AI agents call screenshot to retrieve information from Sl Test without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Screenshots are read-only operations that retrieve visual information about the current state of a system without side effects, modification, or execution. The tool explicitly directs users to 'describe_ui instead' for interactive operations, further confirming it is passive observation only. Low severity because screenshot capture poses minimal risk—it reveals UI state but cannot alter system behavior or data.
From the tool's definition Tool name is 'screenshot' and description states it 'Captures screenshot for visual verification.' This is a non-destructive data retrieval operation that observes the current UI state without modifying or executing anything.
Attacks that exploit this kind of access
Captures screenshot for visual verification. For UI coordinates, use describe_ui instead (don. It is categorised as a Read tool in the Sl Test MCP Server, which means it retrieves data without modifying state.
Register the Sl Test MCP server in PolicyLayer and add a rule for screenshot: 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 Sl Test. Nothing to install.
screenshot 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 screenshot 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 screenshot. 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.
screenshot is provided by the Sl Test MCP server (sampsonky/xcodebuildmcp). 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 →