AI agents use refund to commit financial operations through ClawPay — usually the final step of a payment, billing, or trading workflow. A call moves real money.
Refunds move money back to customers, creating direct financial liability and obligation. This is a core financial operation that, if misused by an AI agent, could result in unauthorized money transfers, customer fraud, or significant financial loss. Severity is critical due to the direct monetary impact and potential for large-scale abuse in an automated shopping context.
From the tool's definition Tool name is 'refund' and description states 'Refund a payment intent.' In the context of a payment processor integration (Lithic, Stripe, PayPal), this directly reverses financial transactions and commits financial obligations by returning funds to customers.
Attacks that exploit this kind of access
Refund a payment intent. It is categorised as a Financial tool in the ClawPay MCP Server, which means it involves financial transactions. Block by default and require explicit approval.
Register the ClawPay MCP server in PolicyLayer and add a rule for refund: 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 ClawPay. Nothing to install.
refund is a Financial 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 refund 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 refund. 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.
refund is provided by the ClawPay MCP server (xodn348/clawpay). 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 →