plugin tools are off by default and can be set to Off, Ask, or Auto.
  • The project includes anonymous telemetry with an opt-out option.
  • Requires installing and running a browser engine (Chromium/Firefox/WebKit) locally or on a server
  • browser_navigate can visit any URL; browser_evaluate can run arbitrary JavaScript in the page context
Risks and side effects
  • An authenticated browser session may allow AI to perform actions permitted to the corresponding service account.
  • Plugin source should be reviewed before enabling it; permissions reset when a plugin is updated.
  • Auto execution reduces opportunities for manual confirmation.
  • Browser actions, network capture, and third-party service calls may create unintended external effects.
  • browser_evaluate executing arbitrary scripts carries risk if a malicious page is visited — restrict to trusted sites or an isolated environment
  • Long-running browser instances consume significant memory and CPU
Supported clientsClaude Desktop, Claude Code, VS Code, Cursor, Windsurf, JetBrains IDEs, Zed, Cline, Amp, Kiro, LM Studio, Antigravity, Goose
Tools010