Playwright MCP

64 tools. 48 can modify or destroy data without limits.

6 destructive tools with no built-in limits. Policy required.

Last updated:

48 can modify or destroy data
16 read-only
64 tools total

Community server · catalogue entry verified 03/07/2026

How to control Playwright MCP ↓

What Playwright MCP exposes to your agents

Read (16) Write / Execute (42) Destructive / Financial (6)
Critical Risk

The most dangerous Playwright MCP tools

48 of Playwright MCP's 64 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

How to control Playwright MCP

PolicyLayer is an MCP gateway — it sits between your AI agents and Playwright MCP, and nothing reaches the server without passing your rules. These are the rules we recommend:

Deny destructive operations
{
  "browser_cookie_clear": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Blocked by default. Requires approval."
      }
    ]
  }
}

Destructive tools should never be available to autonomous agents without human approval.

Rate limit write operations
{
  "browser_cookie_set": {
    "limits": [
      {
        "counter": "browser_cookie_set_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations
{
  "browser_console_messages": {
    "limits": [
      {
        "counter": "browser_console_messages_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

  1. Create a free account and register Playwright MCP — nothing to install.
  2. Add these rules — paste them, or build them visually. Tune the limits to your setup.
  3. Point your MCP client (Claude, Cursor, anything) at your gateway URL.
ENFORCE POLICY ON PLAYWRIGHT →

Instant setup, no code required.

All 64 Playwright MCP tools

EXECUTE 35 tools
Execute browser_click browser_click Execute browser_close browser_close Execute browser_drag browser_drag Execute browser_evaluate browser_evaluate Execute browser_file_upload browser_file_upload Execute browser_fill_form browser_fill_form Execute browser_generate_locator browser_generate_locator Execute browser_handle_dialog browser_handle_dialog Execute browser_hide_highlight browser_hide_highlight Execute browser_highlight browser_highlight Execute browser_hover browser_hover Execute browser_mouse_click_xy browser_mouse_click_xy Execute browser_mouse_down browser_mouse_down Execute browser_mouse_drag_xy browser_mouse_drag_xy Execute browser_mouse_move_xy browser_mouse_move_xy Execute browser_mouse_up browser_mouse_up Execute browser_mouse_wheel browser_mouse_wheel Execute browser_navigate browser_navigate Execute browser_navigate_back browser_navigate_back Execute browser_network_state_set browser_network_state_set Execute browser_pdf_save browser_pdf_save Execute browser_pick_locator browser_pick_locator Execute browser_press_key browser_press_key Execute browser_resize browser_resize Execute browser_resume browser_resume Execute browser_route browser_route Execute browser_run_code browser_run_code Execute browser_select_option browser_select_option Execute browser_start_tracing browser_start_tracing Execute browser_start_video browser_start_video Execute browser_stop_tracing browser_stop_tracing Execute browser_stop_video browser_stop_video Execute browser_type browser_type Execute browser_unroute browser_unroute Execute browser_wait_for browser_wait_for

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

Questions about Playwright MCP

Can an AI agent delete data through the Playwright MCP server? +

Yes. The Playwright MCP server exposes 6 destructive tools including browser_cookie_clear, browser_cookie_delete, browser_localstorage_clear. These permanently remove resources with no undo. PolicyLayer blocks destructive tools by default so they never reach the upstream server.

How do I prevent bulk modifications through Playwright MCP? +

The Playwright MCP server has 7 write tools including browser_cookie_set, browser_localstorage_set, browser_sessionstorage_set. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach Playwright MCP.

How many tools does the Playwright MCP server expose? +

64 tools across 4 categories: Destructive, Execute, Read, Write. 16 are read-only. 48 can modify, create, or delete data.

How do I enforce a policy on Playwright MCP? +

Register the Playwright MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every Playwright MCP tool call.

Deterministic rules across all 64 Playwright MCP tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

Instant setup, no code required.

64 Playwright MCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.

// GET IN TOUCH

Have a question or want to learn more? Send us a message.

Message sent.

We'll get back to you soon.