AI agents call screenshot_compare to retrieve information from InSite without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool retrieves and analyzes screenshot data to generate visual diffs. This is fundamentally a read operation: it queries or extracts information from captured images for comparison purposes. There are no side effects on the screenshots themselves (they are not modified or deleted), no code execution triggered by comparison logic, and no financial or destructive consequences.
From the tool's definition Tool performs 'compare screenshots with diff generation' — a visual analysis and comparison operation that reads and examines existing screenshots without modifying, deleting, or executing code.
Attacks that exploit this kind of access
Compare screenshots with diff generation for visual testing. It is categorised as a Read tool in the InSite MCP Server, which means it retrieves data without modifying state.
Register the InSite MCP server in PolicyLayer and add a rule for screenshot_compare: 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 InSite. Nothing to install.
screenshot_compare 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_compare 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_compare. 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_compare is provided by the InSite MCP server (jowharshamshiri/insite). 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 →