# evaluate

Evaluates a JavaScript script

Agent View of the PolicyLayer registry record for `evaluate`. HTML page: https://policylayer.com/tools/async23-chrome-devtools-mcp/evaluate

## Facts

- Tool: `evaluate`
- Server: Chrome Devtools (`@async23/chrome-devtools-mcp`) — https://policylayer.com/tools/async23-chrome-devtools-mcp.md
- Install: `npx -y @async23/chrome-devtools-mcp`
- Homepage: https://github.com/Async23/chrome-devtools-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": "evaluate",
    "arguments": {}
  }
}
```

## Why evaluate is rated High

Executing arbitrary JavaScript in a live browser session can access cookies, credentials, DOM, local storage, make network requests, exfiltrate data, or manipulate page state. This is the highest-risk Execute action with critical blast radius if misused by an AI agent.

From the tool's own definition: "'Evaluates a JavaScript script' — executes arbitrary JavaScript in a live browser context"

## Use case

AI agents invoke evaluate to trigger actions in Chrome Devtools. 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 Chrome Devtools:

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

## Other tools on Chrome Devtools (49)

- `close_heapsnapshot` — Destructive — https://policylayer.com/tools/async23-chrome-devtools-mcp/close-heapsnapshot.md
- `close_page` — Destructive — https://policylayer.com/tools/async23-chrome-devtools-mcp/close-page.md
- `uninstall_extension` — Destructive — https://policylayer.com/tools/async23-chrome-devtools-mcp/uninstall-extension.md
- `click` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/click.md
- `click_at` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/click-at.md
- `drag` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/drag.md
- `emulate` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/emulate.md
- `evaluate_script` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/evaluate-script.md
- `execute_3p_developer_tool` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/execute-3p-developer-tool.md
- `execute_webmcp_tool` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/execute-webmcp-tool.md
- `fill_form` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/fill-form.md
- `handle_dialog` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/handle-dialog.md
- `hover` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/hover.md
- `install_extension` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/install-extension.md
- `lighthouse_audit` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/lighthouse-audit.md
- `navigate` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/navigate.md
- `performance_start_trace` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/performance-start-trace.md
- `performance_stop_trace` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/performance-stop-trace.md
- `press_key` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/press-key.md
- `reload_extension` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/reload-extension.md
- `resize_page` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/resize-page.md
- `screencast_stop` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/screencast-stop.md
- `select_page` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/select-page.md
- `trigger_extension_action` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/trigger-extension-action.md
- `type_text` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/type-text.md
- `wait_for` — Execute — https://policylayer.com/tools/async23-chrome-devtools-mcp/wait-for.md
- `compare_heapsnapshots` — Read — https://policylayer.com/tools/async23-chrome-devtools-mcp/compare-heapsnapshots.md
- `get_console_message` — Read — https://policylayer.com/tools/async23-chrome-devtools-mcp/get-console-message.md
- `get_heapsnapshot_class_nodes` — Read — https://policylayer.com/tools/async23-chrome-devtools-mcp/get-heapsnapshot-class-nodes.md
- `get_heapsnapshot_details` — Read — https://policylayer.com/tools/async23-chrome-devtools-mcp/get-heapsnapshot-details.md
- …and 19 more: https://policylayer.com/tools/async23-chrome-devtools-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=async23-chrome-devtools-mcp · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/async23-chrome-devtools-mcp