AI agents call get_device to retrieve information from Scholar without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and returns the current device configuration (CPU or GPU) used for embeddings. It is purely informational—no data is created, modified, deleted, or executed. The operation has no side effects and is analogous to checking system status. This clearly falls into the Read category.
From the tool's definition Tool name 'get_device' and description 'Return which device (CPU/GPU) is currently being used for embeddings' indicates a query operation that retrieves system configuration information without modifying or executing external operations.
Attacks that exploit this kind of access
Return which device (CPU/GPU) is currently being used for embeddings. It is categorised as a Read tool in the Scholar MCP Server, which means it retrieves data without modifying state.
Register the Scholar MCP server in PolicyLayer and add a rule for get_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 Scholar. Nothing to install.
get_device is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the get_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 get_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.
get_device is provided by the Scholar MCP server (unlomtrois/little-librarian). 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 →