AI agents call list_cost_centers to retrieve information from Fortnox without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves or queries cost center data from Fortnox's accounting system for display/reference purposes. No data is created, modified, deleted, or financial transactions are committed. The blast radius of misuse is minimal — an agent could at worst enumerate cost centers, which is informational data typically already accessible to authorized users.
From the tool's definition Tool name is 'list_cost_centers' and description states it 'List cost centers' — a retrieval operation with no modification or side effects. The verb 'list' is explicitly a read operation.
Attacks that exploit this kind of access
List cost centers for allocating expenses across departments or projects. It is categorised as a Read tool in the Fortnox MCP Server, which means it retrieves data without modifying state.
Register the Fortnox MCP server in PolicyLayer and add a rule for list_cost_centers: 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 Fortnox. Nothing to install.
list_cost_centers 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 list_cost_centers 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 list_cost_centers. 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.
list_cost_centers is provided by the Fortnox MCP server (proviscale/fortnox-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 →