截取并分析屏幕内容，返回识别到的文字和UI元素。需要先调用 screen_get_layout 绑定目标窗口。
AI agents call screen_capture to retrieve information from Screen Agent without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
screen_capture retrieves visual information from the screen through OCR and UI element detection. While it reads sensitive screen content (which could include passwords, documents, or private information visible on screen), it performs no write, execute, or destructive operations.
From the tool's definition Tool name 'screen_capture' and description indicating it '截取并分析屏幕内容，返回识别到的文字和UI元素' (captures and analyzes screen content, returns recognized text and UI elements). This is a retrieval operation with no modification of data.
Attacks that exploit this kind of access
截取并分析屏幕内容，返回识别到的文字和UI元素。需要先调用 screen_get_layout 绑定目标窗口。. It is categorised as a Read tool in the Screen Agent MCP Server, which means it retrieves data without modifying state.
Register the Screen Agent MCP server in PolicyLayer and add a rule for screen_capture: 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 Screen Agent. Nothing to install.
screen_capture is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the screen_capture 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.
Set action: deny in the PolicyLayer policy for screen_capture. 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.
screen_capture is provided by the Screen Agent MCP server (lqszhsp/screen-agent). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →