AI agents call retrieve_vouchers to retrieve information from Zbd without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool retrieves voucher data, which is a read-only operation. The 'get' verb and context of sibling tools that create/write financial transactions confirm this is a data retrieval tool. Severity is low because reading voucher records poses minimal risk; the blast radius of misuse is limited to information disclosure.
From the tool's definition Tool name 'retrieve_vouchers' and description 'Get Voucher' indicate a retrieval/query operation with no modification or side effects.
Attacks that exploit this kind of access
Get Voucher. It is categorised as a Read tool in the Zbd MCP Server, which means it retrieves data without modifying state.
Register the Zbd MCP server in PolicyLayer and add a rule for retrieve_vouchers: 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 Zbd. Nothing to install.
retrieve_vouchers is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the retrieve_vouchers 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 retrieve_vouchers. 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.
retrieve_vouchers is provided by the Zbd MCP server (zebedeeio/zbd-payments-typescript-sdk). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
retrieve_vouchers is one line of Zbd's registry record.
The record carries the whole server: verified identity, auth posture, risk grade, every tool classified, recommended policy — re-checked continuously.
Teams ship this data inside their own products. See what a licence covers →