Capture screenshot of an element (if supported)
AI agents call capture_screenshot to retrieve information from React Devtools without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool captures visual information about React component rendering state. It retrieves data (a screenshot) with no side effects, no modifications to application state, and no code execution. While it provides visibility into application state, it is fundamentally a read operation.
From the tool's definition Tool name 'capture_screenshot' and description 'Capture screenshot of an element (if supported)' indicate a read-only operation that retrieves a visual representation of UI state without modifying data or executing arbitrary code.
Attacks that exploit this kind of access
Capture screenshot of an element (if supported). It is categorised as a Read tool in the React Devtools MCP Server, which means it retrieves data without modifying state.
Register the React Devtools MCP server in PolicyLayer and add a rule for capture_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 React Devtools. Nothing to install.
capture_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 capture_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 capture_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.
capture_screenshot is provided by the React Devtools MCP server (skylarbarrera/react-devtools-mcp). 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 →