# browser_evaluate

A execute tool on the Playwright MCP server.

Agent View of the PolicyLayer registry record for `browser_evaluate`. HTML page: https://policylayer.com/tools/hanato238-playwright-mcp/browser-evaluate

## Facts

- Tool: `browser_evaluate`
- Server: Playwright (`hanato238/playwright-mcp`) — https://policylayer.com/tools/hanato238-playwright-mcp.md
- Homepage: https://github.com/Hanato238/playwright-mcp
- Risk category: Execute (High risk)
- Registry record: grade F, identity unverified
- Server rate-limited: no
- Parameters: 0
- Recommended policy verdict: Rate-limited

## Example call (MCP tools/call, JSON-RPC 2.0)

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "browser_evaluate",
    "arguments": {}
  }
}
```

## Why browser_evaluate is rated High

The name 'browser_evaluate' is a well-established pattern (e.g., Playwright's page.evaluate()) for executing arbitrary JavaScript in a browser page context. This allows running arbitrary code in the browser, which can access DOM, cookies, storage, make network requests, and interact with any loaded page content.

From the tool's own definition: "Tool name 'browser_evaluate' on a browser automation MCP server; description is empty but the name strongly implies JavaScript evaluation/execution in the browser context."

## Use case

AI agents invoke browser_evaluate to trigger actions in Playwright. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call: builds kicked off, notifications sent, workflows started.

## Recommended policy (PolicyLayer)

Verdict: **Rate-limited**. Enforced by the PolicyLayer MCP gateway (https://policylayer.com/mcp-gateway) before a call reaches Playwright:

```json
{
  "version": "1",
  "default": "deny",
  "tools": {
    "browser_evaluate": {
      "limits": [
        {
          "counter": "browser_evaluate_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}
```

## Other tools on Playwright (65)

- `browser_cookie_clear` — Destructive — https://policylayer.com/tools/hanato238-playwright-mcp/browser-cookie-clear.md
- `browser_cookie_delete` — Destructive — https://policylayer.com/tools/hanato238-playwright-mcp/browser-cookie-delete.md
- `browser_localstorage_clear` — Destructive — https://policylayer.com/tools/hanato238-playwright-mcp/browser-localstorage-clear.md
- `browser_localstorage_delete` — Destructive — https://policylayer.com/tools/hanato238-playwright-mcp/browser-localstorage-delete.md
- `browser_sessionstorage_clear` — Destructive — https://policylayer.com/tools/hanato238-playwright-mcp/browser-sessionstorage-clear.md
- `browser_sessionstorage_delete` — Destructive — https://policylayer.com/tools/hanato238-playwright-mcp/browser-sessionstorage-delete.md
- `browser_click` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-click.md
- `browser_close` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-close.md
- `browser_drag` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-drag.md
- `browser_drop` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-drop.md
- `browser_handle_dialog` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-handle-dialog.md
- `browser_hide_highlight` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-hide-highlight.md
- `browser_hover` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-hover.md
- `browser_mouse_click_xy` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-mouse-click-xy.md
- `browser_mouse_down` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-mouse-down.md
- `browser_mouse_drag_xy` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-mouse-drag-xy.md
- `browser_mouse_move_xy` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-mouse-move-xy.md
- `browser_mouse_up` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-mouse-up.md
- `browser_mouse_wheel` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-mouse-wheel.md
- `browser_navigate` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-navigate.md
- `browser_navigate_back` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-navigate-back.md
- `browser_network_request` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-network-request.md
- `browser_network_state_set` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-network-state-set.md
- `browser_pdf_save` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-pdf-save.md
- `browser_press_key` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-press-key.md
- `browser_resize` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-resize.md
- `browser_resume` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-resume.md
- `browser_route` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-route.md
- `browser_run_code_unsafe` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-run-code-unsafe.md
- `browser_select_option` — Execute — https://policylayer.com/tools/hanato238-playwright-mcp/browser-select-option.md
- …and 35 more: https://policylayer.com/tools/hanato238-playwright-mcp.md

## For agents

This record is a snapshot. Live verdicts and the full registry:

- Check every server in your MCP config at once: `npx -y policylayer stack`
- Vet a server before you add it: install the mcp-precheck skill — `npx skills add https://policylayer.com` (skill text: https://policylayer.com/skill.md)
- Query the registry over MCP: endpoint `https://api.policylayer.com/mcp` — tools `check_mcp_server`, `check_mcp_stack`, `check_tool`, `search_registry`, `get_change_events`

---

Source: the PolicyLayer MCP registry — one continuously verified record per MCP server. Full record: https://policylayer.com/registry?q=hanato238-playwright-mcp · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/hanato238-playwright-mcp