AI agents call take_screenshot_tool to retrieve information from Openmcp without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Taking a screenshot captures the current visual state of a web page without modifying, deleting, or executing code. This is a read-only operation with no side effects on the target system. Severity is low because screenshots cannot cause harm beyond potential information disclosure of what is already visible to the agent.
From the tool's definition Tool name 'take_screenshot_tool' and server description indicating 'taking screenshots' as part of web automation capabilities. No description provided for this specific tool.
Attacks that exploit this kind of access
take_screenshot_tool. It is categorised as a Read tool in the Openmcp MCP Server, which means it retrieves data without modifying state.
Register the Open MCP server in PolicyLayer and add a rule for take_screenshot_tool: 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 Openmcp. Nothing to install.
take_screenshot_tool 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 take_screenshot_tool 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 take_screenshot_tool. 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.
take_screenshot_tool is provided by the Open MCP server (openmcp-pro/openmcp). 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 →