AI agents invoke connect_adb_device to trigger actions in MaaMCP. 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.
Based on the tool name and server context (MaaFramework-based MCP server for Android device automation), this tool likely establishes an ADB (Android Debug Bridge) connection to an Android device, which triggers an external operation. ADB connections enable control over Android devices, making this an Execute-category tool. Severity is high because connecting to a device opens the door for full device control.
From the tool's definition Tool name 'connect_adb_device' on a server providing Android device automation capabilities; description is empty.
Attacks that exploit this kind of access
connect_adb_device. It is categorised as a Execute tool in the MaaMCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Maa MCP server in PolicyLayer and add a rule for connect_adb_device: 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 MaaMCP. Nothing to install.
connect_adb_device 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 connect_adb_device 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 connect_adb_device. 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.
connect_adb_device is provided by the Maa MCP server (maa-ai/maamcp). 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 →