Estimate cost without logging it. Use for planning.
AI agents call estimate_cost to retrieve information from Agent Cost without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool performs a calculation or lookup to project costs for planning purposes. The explicit statement 'without logging it' confirms no data is written or modified. This is a read-only operation with no blast radius if misused by an agent—it merely returns an estimate.
From the tool's definition Tool name 'estimate_cost' and description 'Estimate cost without logging it. Use for planning' indicate this tool retrieves or queries cost projection data without modifying any records or triggering side effects.
Attacks that exploit this kind of access
Estimate cost without logging it. Use for planning. It is categorised as a Read tool in the Agent Cost MCP Server, which means it retrieves data without modifying state.
Register the Agent Cost MCP server in PolicyLayer and add a rule for estimate_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.
estimate_cost 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 estimate_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 estimate_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.
estimate_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 →