AI agents use create_va to create or update resources in Xendit — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Xendit environment.
This tool creates virtual accounts which are financial instruments, but the action itself is a Create (Write) operation—it establishes new payment infrastructure rather than moving funds or deleting data. The severity is high because misconfiguration could create unauthorized payment channels or financial exposure. Confidence is high due to explicit 'create' language and clear financial domain context.
From the tool's definition Tool name 'create_va' and description 'Create a virtual account for payment collection' explicitly indicate creating a new financial account structure. This is a reversible data creation operation in a payment gateway context.
Attacks that exploit this kind of access
Create a virtual account for payment collection. It is categorised as a Write tool in the Xendit MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Xendit MCP server in PolicyLayer and add a rule for create_va: 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 Xendit. Nothing to install.
create_va 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 create_va 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 create_va. 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.
create_va is provided by the Xendit MCP server (theyahia/xendit-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 →