Capture a PNG screenshot using adb exec-out screencap -p.
AI agents call android_screenshot to retrieve information from DevLab MCP Suite without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Screenshot capture is a non-destructive, non-modifying query of the current device display state. It retrieves information (visual data) with no side effects, reversible operations, code execution with external consequences, financial impact, or data deletion. This aligns squarely with the Read category.
From the tool's definition Tool description states it 'Capture[s] a PNG screenshot' — a read-only operation that retrieves visual state without modifying device data or triggering side effects.
Attacks that exploit this kind of access
Capture a PNG screenshot using adb exec-out screencap -p. It is categorised as a Read tool in the DevLab MCP Suite MCP Server, which means it retrieves data without modifying state.
Register the DevLab MCP Suite MCP server in PolicyLayer and add a rule for android_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 DevLab MCP Suite. Nothing to install.
android_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 android_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 android_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.
android_screenshot is provided by the DevLab MCP Suite MCP server (tanguito86/devlab-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 →