// SCAN REPORT

Browser

63 tools. 36 can modify or destroy data without limits.

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

Last updated: 30 Jun 2026

36 can modify or destroy data

27 read-only

63 tools total

Community server · catalogue entry verified 30/06/2026

How to control Browser ↓

TOOL MAP

What Browser exposes to your agents

Read (27) Write / Execute (34) Destructive / Financial (2)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Browser tools

Destructive · Medium browser_console_clear The tool irreversibly clears all captured console logs and stops the listening session. Destructive · High browser_storage_delete_cache This tool performs irreversible deletion of cached data from the browser's Cache Storage API. Execute · Medium browser_console_start Starting console log capture is not a passive read; it triggers an ongoing process within the browser that intercepts and records console output. Execute · High browser_evaluate This tool runs arbitrary JavaScript in the browser, which can trigger external operations, modify state, steal data, or interact with web services depending on the injected code.

36 of Browser's 63 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Browser

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

Deny destructive operations

{
  "browser_console_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_close_page": {
    "limits": [
      {
        "counter": "browser_close_page_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "browser_console_get": {
    "limits": [
      {
        "counter": "browser_console_get_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 Browser — 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 BROWSER →

Instant setup, no code required.

FULL CATALOGUE

All 63 Browser tools

DESTRUCTIVE 2 tools

Destructive browser_console_clear Clear all captured console logs and stop listening (see browser_docs) Destructive browser_storage_delete_cache Delete a specific cache from Cache Storage (see browser_docs)

EXECUTE 32 tools

WRITE 2 tools

Write browser_close_page Close a browser page (tab) (see browser_docs) Write browser_stop_video_recording Stop video recording and save the file (see browser_docs)

READ 27 tools

EXPLORE

Related servers

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

FAQ

Questions about Browser

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

Yes. The Browser server exposes 2 destructive tools including browser_console_clear, browser_storage_delete_cache. 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 Browser? +

The Browser server has 2 write tools including browser_close_page, browser_stop_video_recording. 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 Browser.

How many tools does the Browser MCP server expose? +

63 tools across 4 categories: Destructive, Execute, Read, Write. 27 are read-only. 36 can modify, create, or delete data.

How do I enforce a policy on Browser? +

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 Browser tool call.

Deterministic rules across all 63 Browser tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON BROWSER →

Instant setup, no code required.

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