destroy_sandbox

Force destroy a sandbox container.

SERVERAgent Jail SOURCEidiscoord13-sketch/agent-jail-mcp

Critical RISK CLASS

Category Destructive

Parameters 00 required

Recommended Hiddensee the rule below

Registry record Grade F, identity unverified Pull the record →

This record as markdown: /tools/agent-jail-mcp/destroy-sandbox.md

WHAT IT DOES

What destroy_sandbox does on Agent Jail

AI agents call destroy_sandbox to permanently remove resources in Agent Jail, typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.

RISK

Why destroy_sandbox is rated Critical

An AI agent that decides to call destroy_sandbox doesn't hesitate, doesn't double-check, and doesn't stop at one. Whatever it removes from Agent Jail is gone. There is no undo for destructive operations.

Attacks that exploit this kind of access

RECOMMENDED RULE

The rule that runs destroy_sandbox safely

PolicyLayer is an MCP gateway: it sits between your AI agents and Agent Jail, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For destroy_sandbox, this is the rule to start with:

destroy_sandbox Hidden

destroy_sandbox is removed from the agent's tool list entirely, so the agent never calls it. The rest of the server keeps working.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "hide": [
    "destroy_sandbox"
  ]
}

RESTRICT THIS TOOL → Instant setup, no code required.

The button opens the PolicyLayer dashboard: create your workspace, connect Agent Jail, apply this rule, and every destroy_sandbox call is checked against it from then on.

FAQ

Questions about destroy_sandbox

What does the destroy_sandbox tool do? +

Force destroy a sandbox container. It is categorised as a Destructive tool in the Agent Jail MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.

How do I enforce a policy on destroy_sandbox? +

Register the Agent Jail MCP server in PolicyLayer and add a rule for destroy_sandbox: 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 Agent Jail. Nothing to install.

What risk level is destroy_sandbox? +

destroy_sandbox is a Destructive tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.

Can I rate-limit destroy_sandbox? +

Yes. Add a rate_limit block to the destroy_sandbox 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.

How do I block destroy_sandbox completely? +

Set action: deny in the PolicyLayer policy for destroy_sandbox. 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.

What MCP server provides destroy_sandbox? +

destroy_sandbox is provided by the Agent Jail MCP server (idiscoord13-sketch/agent-jail-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

KEEP EXPLORING

More on Agent Jail, and thousands of servers like it.

This server

All 35 Agent Jail tools → Critical-risk tools on this server →

Across the catalogue

All critical-risk MCP tools → The MCP Attack Database →

Guides

Govern database agents →Blast radius →Tool call approval →