Launch an installed application by package name on the Android box.
AI agents invoke open_app to trigger actions in Android Tester 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 triggers an external operation — launching an application on an Android device. It causes a real side effect (starting a process/app) whose impact depends on which app is launched. It doesn't merely read data or write/modify stored data, making Execute the most appropriate category.
From the tool's definition Launch an installed application by package name on the Android box.
Attacks that exploit this kind of access
Launch an installed application by package name on the Android box. It is categorised as a Execute tool in the Android Tester MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Android Tester MCP server in PolicyLayer and add a rule for open_app: 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 Tester MCP. Nothing to install.
open_app 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 open_app 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 open_app. 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.
open_app is provided by the Android Tester MCP server (zhenweiwang1990/android-tester-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 →