Discard all sandbox changes and re-copy from current production state. This destroys all pending sandbox edits. Requires confirmed: true.
AI agents call reset_sandbox to permanently remove resources in Localization — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
| Parameter | Type | Required | Description |
|---|---|---|---|
confirmed | boolean | — | Must be true to execute. Without confirmation, returns a warning instead. |
projectSlug | string | Yes | Project slug |
Parameters from the server's own tool schema.
This tool irreversibly deletes pending sandbox edits without the ability to recover them (they are overwritten by re-copying production state). This is a classic destructive operation that cannot be undone. While confined to a sandbox environment (limiting blast radius to medium-high rather than critical), the permanent loss of user work justifies 'Destructive' category and 'high' severity.
From the tool's definition Tool description states: 'Discard all sandbox changes and re-copy from current production state. This destroys all pending sandbox edits.' The word 'destroys' and 'discard' indicate irreversible deletion of user work (pending sandbox edits).
Attacks that exploit this kind of access
Discard all sandbox changes and re-copy from current production state. This destroys all pending sandbox edits. Requires confirmed: true. It is categorised as a Destructive tool in the Localization MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
reset_sandbox accepts 2 parameters: confirmed, projectSlug. Required: projectSlug. The full parameter table on this page comes from the server's own tool schema.
Register the Localization MCP server in PolicyLayer and add a rule for reset_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 Localization. Nothing to install.
reset_sandbox is a Destructive tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.
Yes. Add a rate_limit block to the reset_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.
Set action: deny in the PolicyLayer policy for reset_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.
reset_sandbox is provided by the Localization MCP server (localization-mcp-server). 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 →