browser_click

What browser_click does on A11y Agent Skills

AI agents invoke browser_click to trigger actions in A11y Agent Skills. 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.

Why browser_click is rated High

Clicking a web element is an external browser action whose effects are entirely argument-dependent. A click can trigger purchases, deletions, form submissions, or other irreversible operations. Because the outcome depends on context and can span multiple severity levels, Execute is the appropriate category with high severity due to the broad blast radius of arbitrary click actions in a browser session.

From the tool's definition "Perform click on a web page" — triggers browser interactions that can submit forms, navigate pages, confirm dialogs, initiate downloads, or invoke any UI-driven action depending on what element is clicked.

The rule that runs browser_click safely

PolicyLayer is an MCP gateway: it sits between your AI agents and A11y Agent Skills, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For browser_click, this is the rule to start with:

{
  "version": "1",
  "default": "deny",
  "tools": {
    "browser_click": {
      "limits": [
        {
          "counter": "browser_click_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

Questions about browser_click