AI agents invoke power_off_server to trigger actions in Hcloud. 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.
A hard power-off is an external operation that abruptly stops a running server, potentially causing data loss or corruption due to no graceful shutdown. It is not reversible in the sense that any in-flight writes or state may be lost, but the server itself can be powered back on, placing this in Execute rather than Destructive. The blast radius is high as it disrupts live services.
From the tool's definition 'Power off a server (hard)' — forcibly cuts power to a running server, triggering an immediate halt without graceful shutdown
Attacks that exploit this kind of access
Power off a server (hard). It is categorised as a Execute tool in the Hcloud MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Hcloud MCP server in PolicyLayer and add a rule for power_off_server: 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 Hcloud. Nothing to install.
power_off_server 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 power_off_server 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 power_off_server. 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.
power_off_server is provided by the Hcloud MCP server (xodus-co/hcloud-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 →