see

THE RISK

Low Risk

Why see needs a policy

This tool takes a screenshot and identifies UI controls on screen — a read/observe operation with no side effects. It retrieves visual state information without modifying anything.

From the tool's definition 截图并识别当前屏幕上的可操作控件 (Screenshot and identify operable controls on the current screen)

Documented attack patterns abuse exactly the kind of access see gives an agent:

POLICY

How to control see

PolicyLayer is an MCP gateway — it sits between your AI agents and EZComputerCtrl MCP, and nothing reaches the server without passing your rules. This is the rule we recommend for see:

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "see": {}
  }
}

see is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.

Create a free account and register EZComputerCtrl MCP — nothing to install.
Add this policy — paste it, or build it visually.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

CAP THIS TOOL →

Free to start. No card required.

EXPLORE

Related tools and policies

More EZComputerCtrl MCP tools

Execute click 单击指定控件，支持 left 和 right 两种按键。 Execute hotkey 发送组合快捷键，支持 ctrl+c 或 ["ctrl", "c"]。 Execute move_to 移动鼠标到指定控件中心点，用于悬停触发。 Execute scroll 在指定控件处滚动鼠标滚轮。 Execute type_text 输入文本；若提供 object_id，则自动先点击再输入。

All 6 EZComputerCtrl MCP tools →

Read tools on other servers

M-Team MCP Server get_torrent_detail YouTube MCP Server search_channel Android Forensics ADB MCP Server parse_browser_history 0xarchive web3_challenge

Go deeper

The MCP Attack Database → Documented attack patterns against MCP deployments — and the policies that stop them.
Data exfiltration → How read access becomes an exfiltration channel when chained with untrusted content.
MCP token cost → What connecting this server costs in context-window tokens on every request.
Rate Limiting MCP Tool Calls: A Practical Guide →
MCP Security: Why Prompt Guardrails Aren't Enough →

FAQ

Questions about see

What does the see tool do? +

截图并识别当前屏幕上的可操作控件。. It is categorised as a Read tool in the EZComputerCtrl MCP MCP Server, which means it retrieves data without modifying state.

How do I enforce a policy on see? +

Register the EZComputerCtrl MCP server in PolicyLayer and add a rule for see: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches EZComputerCtrl MCP. Nothing to install.

What risk level is see? +

see is a Read tool with low risk. Read-only tools are generally safe to allow by default.

Can I rate-limit see? +

Yes. Add a rate_limit block to the see rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.

How do I block see completely? +

Set action: deny in the PolicyLayer policy for see. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.

What MCP server provides see? +

see is provided by the EZComputerCtrl MCP server (jucieovo/ezcomputerctrl-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Enforce policy on every EZComputerCtrl MCP tool call.

Start from EZComputerCtrl MCP, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.

CHECK YOUR STACK →