input_back

THE RISK

High Risk

Why input_back needs a policy

This tool triggers a UI interaction on an external Android device (simulating a hardware back button press). It executes an action with real side effects — it can dismiss screens, cancel dialogs, close apps, or navigate away — making it an Execute-category tool. Misuse could disrupt ongoing operations on the device, hence medium severity.

From the tool's definition Simulates a back button event on the connected Android device

Documented attack patterns abuse exactly the kind of access input_back gives an agent:

POLICY

How to control input_back

PolicyLayer is an MCP gateway — it sits between your AI agents and Ultimate Android MCP, and nothing reaches the server without passing your rules. This is the rule we recommend for input_back:

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "input_back": {
      "limits": [
        {
          "counter": "input_back_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

input_back 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.

Create a free account and register Ultimate Android MCP — nothing to install.
Add this policy — paste it, or build it visually.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

RATE-LIMIT THIS TOOL →

Free to start. No card required.

EXPLORE

Related tools and policies

More Ultimate Android MCP tools

Destructive uninstall_package Uninstalls the specified package from the connected Android device. Requires the package_n Execute execute_adb_shell Executes a raw ADB shell command on the connected Android device and returns the output. R Execute launch_app Launches the app with the specified package name on the connected Android device. Requires Execute input_home Simulates a home button event on the connected Android device. No parameters are required. Execute input_keyevent Simulates a key event with the specified keycode on the connected Android device. Requires Execute input_roll Simulates a rolling gesture with the specified dx and dy values on the connected Android d Execute input_text Simulates typing the given text input into the currently focused field on the connected An Execute input_press Simulates a key press event with the specified keycode on the connected Android device. Re

All 35 Ultimate Android MCP tools →

Execute tools on other servers

Android Forensics ADB MCP Server adb_shell_command EMBA-MCP run_emba_scan Frida Game Hacking MCP remove_breakpoint Procmon request_elevation

Go deeper

High-risk tools in Ultimate Android MCP → Every high-risk tool this server exposes, in one view.
All high-risk MCP tools → Every high-risk tool across the catalogue, and the policy pattern that contains them.
The MCP Attack Database → Documented attack patterns against MCP deployments — and the policies that stop them.
Agent sandbox → Bounding what an agent can execute and where the boundary should sit.
Least privilege for MCP → Granting agents the minimum tool surface the task needs.
Rate Limiting MCP Tool Calls: A Practical Guide →
MCP Security: Why Prompt Guardrails Aren't Enough →

FAQ

Questions about input_back

What does the input_back tool do? +

Simulates a back button event on the connected Android device. No parameters are required. It is categorised as a Execute tool in the Ultimate Android MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on input_back? +

Register the Ultimate Android MCP server in PolicyLayer and add a rule for input_back: 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 Ultimate Android MCP. Nothing to install.

What risk level is input_back? +

input_back is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.

Can I rate-limit input_back? +

Yes. Add a rate_limit block to the input_back 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.

How do I block input_back completely? +

Set action: deny in the PolicyLayer policy for input_back. 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.

What MCP server provides input_back? +

input_back is provided by the Ultimate Android MCP server (oddlyspaced/ultimate-android-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Enforce policy on every Ultimate Android MCP tool call.

Start from Ultimate Android MCP, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.

CHECK YOUR STACK →