# run_javascript

Execute JavaScript in a Safari tab. Use list_tabs to find window/tab indices. Returns the result as a string.

Agent View of the PolicyLayer registry record for `run_javascript`. HTML page: https://policylayer.com/tools/heznpc-airmcp/run-javascript

## Facts

- Tool: `run_javascript`
- Server: AirMCP (`airmcp`) — https://policylayer.com/tools/heznpc-airmcp.md
- Install: `npx -y airmcp`
- Homepage: https://github.com/heznpc/AirMCP
- 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": "run_javascript",
    "arguments": {}
  }
}
```

## Why run_javascript is rated High

This tool executes arbitrary JavaScript in a Safari tab with full access to the DOM, local storage, and potentially sensitive data within the browser context. An AI agent could use this to steal credentials, modify webpage content, exfiltrate data from web applications, perform unauthorized actions on behalf of the user, or establish persistence.

From the tool's own definition: "'Execute JavaScript in a Safari tab' and 'Returns the result as a string' directly indicate arbitrary code execution capability."

## Use case

AI agents invoke run_javascript to trigger actions in AirMCP. 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 AirMCP:

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

## Other tools on AirMCP (278)

- `close_tab` — Destructive — https://policylayer.com/tools/heznpc-airmcp/close-tab.md
- `delete_contact` — Destructive — https://policylayer.com/tools/heznpc-airmcp/delete-contact.md
- `delete_event` — Destructive — https://policylayer.com/tools/heznpc-airmcp/delete-event.md
- `delete_note` — Destructive — https://policylayer.com/tools/heznpc-airmcp/delete-note.md
- `delete_photos` — Destructive — https://policylayer.com/tools/heznpc-airmcp/delete-photos.md
- `delete_playlist` — Destructive — https://policylayer.com/tools/heznpc-airmcp/delete-playlist.md
- `delete_reminder` — Destructive — https://policylayer.com/tools/heznpc-airmcp/delete-reminder.md
- `delete_reminder_list` — Destructive — https://policylayer.com/tools/heznpc-airmcp/delete-reminder-list.md
- `delete_shortcut` — Destructive — https://policylayer.com/tools/heznpc-airmcp/delete-shortcut.md
- `memory_forget` — Destructive — https://policylayer.com/tools/heznpc-airmcp/memory-forget.md
- `move_note` — Destructive — https://policylayer.com/tools/heznpc-airmcp/move-note.md
- `quit_app` — Destructive — https://policylayer.com/tools/heznpc-airmcp/quit-app.md
- `remove_from_playlist` — Destructive — https://policylayer.com/tools/heznpc-airmcp/remove-from-playlist.md
- `semantic_clear` — Destructive — https://policylayer.com/tools/heznpc-airmcp/semantic-clear.md
- `semantic_index` — Destructive — https://policylayer.com/tools/heznpc-airmcp/semantic-index.md
- `spotlight_clear` — Destructive — https://policylayer.com/tools/heznpc-airmcp/spotlight-clear.md
- `system_power` — Destructive — https://policylayer.com/tools/heznpc-airmcp/system-power.md
- `trash_file` — Destructive — https://policylayer.com/tools/heznpc-airmcp/trash-file.md
- `activate_tab` — Execute — https://policylayer.com/tools/heznpc-airmcp/activate-tab.md
- `ai_agent` — Execute — https://policylayer.com/tools/heznpc-airmcp/ai-agent.md
- `ai_chat` — Execute — https://policylayer.com/tools/heznpc-airmcp/ai-chat.md
- `ai_plan_metrics` — Execute — https://policylayer.com/tools/heznpc-airmcp/ai-plan-metrics.md
- `connect_bluetooth` — Execute — https://policylayer.com/tools/heznpc-airmcp/connect-bluetooth.md
- `disconnect_bluetooth` — Execute — https://policylayer.com/tools/heznpc-airmcp/disconnect-bluetooth.md
- `drop_pin` — Execute — https://policylayer.com/tools/heznpc-airmcp/drop-pin.md
- `duplicate_shortcut` — Execute — https://policylayer.com/tools/heznpc-airmcp/duplicate-shortcut.md
- `edit_shortcut` — Execute — https://policylayer.com/tools/heznpc-airmcp/edit-shortcut.md
- `event_subscribe` — Execute — https://policylayer.com/tools/heznpc-airmcp/event-subscribe.md
- `generate_image` — Execute — https://policylayer.com/tools/heznpc-airmcp/generate-image.md
- `generate_structured` — Execute — https://policylayer.com/tools/heznpc-airmcp/generate-structured.md
- …and 248 more: https://policylayer.com/tools/heznpc-airmcp.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=heznpc-airmcp · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/heznpc-airmcp