AI agents use attach_drive to create or update resources in Mcp Utm — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Mcp Utm environment.
Attaching a drive is a write operation that modifies VM configuration but does not execute arbitrary code, delete data, or cause irreversible damage. The action is reversible (drive can be detached). While it could indirectly enable malicious activities (e.g., booting from a malicious ISO), the tool itself performs a configuration modification rather than executing those operations.
From the tool's definition Tool description states 'Attach an ISO or disk image to a removable drive' - this modifies VM configuration by adding/mounting a drive attachment, which is a reversible change to VM state.
Attacks that exploit this kind of access
Attach an ISO or disk image to a removable drive. It is categorised as a Write tool in the Mcp Utm MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Mcp Utm MCP server in PolicyLayer and add a rule for attach_drive: 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 Mcp Utm. Nothing to install.
attach_drive 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 attach_drive 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 attach_drive. 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.
attach_drive is provided by the Mcp Utm MCP server (neverprepared/mcp-utm). 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 →