device_set
Select active device for subsequent commands. Sets global state — all following tool calls will target this device until changed. For parallel multi-device workflows, prefer passing deviceId directly to each tool call instead of using device_set, which avoids race conditions from shared mutable s...
This record as markdown: /tools/alexgladkov-claude-in-mobile/device-set.md
What device_set does on Claude Mobile
AI agents use device_set to create or update resources in Claude Mobile, usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Claude Mobile environment.
Why device_set is rated Medium
This tool modifies mutable global state (the active device setting) that persists across subsequent tool invocations. While it doesn't directly execute actions or delete data, changing the active device target can redirect all future operations to a different device, potentially causing commands intended for one device to execute on another.
From the tool's definition device_set 'Sets global state — all following tool calls will target this device' and is used to configure which device subsequent commands target
Attacks that exploit this kind of access
The rule that runs device_set safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Claude Mobile, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For device_set, this is the rule to start with:
device_set stays usable, but capped: an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Claude Mobile, apply this rule, and every device_set call is checked against it from then on.
Questions about device_set
Select active device for subsequent commands. Sets global state — all following tool calls will target this device until changed. For parallel multi-device workflows, prefer passing deviceId directly to each tool call instead of using device_set, which avoids race conditions from shared mutable state. It is categorised as a Write tool in the Claude Mobile MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Claude Mobile MCP server in PolicyLayer and add a rule for device_set: 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 Claude Mobile. Nothing to install.
device_set is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the device_set 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 device_set. 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.
device_set is provided by the Claude Mobile MCP server (alexgladkov/claude-in-mobile). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Claude Mobile, and thousands of servers like it.
This server
Across the catalogue