Log token usage and cost for a task. Call this after each AI interaction.
AI agents use log_cost to create or update resources in Agent Cost — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Agent Cost environment.
This tool writes cost and token usage records to the tracking system after each AI interaction. It creates new log entries (a reversible write operation) rather than deleting data, executing code, or moving money. Misuse could corrupt cost tracking data or inflate/deflate reported spending metrics, hence medium severity.
From the tool's definition 'Log token usage and cost for a task. Call this after each AI interaction.' - creates/records cost data entries
Attacks that exploit this kind of access
Log token usage and cost for a task. Call this after each AI interaction. It is categorised as a Write tool in the Agent Cost MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Agent Cost MCP server in PolicyLayer and add a rule for log_cost: 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 Agent Cost. Nothing to install.
log_cost 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 log_cost 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 log_cost. 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.
log_cost is provided by the Agent Cost MCP server (vanthienha199/agent-cost-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 →