LLM spend so far: total + today, vs the daily budget.
AI agents call cost_summary to retrieve information from surveyHelper without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and displays financial tracking information (LLM spend totals and comparisons) for informational purposes only. It does not move money, create subscriptions, or commit financial obligations—it merely reads and reports cost data. The blast radius of misuse is minimal, as an agent cannot incur unexpected charges through reporting alone.
From the tool's definition Tool name 'cost_summary' and description 'LLM spend so far: total + today, vs the daily budget' indicate retrieval of cost/usage metrics with no modification or execution.
Attacks that exploit this kind of access
LLM spend so far: total + today, vs the daily budget. It is categorised as a Read tool in the surveyHelper MCP Server, which means it retrieves data without modifying state.
Register the surveyHelper MCP server in PolicyLayer and add a rule for cost_summary: 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 surveyHelper. Nothing to install.
cost_summary 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 cost_summary 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 cost_summary. 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.
cost_summary is provided by the surveyHelper MCP server (rich7420/surveyhelper). 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 →