AI agents invoke input_scroll to trigger actions in Android Debug Bridge MCP. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool executes a scroll gesture on an Android device, which is an external operation that simulates user input. It falls under Execute because it triggers a device-side action whose effects depend on the current UI state and arguments (direction, distance, target element).
From the tool's definition 'Perform scroll action' — triggers a UI input action on an Android device via ADB
Documented attack patterns abuse exactly the kind of access input_scroll gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Android Debug Bridge MCP, and nothing reaches the server without passing your rules. This is the rule we recommend for input_scroll:
{
"version": "1",
"default": "deny",
"tools": {
"input_scroll": {
"limits": [
{
"counter": "input_scroll_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
} input_scroll stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Perform scroll action. It is categorised as a Execute tool in the Android Debug Bridge MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Android Debug Bridge MCP server in PolicyLayer and add a rule for input_scroll: 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 Android Debug Bridge MCP. Nothing to install.
input_scroll is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the input_scroll 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 input_scroll. 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.
input_scroll is provided by the Android Debug Bridge MCP server (tiagodanin/android-debug-bridge-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Android Debug Bridge MCP, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
9 Android Debug Bridge MCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.