AI agents use wave_approve_invoice to create or update resources in Waveapps — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Waveapps environment.
Approving an invoice is a Write operation because it modifies existing financial data (invoice status) reversibly—the approval can typically be undone by returning to draft status. It is not Destructive (data remains and can be reverted), not Financial (no money movement), and not Execute (no arbitrary code execution).
From the tool's definition Tool name 'wave_approve_invoice' and description indicate it changes invoice status from draft to approved, which is a state modification of financial records.
Attacks that exploit this kind of access
Approves a draft invoice (status. It is categorised as a Write tool in the Waveapps MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Waveapps MCP server in PolicyLayer and add a rule for wave_approve_invoice: 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 Waveapps. Nothing to install.
wave_approve_invoice 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 wave_approve_invoice 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 wave_approve_invoice. 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.
wave_approve_invoice is provided by the Waveapps MCP server (lunaparker/waveapps-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 →