// SCAN REPORT

BrowserMCP

44 tools. 30 can modify or destroy data without limits.

30 write tools that can modify data. Rate limits recommended.

Last updated: 29 Jun 2026

30 can modify or destroy data

14 read-only

44 tools total

Community server · catalogue entry verified 29/06/2026

How to control BrowserMCP ↓

TOOL MAP

What BrowserMCP exposes to your agents

Read (14) Write / Execute (30) Destructive / Financial (0)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

High Risk

The most dangerous BrowserMCP tools

Execute · Medium browser_execute_js Execute JavaScript code in the context of the current page and return the result. Execute · Medium browser_navigate Navigate to a URL in the browser and return page content. Uses the user Execute · Medium browser_new_tab Open a new browser tab. Execute · Medium browser_new_window Open a new browser window.

30 of BrowserMCP's 44 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control BrowserMCP

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

Rate limit write operations

{
  "browser_back": {
    "limits": [
      {
        "counter": "browser_back_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "browser_extract_images": {
    "limits": [
      {
        "counter": "browser_extract_images_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

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

Create a free account and register BrowserMCP — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON BROWSERMCP →

Instant setup, no code required.

FULL CATALOGUE

All 44 BrowserMCP tools

EXECUTE 5 tools

Execute browser_execute_js Execute JavaScript code in the context of the current page and return the result. Execute browser_navigate Navigate to a URL in the browser and return page content. Uses the user Execute browser_new_tab Open a new browser tab. Execute browser_new_window Open a new browser window. Execute browser_wait_for Wait for an element matching a CSS selector to appear on the page.

WRITE 25 tools

Write browser_back Navigate back in browser history. Write browser_click Click on an element in the page identified by CSS selector. Write browser_close_tab Close a specific browser tab by its ID. Write browser_close_window Close a browser window by its ID. Write browser_console_log Capture console output (log/warn/error) on the page. Use Write browser_double_click Double-click on an element by CSS selector. Write browser_drag_drop Drag an element and drop it onto another element. Write browser_forward Navigate forward in browser history. Write browser_highlight Highlight elements on the page with a colored outline. Write browser_hover Hover over an element identified by CSS selector. Write browser_inject_css Inject custom CSS styles into the page. Write browser_keyboard Press a key or key combination on the page. Write browser_move_tab Move a tab to a different window. Write browser_mute_tab Mute or unmute a browser tab. Write browser_pin_tab Pin or unpin a browser tab. Write browser_readability Extract the main article content from the page (Reader Mode). Write browser_reload Reload a browser tab. Write browser_resize_window Resize a browser window. Write browser_right_click Right-click (context menu) on an element by CSS selector. Write browser_scroll Scroll the page in a direction or to a specific element. Write browser_select Select an option in a <select> dropdown element. Write browser_set_cookies Set a cookie for a specific URL. Write browser_set_storage Write a value to localStorage or sessionStorage. Write browser_switch_tab Activate/focus a specific browser tab. Write browser_type Type text into an input field identified by CSS selector.

READ 14 tools

Read browser_extract_images Extract all images from the page with their attributes. Read browser_extract_meta Extract all meta tags, Open Graph, Twitter Cards, JSON-LD, and canonical URL from the page. Read browser_extract_table Extract table data as structured JSON (headers + rows). Read browser_find_text Search for text on the page and return matches with surrounding context. Read browser_get_computed_style Get computed CSS styles for an element. Read browser_get_cookies Get cookies for a specific URL. Read browser_get_elements Get elements matching a CSS selector with their attributes. Read browser_get_links Extract all links from the page, optionally filtered. Read browser_get_storage Read localStorage or sessionStorage entries. Read browser_list_tabs List all open browser tabs with their IDs, URLs, and titles. Read browser_network_log Capture network requests (fetch/XHR) on the page. Use Read browser_read_page Read content from the current active tab or a specific tab. Read browser_screenshot Take a screenshot of the current browser tab. Returns a base64-encoded PNG image. Read browser_watch_changes Watch for DOM changes on the page using MutationObserver. Use

EXPLORE

Related servers

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

FAQ

Questions about BrowserMCP

How do I prevent bulk modifications through BrowserMCP? +

The BrowserMCP server has 25 write tools including browser_back, browser_click, browser_close_tab. 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 BrowserMCP.

How many tools does the Browser MCP server expose? +

44 tools across 3 categories: Execute, Read, Write. 14 are read-only. 30 can modify, create, or delete data.

How do I enforce a policy on BrowserMCP? +

Register the Browser 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 BrowserMCP tool call.

Deterministic rules across all 44 BrowserMCP tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON BROWSERMCP →

Instant setup, no code required.

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