AI agents use customs_create_order_draft to create or update resources in Customs — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Customs environment.
This tool creates audit entries and validates draft orders but does not execute actual orders, delete data, or move funds. It is a reversible write operation with moderate blast radius: malicious use could pollute audit logs or create misleading draft records that might influence downstream customs decisions, but the pre-check-only nature limits immediate impact.
From the tool's definition Tool description states 'write an audit row' which is a data creation/modification operation. The 'PRE-CHECK ONLY' qualifier and 'validate the payload' indicate reversible operations limited to draft creation and logging.
Attacks that exploit this kind of access
PRE-CHECK ONLY: validate the payload and write an audit row. It is categorised as a Write tool in the Customs MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Customs MCP server in PolicyLayer and add a rule for customs_create_order_draft: 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 Customs. Nothing to install.
customs_create_order_draft 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 customs_create_order_draft 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 customs_create_order_draft. 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.
customs_create_order_draft is provided by the Customs MCP server (yak33/customs-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 →