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.

DimensionPlaywright MCPGive AI agents full browser control via structured accessibility snapshotsFSB MCP ServerControl the FSB browser extension from any MCP client for DOM-first AI browser automation.
FMRS43 / 100 · D66 / 100 · C
Reliability3 / 2010 / 20
Security and permissions8 / 2014 / 20
Maintenance17 / 2015 / 20
Documentation9 / 2015 / 20
Setup experience6 / 2012 / 20
Best for
  • Users who want an LLM to drive open-ended browser tasks
  • Developers already using MCP-capable clients like Claude Desktop, VS Code, Cursor, or Windsurf
  • Cross-browser (Chromium/Firefox/WebKit) automation without hand-writing test scripts
  • Developers who want AI clients to directly control the browser
  • Scenarios requiring DOM-level understanding and low-latency automation (50-200 ms)
  • Multi-client support (Claude Code, Cursor, VS Code, etc.)
  • Internal tools or automation scripts needing browser interaction
Not for
  • Teams needing a full end-to-end test runner with parallel execution and trace viewer — the README points to Playwright Test for that instead
  • Coding-agent setups that prioritize token efficiency and want to avoid loading large tool schemas and accessibility trees — the README recommends the lighter-weight Playwright CLI for that case
  • Environments requiring strict isolation from real browser sessions, cookies, or local files, since the agent gets genuine browser control
  • Headless browser or server-side monitoring
  • Bypassing website restrictions or unsupervised automation
  • Automation on sensitive pages without human supervision
  • Complex scenarios that require visual understanding (e.g., screenshots)
Required permissions
  • Local Node.js/npx runtime to launch the server process
  • Ability to download and run Chromium/Firefox/WebKit browser binaries
  • Outbound network access to load whatever pages the agent is directed to visit
  • Requires installing the Chrome extension, which requests permissions for active tab, debugger, storage, etc.
  • MCP server needs to connect to the extension via WebSocket (default ws://localhost:7225)
  • Optional Streamable HTTP mode requires allowing connections to http://127.0.0.1:7226/mcp
  • Uploading files requires providing absolute local paths
Risks and side effects
  • The server grants agents real control of a browser, so a manipulated prompt could cause navigation to or submission of data on unintended sites
  • If a saved storage state (login session) is loaded, the agent may gain access to authenticated account data
  • The README does not describe any authentication or access-control mechanism built into the MCP server itself — safety depends on the host client and how target URLs are constrained
  • Browser automation can be unpredictable on complex or sensitive sites; supervision is required
  • Credentials and payment info are stored in a vault but require user unlock flows
  • API keys are encrypted in Chrome storage, but clients may send page context to model providers
  • MCP server is a local process, posing local security risks
  • Capability catalog allows most non-denied origins but is gated by multiple checks
Supported clientsClaude Desktop, Claude Code, VS Code, Cursor, WindsurfClaude Desktop, Claude Code, Cursor, VS Code, Windsurf, Codex
Tools030